A Review of RFID Sensors, the New Frontier of Internet of Things

sensors

Review A Review of RFID Sensors, the New Frontier of Internet of Things

Filippo Costa 1,* , Simone Genovesi 1, Michele Borgese 2 , Andrea Michel 1 , Francesco Alessio Dicandia 3 and Giuliano Manara 1

1 Dipartimento di Ingegneria dell’Informazione, Università di Pisa, 56126 Pisa, Italy; [email protected] (S.G.); [email protected] (A.M.); [email protected] (G.M.) 2 Research and Development Department, Siae Microelettronica, 20093 Cologno Monzese, Italy; [email protected] 3 IDS Ingegneria dei Sistemi SpA, 56121 Pisa, Italy; [email protected] * Correspondence: ﬁ[email protected]

Abstract: A review of technological solutions for RFID sensing and their current or envisioned applications is presented. The fundamentals of the wireless sensing technology are summarized in the first part of the work, and the benefits of adopting RFID sensors for replacing standard sensor- equipped Wi-Fi nodes are discussed. Emphasis is put on the absence of batteries and the lower cost of RFID sensors with respect to other sensor solutions available on the market. RFID sensors are critically compared by separating them into chipped and chipless configurations. Both categories are further analyzed with reference to their working mechanism (electronic, electromagnetic, and acoustic). RFID sensing through chip-equipped tags is now a mature technological solution, which is continuously increasing its presence on the market and in several applicative scenarios. On the other

hand, chipless RFID sensing represents a relatively new concept, which could become a disruptive solution in the market, but further research in this ﬁeld is necessary for customizing its employment in

Citation: Costa, F.; Genovesi, S.; specific scenarios. The benefits and limitations of several tag configurations are shown and discussed. Borgese, M.; Michel, A.; Dicandia, A summary of the most suitable applicative scenarios for RFID sensors are finally illustrated. Finally, F.A.; Manara, G. A Review of RFID a look at some sensing solutions available on the market are described and compared. Sensors, the New Frontier of Internet of Things. Sensors 2021, 21, 3138. Keywords: RFID; sensors; RFID sensors; RFID antennas; RFID readers; chipless RFID; wireless sen- https://doi.org/10.3390/s21093138 sors; wireless sensors networks; internet of things; metamaterials; metasurfaces; printed electronics

Academic Editor: Federico Alimenti

Received: 9 February 2021 1. Introduction Accepted: 30 March 2021 Published: 30 April 2021 Radio frequency identiﬁcation (RFID) is a low-cost wireless technology that makes possible the connection of billions of things, enabling consumers and businesses to engage,

Publisher’s Note: MDPI stays neutral identify, locate, transact, and authenticate products [1]. The general RFID market has seen with regard to jurisdictional claims in a considerable growth over the past few years in terms of the number of RFID tags sold. published maps and institutional affil- The exploration of allied technologies such as RFID sensors has been enabled thanks to new iations. chipsets, both within the High Frequency (HF) band (NFC—near field communication) at 13.56 MHz, and within the Ultra-High Frequency (UHF) (RAIN—radio frequency identification) frequency band around 866 MHz (ETSI - European Telecommunications Standards Institute) or 910 MHz (FCC - Federal Communications Commission), which are dedicated to supporting several sensor platforms [2]. Even though battery-powered sensors have an Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. obvious advantage for data transmission over large distances, at the same time, a battery in- This article is an open access article creases the system’s complexity and maintenance issues, reduces system life, and limits the distributed under the terms and temperature range of sensor applications. Small form factors of batteries, such as thin-film conditions of the Creative Commons or other types of micro-batteries, are available on the market, but they need replacement Attribution (CC BY) license (https:// every few days [3]. RFID sensors can be fully passive, or battery powered; in the latter case, creativecommons.org/licenses/by/ they do not need a frequent change of the batteries as in the case of traditional wireless 4.0/). sensor nodes. In case of battery assisted RFID sensors or battery-assisted passive mode,

Sensors 2021, 21, 3138. https://doi.org/10.3390/s21093138 https://www.mdpi.com/journal/sensors Sensors 2021, 21, x FOR PEER REVIEW 2 of 35

Sensors 2021, 21, 3138 2 of 34

(http://creativecommons.org/li- or battery powered; in the latter case, they do not need a frequent change of the batteries censes/by/4.0/). as in the case of traditionala simple wireless circuit sensor is nodes. built around In case theof battery memory assisted chip, thusRFID enabling sensors the chip to switch to a or battery-assisted passivelocal mode, energy-assisted a simple circuit mode is only built when around it senses the memory a certain chip, stimulus thus [4]. Instead, chip-based enabling the chip to switchpassive to a sensorslocal energy acquire-assisted the power mode required only when for activation it senses a from certain the reader through wireless stimulus [4]. Instead, chippower-based transfer. passive sensors acquire the power required for activation from the reader through wirelessDifferent power RFID transfer. sensors are currently proposed in terms of architecture, complexity, Different RFID sensorsand are system currently requirements. proposed A in chip-based terms of architecture, design, where complexity, the sensor is integrated inside the and system requirements.chip, A chip provides-based a design, reliable where configuration, the sensor since is integrated the sensing inside and the communication functions chip, provides a reliableare configuration separated. Since, since embedding the sensing the and sensor communication increases both functions tag size and cost, an alternative are separated. Since embeddingsolution the is the sensor functional increases integration both tag ofsize the and antenna cost, an and alternative the sensor component. The chal- solution is the functionallenge integration is then of to the transform antenna theand RFID the sensor tag antenna component. into aThe sensor. chal- In antenna-based RFID lenge is then to transformsensors, the RFID the tag response antenna is moreinto a dependent sensor. In onantenna the environment.-based RFID Today’ssen- generation of passive sors, the response is moretags dependent has the abilityon the toenvironment. sense several Today’s environmental generation parameters of passive such as light, humidity, tags has the ability to senseand temperature,several environmental becoming aparameters key technology such foras “object-based”light, humidity, services. The role of RFID and temperature, becomingsystems a key can technology be extended, for “object and even-based” involve services. ubiquitous The role computing of RFID by moving from simple systems can be extendedpassive, and even tags involve to smart ubiquitous tags that cancomputing perform by different moving functions, from simple thanks to integrated sensors passive tags to smart tagsand that a microcontrollercan perform different unit (MCU). functions Research, thanks is to also integrated active onsen- the so-called chipless RF sors and a microcontrollersensors unit (MCU). that do Research not employ is also Integrated active on Circuits the so- (ICs)called and chipless may offer RF the benefits of a longer sensors that do not employlife Integrated and lower Circuits cost. Chipless (ICs) and RFID, may also offer known the benefits as passive of a longer RFID sensors, are compatible life and lower cost. Chiplesswith planarRFID, also technology, known allowingas passive them RFID to sensors, be produced are compatible by roll-to-roll processing. with planar technology, allowingRFID them sensors to be are produced a new paradigm by roll-to for-roll the processing. internet of things (IoT). They have a limited RFID sensors are a newcost paradigm and negligible for the maintenance, internet of things which (IoT). make They them have appealing a limited for numerous applicative scenarios such as manufacturing, logistics, healthcare, agriculture, and food. They have cost and negligible maintenance, which make them appealing for numerous applicative attracted numerous research efforts due to their innovative potential in various application scenarios such as manufacturing, logistics, healthcare, agriculture, and food. They have fields [5]. In order to also address the relevance of the research topic at an industry level, attracted numerous research efforts due to their innovative potential in various applica- we report some figures obtained from the database Scopus. In particular, the number of tion fields [5]. In order to also address the relevance of the research topic at an industry patents published every year according to the keywords ‘RFID’ and ‘RFID sensor’ are level, we report some figures obtained from the database Scopus. In particular, the num- reported in Figure1a. The patents classified by registration offices are reported Figure1b ber of patents published every year according to the keywords ‘RFID’ and ‘RFID sensor’ to highlight the most relevant countries active on this technology. We observe that a very are reported in Figure 1a. The patents classified by registration offices are reported Figure high number of patents, i.e., more than 129,000, have been registered with the United States 1b to highlight the most relevant countries active on this technology. We observe that a patent office. Surprisingly, the number of papers published in journals and conferences very high number of patents, i.e., more than 129,000, have been registered with the United every year is one order of magnitude lower than the number of registered patents. States patent office. Surprisingly, the number of papers published in journals and conferences every year is one order of magnitude lower than the number of registered patents. 2,733 15,222

-USPO 16,455 -Japan PO I IEU PO -WIP

I IUKI PO 26,947

159,704

(a) (b) (b)

FigureFigure 1. 1.Number Number of of ( a(a)) patents patents indexed indexed on on the the Scopus Scopus database database in in February February 2021 2021 according according to to the the keyword ‘RFID’ and the keywordkeyword ‘RFID‘RFID’ Sensor’. and the (keywordb) Number ‘RFID of patents Sensor’ issued. (b) Number by the of main patents patent issued offices. by the Acronyms: main patent United States Patent and Trademarkoffices. Acronyms: Office (US United PO), States Japan Patent and Office Trademark (Japan PO), Office European (US PO), Patent Japan Office Patent (EU Office PO), (Japan World Intellectual Property PO), European Patent Office (EU PO), World Intellectual Property Organization (WIP), United Organization (WIP), United Kingdom Intellectual Property Office (UKI PO). Source: Scopus. Kingdom Intellectual Property Office (UKI PO). Source: Scopus.

2. From Wi-Fi Nodes to RFID Sensors

Sensors 2021, 21, 3138 3 of 34

2. From Wi-Fi Nodes to RFID Sensors The typical hardware platform of a wireless sensor node consists of a sensor, a microcontroller, a radio frequency transceiver, and a power source. Each node is equipped with a physical sensor for revealing parameters of interest, such as light, temperature, sound, pressure, or other physical phenomena. The architecture of a wireless sensor node is shown in Figure2. Power consumption, chip footprint, and computational power, as well as on-chip memory are very important features for a microcontroller. To this aim, choosing a low-power-consumption transceiver that connects the node to the network is crucial for saving power. In fact, current consumption of a transceiver takes up most of the power consumption budget. Transceivers are available from various manufacturers, such as Inﬁneon Technologies AG (Neubiberg, Germany) [6], Analog Devices Inc. (Norwood, MA, USA) [7], and others. A crucial characteristic of a sensor node is the power source and battery management, especially in wireless sensor networks (WSNs), where the battery is irreplaceable. The node typically also includes a circuit that adjusts the RF front-end to compensate for changes of the RF front-end caused by the sensing element [8].

Sensors Antenna MCU Transceiver

MCU

Battery

Figure 2. Simpliﬁed sketch of classical wireless sensor architecture.

Although battery technology is mature, the mean time to replacement is only one year, or two, even, for relatively large batteries [9]. This means that for a sensor network with a considerable number of devices installed, several batteries need to be replaced every few days, which is unfeasible for many applications. Finally, systems based on battery- powered communication still cost more than USD 5 a unit, with no clear commercial or technical solution to achieve a cost lower than one dollar [9]. A possible way to overcome this problem is represented by ambient-power scavenging, which can, in principle, supply power indefinitely. In this context, an augmented version of RFID technology with sensing capabilities can represent the turn-around point for the massive development of the internet of things (IoT). RFID sensors can be of different types, and are based on many different approaches. Some of the sensors are already available on the market, while other are still in development and are not yet mature enough for market applications. A classification of RFID sensors covering both chip-equipped tags and chipless tags is shown in Figure3. The former set includes configurations in which the sensor is integrated directly into the tag (electronic sensors), whereas the latter rely on the modification of the tag response due to a detuning of the tag antenna for sensing (electromagnetic sensors). On the other hand, the chipless coun- terpart can exploit the properties of piezoelectric materials (acoustic sensors), the changes in the electromagnetic response of the tag (electromagnetic sensors), or thin-film transistors (electronic sensors) for sensing purposes. Sensors 2021, 21, 3138 4 of 34 Sensors 2021, 21, x FOR PEER REVIEW 5 of 36

Sensors 2021, 21, x FOR PEER REVIEW 5 of 36

FigureFigure 3. 3. ClassificationClassification of of RFID RFID sensors. sensors. Acronyms: Acronyms: SAW SAW (surface (surface acoustic acoustic wave ), wave),TFT (thin TFT film (thin transistor). Figure 3. Classificationfilm transistor). of RFID sensors. Acronyms: SAW (surface acoustic wave), TFT (thin film transistor). Finally,Finally, it it is is important important to to point point out out that that each each one one of of the the mentioned mentioned solutions solutions for for sensing sens- Finally, itpossesses ising important posses prosses to andpros point cons. and out cons. Consequently, that eachConsequently, one of the the user mentionedthe has user to has decide solutions to decide the best for the sens- trade-off, best trade depending-off, de- ing possesses onprospending the and goal cons.on they the Consequently, wantgoal the toy achieve. want the to user Aachieve. summary has to A decide summary of the the characteristics best of the trade characteristics-off,of de- various of kindsvarious of pending on thesensorskind goals oft ishe illustratedsensorsy want isto illustrated inachieve. Table1, A wherein summary Table it is1, emphasized, whereof the itcharacteristics is emphasized, once again, of that oncevarious higher again, performance that higher kinds of sensorscomesperformance is illustrated at the cost come in of Tables a at higher the 1, costwhere price. of ita higheris emphasized, price. once again, that higher performance comes at the cost of a higher price. Table 1. Industrial IoT Technologies. Table 1. Industrial IoT Technologies. Table 1. Industrial IoT Technologies. Wired Wireless RFID Chipless RFID Wired Wireless RFID Chipless RFID Wired WirelessSensors RFIDSensors SensorsChipless RFID Sensors Sensors Sensors Sensors Sensors InstallationSensors cost SensorsHigh SensorsLow LowSensors Low Installation costInstallationMaintenance costHigh cost HighLowLow LowHigh Low Low Low Low Low Low Maintenance costMaintenance HardwareLow cost cost LowHighHigh LowMediumHigh Low Low Low Low Low Hardware cost High Medium Low Low Hardware cost ScalabilityHigh MediumLow LowHigh HighLow High Scalability Low High High High Scalability Sensor accuracyLow HighHigh HighHigh MediumHigh Low Sensor accuracy High High Medium Low Sensor accuracy High High Medium Low 3. Chip Based RFID Sensors 3. Chip Based3. RFID ChipThere Sensors Based existRFID two Sensorsimplementations of a sensor employing a chip, namely the electronic There existRFID twoThere implementations sensor exist and two the implementations electromagnetic of a sensor employing of one. a sensor A a sketchchip, employing namely representing the a chip, electronic the namely two topologies the electronic is RFID sensor andRFIDshown the sensor electromagneticin Figure and 4 the. Both electromagnetic one.configurations A sketch one.representing contain A sketch the same the representing two functional topologies units the two is, but topologies with a fun- is shown in Figureshowndamental 4. Both in Figureconfigurationsdifference4. Both in the contain configurations sensing the part. same containIn functional the electronic the units same ,configuration but functional with a fun- units,, the sensing but with unit a damental differencefundamentalinteracts in thewith differencesensing the IC ,part. whereas in theIn the sensing in electronicthe electromagnetic part. Inconfiguration the electronic configuration, the configuration, sensing, the unit sensing the sensing operation unit interacts withinteracts theis based IC, whereas withon the the changein IC, the whereas electromagnetic in the tag in thefrequency electromagnetic configuration response., theIn configuration, both sensing cases, operation the the primary sensing constraint operation is based on theis thatchange based determines onin the the tag change the frequency maximum in the response. tag reading frequency In distanceboth response. cases, of thethe In primary bothsensor cases, tag constraint theis represented primary constraint by the that determinesthatvalue the determines maximumof the voltage thereading maximumthat isdistance available reading of atthe the distance sensor chip [10]. tag of theis represented sensor tag by is representedthe by the value of the voltagevalue ofthat the is voltageavailable that at isthe available chip [10]. at the chip [10].

(a) (b) Figure 4. RFID(a) -based sensor topology: (a) electronic RFID sensor(b) and (b) electromagnetic RFID sensor architecture. Figure 4. RFID-basedFigure 4. RFID sensor-based topology: sensor topology: (a) electronic (a) electronic RFID sensor RFID and sensor (b) electromagneticand (b) electromagnetic RFID sensor RFID architecture. sensor architecture.

Sensors 2021, 21, 3138 5 of 34

3.1. Electronic RFID Sensors Tags Electronic sensor tags have the peculiarity of separating the sensing functions from the communication functions [11]. The transmitted information is digitally encoded, and thus the sensed information is essentially insensible to the environment. In particular, some bits of the ID code can be used to transmit the value of the sensed parameter. The sensor can be integrated inside the chip, or interfaced though an external microcontroller, thus obtaining an augmented tag architecture. In both cases, the energy efﬁciency of the tag is a key issue for achieving an acceptable reading range. When the reading range is not enough, a battery is included within the tag in the so-called battery assisted tags. In these cases, the tag acts like a transponder, and thus it sends the information to the reader only if interrogated. When the tag has to continuously broadcast the sensor information, an active conﬁguration has to be used.

3.1.1. Active Sensor Tags In active RFID tags equipped with sensing capabilities, a sensor is integrated into the tag, and the tag IC communicates with the sensor in order to get the information on the monitored quantity, and thus to include it into the backscattered bit sequence. The sensing activity is therefore performed in DC with a dedicated electronic component. The integration of the sensors within the chip circuit requires the design of dedicated chips with specific sensing features. Being equipped with a battery, active tags provide a much longer read range than passive tags. Active RFID systems generally work at 433 MHz or 2.45 GHz [12], but 433 MHz is usually preferred by companies because of the longer wavelengths, which result in being more suitable for materials like metal and water. Typically, active RFID tags are powered by a battery that will last between 3–5 years. Active RFID tags can be employed both as transponders and beacons. Beacons have a read range of hundreds of meters but, in order to limit battery consumption, their transmitting power can be set to reach a read range of around 100 m. Active RFID tags can cost from $20 to $100+ depending on the tag specifications and on its capability to withstand harsh conditions [12]. Active RFID tags can be equipped with motion and temperature sensors. Recently, novel configurations of 2.4 GHz and 5.7 GHz active tags are available, which can be configured to activate when moved, record temperature, humidity or other environment parameters, and transmit signals periodically. The most typical sensing applications of active tags are as follows [13]: - Real-time locating systems (RTLS): it is possible to determine the location of active tags with correct antenna placement. - Patient tracking: the movements of patients can be monitored within a certain area with a wrist band tag. - Temperature sensing: active tags can be used to sense temperature variation in time with alarms set if a certain threshold is overtaken. - Motion sensing: the tag can be set up to transmit a signal when moved or a signal until movement stops. This is used to control the movement of high value assets and for avoiding intrusions.

3.1.2. Battery Assisted Sensor Tags The battery assisted passive (BAP) class is comprised of RFID tags with an embedded battery. When the reader interrogates the BAP tag, the embedded battery is turned on, as well as the RFID tag IC and other sensors or actuators within the tag [14]. BAP RFID tags’ read range is clearly greater than passive ones, but is shorter than active tags. The operative life of these tags is limited by the battery, which is usually not replaceable. As an example, the Monza X-2K RFID chip manufactured by Impinj Inc. (Seattle, WA, USA) [15] provides a performance boost via a ‘battery-assisted passive mode’ [4], when both read sensitivity (−17 dBm) and write sensitivity (−12 dBm) of the chip increase up to −24 dBm when a DC voltage is provided [4]. Figure5 illustrates a scheme of RFID tags relying on battery- assisted sensors. To limit the use of the battery, thus increasing the battery-life, energy Sensors 2021, 21, x FOR PEER REVIEW 7 of 36 Sensors 2021, 21, 3138 6 of 34

on battery-assisted sensors. To limit the use of the battery, thus increasing the battery-life, harvestingenergy harvesting solutions solutions have been have proposedbeen proposed and integratedand integrated into into both both active active and and battery- bat- assistedtery-assisted RFID RFID tags [ 16tags]. Technologies[16]. Technologies relying relying on thermoelectric on thermoelectric effects, effect photovoltaics, photovoltaic effects, oreffect piezoelectrics, or piezoelectric effects have effect beens have adopted been adopted for harvesting for harvesting energy energyfrom the from surrounding the sur- environment.rounding environment. Conversion Conversion of biomechanical of biomechanical energy into energy electricity into electricity has also beenhas also proposed been inproposed [17]. in [17].

FigureFigure 5. 5.RFID RFID tagstags withwith battery-assistedbattery-assisted sensors.

3.1.3.3.1.3. Battery-LessBattery-Less SensorsSensors (Passive)(Passive) SolutionsSolutions thatthat provideprovide sensors integrated integrated within within the the chip chip have have been been recently recently pro- pro- posedposed [ 10[10,18].,18]. AA sensorsensor is embedded and communicates communicates with with the the IC IC to to deliver deliver information information onon thethe monitoredmonitored quantity,quantity, whichwhich has to be encoded into into the the backscattered backscattered signal. signal. Elec- Elec- tronictronic RFIDRFID sensorssensors cancan exploitexploit the HF or UHF frequency frequency range. range. An An example example of of a areader reader forfor HFHF sensorssensors isis representedrepresented byby smartphonessmartphones exploiting exploiting the the near near field field communication communication (NFC)(NFC) paradigm. paradigm. For For UHF UHF tags, tags, which which cover cover a a longer longer communication communication range, range, the the amount amount of powerof power rectified rectified by theby the tag tag must must be be sufficient sufficient to powerto power up up both both the the chip chip and and the the embedded embed- sensor.ded sensor. In this In case,this case the,value the value of the of voltagethe voltage that that is available is available at theat the chip chip [10 [10]] is pivotalis pivotal for thefor readingthe reading range range extension extension [19 [19].]. Obviously, Obviously, the the sensor sensor must must bebe able to to be be miniaturized miniaturized inin order order toto bebe embeddedembedded intointo thethe chip,chip, and few examples examples are are currently currently available available in in the the literatureliterature dealing dealing withwith temperature, light light,, and and pressure pressure monitoring. monitoring. However, However, the the passive passive RFIDRFID tagtag chipchip isis soso sensitivesensitive to power consumption that that it it is is difficult difficult to to embed embed sensors sensors andand an an analogue-to-digital analogue-to-digital converter converter (ADC) (ADC) maintaining maintaining a reasonable a reasonable operating operating range range [20 ]. Even[20]. Eve an ADCn an ADC with with a power a power consumption consumption of of several severalµ µWW[ [21]21] willwill reduce the the operating operating distancedistance ofof thethe tagtag dramatically.dramatically. AA newnew augmentedaugmented RFID tag, the WISP WISP ( (wirelesswireless identification identification and and sensing sensing platform platform),), hashas beenbeen developeddeveloped byby the Intel Research Centre Centre [22]. [22]. WISP WISP o operatesperates as as a astandard standard tag tag,, butbut offersoffers thethe possibilitypossibility toto be connected to an external external sensor sensor by by using using a aprogrammable programmable microcontrollermicrocontroller unit.unit. However,However, thethe flexibilityflexibility offered by the the WISP WISP solution solution is is paid paid for for with with anan increasedincreased price. Battery Battery-less-less configurations configurations have have recently recently been been developed developed in which in which the themicrocontroller microcontroller requires requires a few a fewmilliwatts milliwatts [4]. Moreover, [4]. Moreover, they offer they the offer possibility the possibility of gath- of gatheringering the energy the energy required required to drive to drive the data the acquisition data acquisition from the from interrogat the interrogationion signal, signal,as in asthe in case the caseof conventional of conventional passive passive tags. tags.Another Another option option is represented is represented by piezoelectric by piezoelectric en- energyergy scavengers scavengers transforming transforming micro micro-oscillations-oscillations into electrical into electrical energy, energy, which which are the are most the mostpowerful powerful and versatile and versatile ones, ones,although although the power the power requirement requirementss may reduce may reduce the real the data real datarate. rate.

3.2.3.2. ElectromagneticElectromagnetic RFIDRFID Sensors Tags TheThe effectiveeffective permittivitypermittivity of an object can can be be inferred inferred by by using using any any conventional conventional RFID RFID tagtag attachedattached to it. it. In In fact, fact, quantities quantities such such as the asthe input input impedance impedance or gain or gainof the of RFID the RFIDtag, tag,as in as the inthe case case of of any any antenna, antenna, are are altered altered by by the the surrounding surrounding environment. environment. The The RFID RFID backscatteredbackscattered signalsignal is affected by any change of of the the tagged tagged object object,, since since this this is is shifted shifted into a correspondent change in the tag’s antenna parameters. Therefore, the antenna itself becomes the sensor, even though it is completely sensor-less [23].

Sensors 2021, 21, x FOR PEER REVIEW 8 of 36

into a correspondent change in the tag’s antenna parameters. Therefore, the antenna itself becomes the sensor, even though it is completely sensor-less [23]. Electromagnetic tags rely on the modification of the tag response due to an occasional modification of the antenna. The antenna behavior can be modified essentially for two reasons: a change in the electrical conductivity of the antenna, or a part of it; or because of a change of the dielectric permittivity of the medium surrounding the antenna, or a part of it. In the first case, the sensor is classified as resistive, and in the second case, the sensor can be classified as capacitive. Notably, in both cases, the transduction mechanism is influencing radio frequency waves and thus the sensors are referred to as electromagnetic. A possible layout of the RFID sensor is shown Figure 6a, where a dipole antenna is connected in series with a sensing unit and the RFID chip [24]. A simplified equivalent Sensors 2021, 21, 3138 circuit model of an electromagnetic RFID-based sensor tag is shown in Figure 6b. The7 of 34 sensing material, e.g., carbon nanotubes, Kapton, or other substances, which change their properties as a function of an external stimulus, are placed in the sensing unit. The variations of the sensing unit properties impact on the matching between the antenna and the Electromagnetic tags rely on the modification of the tag response due to an occasional chip impedance, and thus they induce a change of the reflected power. Depending on the modification of the antenna. The antenna behavior can be modified essentially for two value of the sensed parameter, the impedance mismatch between the antenna port and reasons: a change in the electrical conductivity of the antenna, or a part of it; or because of a the microchip is enough to enable (or not) the microchip to harvest the necessary power change of the dielectric permittivity of the medium surrounding the antenna, or a part of it. and respond to the reader with its own digital identification code (IDn) [23]. In the first case, the sensor is classified as resistive, and in the second case, the sensor can be The two main mechanisms for detecting a variation in the sensor response are a fre- classified as capacitive. Notably, in both cases, the transduction mechanism is influencing quency shift (f) of a resonant peak or a magnitude change (α) in the reflection coeffi- radio frequency waves and thus the sensors are referred to as electromagnetic. cient, as depicted in Figure 6c–d. The frequency shift method is not compatible with RFID A possible layout of the RFID sensor is shown Figure6a, where a dipole antenna is chip-enabled technology, since the resonant frequency shift of the antenna leads the in- connected in series with a sensing unit and the RFID chip [24]. A simplified equivalent cir- formation content far away from the operating frequency (868 MHz or 915 MHz). There- cuit model of an electromagnetic RFID-based sensor tag is shown in Figure6b. The sensing fore, magnitude change configuration is the most popular approach for RFID-based sen- material, e.g., carbon nanotubes, Kapton, or other substances, which change their prop- sors. erties as a function of an external stimulus, are placed in the sensing unit. The variations An alternative sensor topology consists of equipping the tag with a real sensor (mo- of the sensing unit properties impact on the matching between the antenna and the chip tion, temperature, pressure, or other), which could be either connected in a region of the impedance,tag’s antenna and or distributed thus they induceover the a antenna change surface of the reflectedas a paint. power. The variation Depending in the on im- the valuepedance of theloading sensed caused parameter, by the the change impedance of the environment mismatch between will produce the antenna a change port of and the the microchiptag’s radar is cross enough-section to enable and, hence (or not), a backscattered the microchip power to harvest modulation the necessary, as in the power case andof respondsensor-less to thetags reader [23]. with its own digital identification code (IDn) [23].

1 Sensing unit

Ra Ra Chip unit

La La Zchip Antenna

Ca Sensing Ca

Sensors 2021, 21, x FOR PEER REVIEW 9 of 36 Chip 1’

(a) (b)

f

a

Amplitude Amplitude

frequency frequency

(c) (d) Figure 6. (a) Equivalent circuit model of an electromagnetic RFID sensor. (b) Possible layout of the sensor. Working prin- Figure 6. (a) Equivalent circuit model of an electromagnetic RFID sensor. (b) Possible layout of the sensor. Working principle ciple of frequency domain RFID sensors based on (c) capacitive or (d) resistive transduction mechanisms. of frequency domain RFID sensors based on (c) capacitive or (d) resistive transduction mechanisms. 3.2.1. Self-Tuned Chips The two main mechanisms for detecting a variation in the sensor response are a frequencyIn the shift case (∆ f)of ofelectromagnetic a resonant peak RFID or sensor a magnitude tags, the changeantenna ( ∆accomplishesα) in the reﬂection two dif- coefﬁcient,ferent functions, as depicted namely in Figure the communication6c,d. The frequency and the shift sensing method tasks. is However, not compatible they cannot with RFIDbe pursued chip-enabled independently technology, or maximized since the resonant at the same frequency time, since shift the of sensing the antenna is performed leads theat information the expense content of the communication far away from function. the operating In fact, frequency the tag must (868 suffer MHz ora certain 915 MHz). level of mismatch in order to indirectly provide information about the variation of the measurements. The reason for this is that the sensing information is strictly related to the detuning level of the RFID tag, and hence the communication and sensing capabilities have conflicting requirements. A recent solution to this problem has been offered by self-tuning microchips [25–27] that can cope with the undesired tag detuning without harming the sensing function [28]. The equivalent circuit of an RFID tag with self-tuning capability is illustrated in Figure 7a. The working principle can be explained by considering that the RFID tag is placed on an item that drastically alters its antenna impedance with respect to the free space case. This means that the original antenna is detuned, and the impedance ZA = RA + jXA, which was designed to properly match the impedance of the chip, ZC, may be far from the optimal condition ZC = (ZA)*, and therefore the communication may be severely compromised.

(a) (b) Figure 7. (a) Equivalent circuit of an RFID tag equipped with a self-tuning chip and (b) an example of the exploitation of the autotuning feature for indirect sensing.

Sensors 2021, 21, x FOR PEER REVIEW 9 of 36

f

Sensors 2021, 21, 3138 a 8 of 34

Amplitude Amplitude

Therefore, magnitude change conﬁguration is the most popular approach for RFID-based sensors. An alternative sensor topology consists of equipping the tag with a real sensor (motion, temperature,frequency pressure, or other), which could be either connectedfrequency in a region of the tag’s antenna or distributed over the antenna surface as a paint. The variation in the impedance (loadingc) caused by the change of the environment will produce(d) a change of the tag’s radar cross-section and, hence, a backscattered power modulation, as in the case of sensor-less Figure 6. (a) Equivalent circuit model of an electromagnetic RFID sensor. (b) Possible layout of the sensor. Working prin- tags [23]. ciple of frequency domain RFID sensors based on (c) capacitive or (d) resistive transduction mechanisms. 3.2.1. Self-Tuned Chips 3.2.1. Self-Tuned Chips In the case of electromagnetic RFID sensor tags, the antenna accomplishes two different In the case of electromagnetic RFID sensor tags, the antenna accomplishes two dif- functions, namely the communication and the sensing tasks. However, they cannot be ferent functions, namely the communication and the sensing tasks. However, they cannot pursued independently or maximized at the same time, since the sensing is performed be pursued independently or maximized at the same time, since the sensing is performed at the expense of the communication function. In fact, the tag must suffer a certain at the expense of the communication function. In fact, the tag must suffer a certain level level of mismatch in order to indirectly provide information about the variation of the of mismatch in order to indirectly provide information about the variation of the meas- measurements. The reason for this is that the sensing information is strictly related to the urements. The reason for this is that the sensing information is strictly related to the de- detuning level of the RFID tag, and hence the communication and sensing capabilities have tuning level of the RFID tag, and hence the communication and sensing capabilities have conﬂicting requirements. A recent solution to this problem has been offered by self-tuning conflicting requirements. A recent solution to this problem has been offered by self-tuning microchips [25–27] that can cope with the undesired tag detuning without harming the microchips [25–27] that can cope with the undesired tag detuning without harming the sensing function [28]. The equivalent circuit of an RFID tag with self-tuning capability is sensing function [28]. The equivalent circuit of an RFID tag with self-tuning capability is illustrated in Figure7a. The working principle can be explained by considering that the illustrated in Figure 7a. The working principle can be explained by considering that the RFID tag is placed on an item that drastically alters its antenna impedance with respect to RFID tag is placed on an item that drastically alters its antenna impedance with respect to the free space case. This means that the original antenna is detuned, and the impedance the free space case. This means that the original antenna is detuned, and the impedance ZA = RA + jXA, which was designed to properly match the impedance of the chip, ZC, ZA = RA + jXA, which was designed to properly match the impedance of the chip, ZC, may may be far from the optimal condition ZC = (ZA)*, and therefore the communication may be far from the optimal condition ZC = (ZA)*, and therefore the communication may be be severely compromised. severely compromised.

(a) (b)

FigureFigure 7. 7.( a()a) Equivalent Equivalent circuit circuit ofof anan RFIDRFID tagtag equippedequipped with a a self self-tuning-tuning chip chip and and (b (b) )an an example example of of the the exploitation exploitation of of the autotuning feature for indirect sensing. the autotuning feature for indirect sensing.

The integrated circuit of a self-tuning RFID tag is able to guarantee a proper level of matching thanks to its ability to autonomously adjust the chip’s impedance to the current antenna impedance. This goal is achieved for the chip via a set of capacitors that can be selectively activated. More speciﬁcally, the chip reactance spans from a minimum value of Cmin, where all the other N selectable capacitors, CS, are switched out, up to the maximum value of Cmax = Cmin + N CS that is attained when all the capacitors are activated. The con- ﬁguration status of all the capacitors in the switching network is automatically selected to assure the maximum power transfer from the antenna to the microchip. This means that the microchip neutralizes the undesired changes of the RFID antenna impedance with the aim of reaching the closest condition to perfect matching. Sensors 2021, 21, 3138 9 of 34

The adaptive matching feature exhibited by this kind of RFID tag can also be advan- tageously exploited for sensing purposes. In fact, the self-tunable chip sends to the RFID reader the information written in its memory, as well as the state of the selectable capacitors. This string of bits is related to the detuning undergone by the RFID tag, and indirectly provides information on the environmental change in which the tag is operating. It is then possible to infer the value of a physical parameter that affects the tag from the state of the automatic tuning network. Figure7b illustrates an applicative scenario in which an RFID tag is applied to a container in which the level of water (hWATER) can change over time. Every time the tag is interrogated, it also provides the status of the internal matching network that allows it to be matched and running in the complex operational environment. The encoded sequence of the capacitors selected in the tunable matching network at time t1 is embedded in the backscattered signal. When the water level is changed at time t2, the tunable matching network is set to the new status of the capacitors, since the effect of the environment has changed. The change of the matching network settings that has been encoded in the backscattered signal allows for indirectly estimating the level of the water. Obviously, the limited number of capacitors in the switching network, and also their range of values, have an impact on the resolution of the monitored physical parameter [29].

3.2.2. Harmonic Tags An alternative implementation of RFID sensors is based on the so-called harmonic or secondary radar [30]. In this configuration, the tag receives the interrogation at a certain frequency, and it scatters back at a harmonic frequency. A typical implementation of harmonic tags employs the second harmonic frequency (2f0). The reason for using harmonic tag configuration relies on the possibility of obtaining a better immunity of the tag response in the presence of clutter. Indeed, it can be reasonably assumed that most of the objects surrounding that tag do not possess nonlinear properties, which may cause the reflection of harmonic frequencies. An harmonic transponder is usually synthesized through a Schottky diode and an antenna [30–32]. The Schottky diode should be able to perform the frequency multipli- cation as efficiently as possible to 2f0. The antenna allows for radiating the upconverted frequency signal. One issue is that the antenna matching should be guaranteed both at the fundamental and the second harmonic frequencies. This aspect makes the transponder design more complex if compared to standard RFID tags. An additional challenge is that the tags need to comply with existing frequency regulations.

4. Chipless Sensors Chipless RFID sensor tags, like the electromagnetic RFID sensors, exploit the changes in antenna behavior that is dependent on the change in the physical environmental parameter that has to be measured [33]. However, differently from classical electromagnetic RFID sensors, the tag does not include a chip. It is basically a simple passive antenna or resonator. An electromagnetic wave impinging onto the chipless tag is mostly backscattered at the resonance frequency of the resonator, which acts as a spatial filter. Therefore, by observing the spectral response of the backscattered signal, a peak can be observed at the resonance frequency of the resonator, or resonant frequencies, in the case of multiple resonators. This technology appears to be promising for designing low-cost, green, and printable sensors [34]. The sensing capabilities are obtained through a frequency shift of the resonant peaks in the backscattered response generated by the change in external parameters surrounding the tag. The absence of a chip and a battery gives the opportunity to significantly decrease the cost of the sensor and to achieve a theoretically infinite lifetime. Given the absence of any electronic circuit, chipless RFID sensors are potentially suitable for harsh environments [33,35–37]. Clearly, one of the main limitations is that the reliable reading of the sensor can be guaranteed only under specific conditions. Chipless RFIDs rely on passive transduction mechanisms such as capacitance or surface resistance change, but even a mechanical perturbation of an RF resonatorallows for realizing cheap devices [36,38,39]. Sensors 2021, 21, 3138 10 of 34

Printable electronic circuits can also be included in the chipless category. A summary of three main categories of chipless sensors is shown in Table2. The accuracy of chipless tags is expected to be more limited with respect to chip-equipped sensors, but their study is justiﬁed by their extreme fabrication simplicity, which would allow for integrating these sensors inside conventional packaging at a minimal cost. Some examples of chipless RFID sensors that are commercially available are summarized in Table3. As it is apparent, the employed technologies are quite broad-spectrum, spanning from the exploitation of the Barkhausen effect to radar imaging.

Table 2. Comparison between main chipless RFID technologies.

Printed Electronics EM Scattering SAW Read range <10 cm <1 m <10 m Working band 13 MHz 10–100 MHz/3–8 GHz 2.45 GHz Maturity level Medium Medium High Temperature range −100◦–300◦ −100◦–300 Standard HF-RFID No Some Cost NA Very low (<1 cent) Low (<10 cents) Anticollision Good Low Good Read speed Low Low High Threshold power Yes No No Bits High Low High Rewritable Yes No No Market available No No Yes Integration with packaging Yes Yes No Sensing Yes (various) Yes (various) Yes (temperature)

4.1. Electromagnetic Chipless RFID Sensors Two general types of chipless RFID tags can be identiﬁed: time domain (TD)-based and spectral (frequency) signature-based (frequency domain, FD). Time domain chipless sensors are comprised of a wideband antenna and a delay line. The tag receives and retransmits a modiﬁed version of the pulse transmitted by the reader. The time domain radar cross section (RCS) of the tag is comprised of a structural and an antenna component. The delay line is used to separate the antenna RCS peak from the structural one, which is the most intense. Depending on the load of the delay line, the antenna response of the tag can be modulated or delayed. Both these mechanisms can be used to design a chipless sensor. Amplitude modulation can be obtained by ending the delay line with a material able to change its surface resistance as a function of an external phenomenon. The sensed information can be also included in the time shift from the antenna peak [40]. Time domain- based chipless RFID tags manufactured with printed circuit board technology are large and are unable to encode a high number of bits [41].

Table 3. Chipless RFID sensor tags that are commercially available. From [42].

Company Tecnology HID Barkhausen magnetic access card AstraZeneca Acoustomagnetic Syringe tag RF saw, Microdesign SAW Menippos, M-real, Acreo Conductive stripes Inksure Radar Array Myyake, Navitas, RF code LC array Flying null, Conﬁrm technologies Electromagnetic

The second class of chipless tags encodes data into the spectrum using resonant structures [43–47]. Two conﬁgurations have emerged for FD chipless tags. The former is based on two orthogonally polarized antennas with a series of resonators in between [46], whereas the latter employs several resonators of different size [47]. In FD chipless sensors, Sensors 2021, 21, 3138 11 of 34

the idea is to add a sensor function to the tag. This goal can be achieved by exploiting the Sensors 2021, 21, x FOR PEER REVIEW 13 of 36 permittivity variation of a chemical interactive material (CIM) that is placed on the tag antenna. The CIM can cause a frequency shift in the backscattered signal or a change in the amplitude response, as illustrated in Figure6. Different environmental parameters can be monitoredBarium by exploitingstrontium titanate the interaction (BST) of the CIM withTemperature resonators. Temperature, pressure, humidity, andZeolite gas sensors Material have been proposed in [48Ammonia–51]. Inkjet printing can be also efﬁciently exploited forPaper the fabrication of chipless sensorsHumidity [52–54]. Several polymers and nanomaterials have been used to synthesize sensors because of their ability to change theirThe properties use of a assmall a function tag provides of environmental a weak backscattered parameters signal such, as which humidity, can hardly temperature, be distinguishedand gasses, or from due the to mechanicalbackground stresses response. and Beyond pressure. the Thesespecific materials topology are of namedthe tag, CIMthe as mainthey limitation are able to of operate chipless a transductiontechnology is fromthat the a physical tag detection quantity requires to an electricallya calibration measured pro- cedureone. In often Table based4, some on background materials reported subtraction. in the For recent this reason, literature some are strategies categorized to increase with their thefunctionalization immunity of the property. tag response The tablewith respect is based to on the [55 environment,56]. A simpliﬁed have been scheme investigated. of the sensing Severalmechanism solutions based have on chiplessbeen proposed tags is shownin order in to Figure overcome8. this fundamental problem, such as adopting encoding/decoding schemes based on cross-polarization [57,58], circular polarizationTable [59], 4. differentialSensing polymers polarization and nanomaterials. encoding [60], or synthetic aperture radar (SAR) based approaches [44]. To further improve the system reliability, the above strategies can Category Material Name Sensing be jointly adopted with time domain gating. Time domain gating allows for filtering out ® Polymers some undesiredStanyl TE200F6 contributions, such as antenna coupling and multipath.Temperature However, the Polyamide kapton Humidity presence of large objects behind the tag is difficult to remove with time domain gating, Polyvinyl alcohol (PVA) Humidity sincePoly (2-Hydroxyethylthe time window methacrylate) should fulfill a couple of opposite requirements: Humidity the time window shouldPolydimethylsiloxane be long enough to (PDMS) include all the time domain responses of pH, the pressure resonant tag, which isPolymethyl inversely related methacrylate to the frequency (PMMA) bandwidth of the peak, but, Humidity, at the same methanol time, it should PEDOT:PSSbe trunca (poly(3,4-ethylenedioxythiophene)ted to remove the contribution of the objects close to the tag [60]. If the object is Temperature, humidity, pH too closepolystyrene to the tag, sulfonate) the time gating may be not sufficient to completely remove the undesired contributions.cellulose The possibility of correctly reading Chemical the information sensors (humidity, encoded ion, within gas) Metal and metal oxide semiconductor (MOS): SnO2, Nanomaterials the tag (detection probability) is strictly related to the RCS of the tag.Gasses The RCS of the tag ZnO and TiO2 is proportional to the square of the footprint of the label if all the particles radiate in phase Nanoparticle-based inks Strain [61]. InCarbon the case nanotubes of periodic (CNTs) tags, the average value Gas, of RCS temperature, can be controlled pressure/mechanical by increasing strain or decreasing the number of unit cells, and thus theStrain, footprint temperature, of the tag biological, [61]. Finally, gasses it (NO2,has Graphene to be mentioned that readers for chipless RFID require complexNH3 and andspecialized CO) architec- turesBarium based strontium on their titanate mode (BST)of operation, and competitive commercial Temperature products are currently available.Zeolite Examples Material of reader prototypes include frequency Ammonia-modulated continuous wave (FMCW) Paperor ultra-wideband (UWB) impulse radar-based devices Humidity [62,63].

FigureFigure 8. 8.ChiplessChipless RFID RFID sensor sensor architecture architecture and and possible possible layout layout of a of chipless a chipless RFID RFID sensor. sensor.

4.2. AcousticThe use Chipless of a small RFID tagSensors provides: SAW aTags weak backscattered signal, which can hardly be distinguishedA surface acoustic from the wave background (SAW) tag response. [64] is comprise Beyondd of the a piezoelectric speciﬁc topology substrate, of the an tag, interthe- maindigital limitation transducer of (IDT), chipless some technology metal reflector is thats, the and tag antennas. detection When requires the RF a calibration reader interrogatesprocedure the often tag, based the tag on antenna background collects subtraction. the EM probing For wave. this reason, The IDT some then strategiesconverts to theincrease electronic the energy immunity into ofmechanical the tag response energy through with respect the inverse to the piezoelectric environment effect. have The been SAWinvestigated. undergoes Several a series solutions of reflections have been that proposedoperate the in encoding order to overcome of the signal this, fundamentaland then theseproblem, acoustic such waves as adopting are reconverted encoding/decoding by the IDT schemesinto an EM based wave on by cross-polarization piezoelectric effect. [57, 58], Thiscircular signal polarization is finally transmitted [59], differential back to polarization the reader encodingby the antenna [60], or tag. synthetic The RF aperture operation radar frequency is limited by the substrate size and by the photolithographic process. SAW devices are manufactured in the frequency range between 30 MHz and approximately 3 GHz.

Sensors 2021, 21, 3138 12 of 34

(SAR) based approaches [44]. To further improve the system reliability, the above strategies can be jointly adopted with time domain gating. Time domain gating allows for filtering out some undesired contributions, such as antenna coupling and multipath. However, the presence of large objects behind the tag is difficult to remove with time domain gating, since the time window should fulfill a couple of opposite requirements: the time window should be long enough to include all the time domain responses of the resonant tag, which is inversely related to the frequency bandwidth of the peak, but, at the same time, it should be truncated to remove the contribution of the objects close to the tag [60]. If the object is too close to the tag, the time gating may be not sufficient to completely remove the undesired contributions. The possibility of correctly reading the information encoded within the tag (detection probability) is strictly related to the RCS of the tag. The RCS of the tag is proportional to the square of the footprint of the label if all the particles radiate in phase [61]. In the case of periodic tags, the average value of RCS can be controlled by increasing or decreasing the number of unit cells, and thus the footprint of the tag [61]. Finally, it has to be mentioned that readers for chipless RFID require complex and specialized architectures based on their mode of operation, and competitive commercial products are currently available. Examples of reader prototypes include frequency-modulated continuous wave (FMCW) or ultra-wideband (UWB) impulse radar-based devices [62,63].

4.2. Acoustic Chipless RFID Sensors: SAW Tags A surface acoustic wave (SAW) tag [64] is comprised of a piezoelectric substrate, an inter-digital transducer (IDT), some metal reflectors, and antennas. When the RF reader interrogates the tag, the tag antenna collects the EM probing wave. The IDT then converts the electronic energy into mechanical energy through the inverse piezoelectric effect. The SAW undergoes a series of reflections that operate the encoding of the signal, and then these acoustic waves are reconverted by the IDT into an EM wave by piezoelectric effect. This signal is finally transmitted back to the reader by the antenna tag. The RF operation frequency is limited by the substrate size and by the photolithographic process. SAW devices are manufactured in the frequency range between 30 MHz and approximately 3 GHz. SAW delay lines are based on the SAW propagation time delay, calculated as the ratio of acoustical length and SAW velocity (L/vSAW). In known sensor applications, L and vSAW, respectively, vary as a transducer effect determined by a temperature change, mechanical stress, and strain, and because of a mass loading from a thin surface layer. SAW sensors can measure physical, chemical, and biological parameters. As an example, a pressure sensor is reported in [65], where the monitored parameter is converted into a change in the sensor’s surface acoustic wave’s velocity. As shown in [66], by bending, stretching, and compressing the SAW substrate, sensors for torque, force, displacement, vibration, and acceleration can be synthesized [67]. The first SAW sensors were for temperature monitoring, and sensitive materials such as quartz (SiO2) were employed. In multi-sensor systems, different sensors are typically distinguished through frequency division multiplexing. However, space division multiple access (SDMA) and time division multiple access (TDMA) can also be employed [64]. Orthogonal frequency coding (OFC) has also been proposed for encoding the SAW sensor in multisensory environments [68].

4.3. Electronic Chipless RFID Sensors: Thin Film Transistors (TFT) An attractive solution for designing chipless RFIDs would be to manufacture the antenna and electronics on the same substrate. Organic electronics could represent the solution that enables the production of complete ultra-low-cost RFID tags. Thin ﬁlm transistor circuits (TFTCs) can be also employed in designing TD tags [69]. However, the current development stage of printed RFID tags is still not sufﬁcient to enable real-world applications [70]. Despite the availability of printed organic transistors with carrier mobilities over 1 cm2 V−1 s−1 and switching speed in the MHz range [71], silicon electronics strongly out- perform printed electronics in terms of carrier mobilities and switching speeds, as shown Sensors 2021, 21, 3138 13 of 34

in Table5. However, printed electronics still provide interesting characteristics related to manufacturing on soft substrates in large areas. Currently, transistor architectures have not been optimized for ICs because of the limited number of metal layers available [72].

Table 5. Performance comparison and utilization of printed electronics and silicon ICs in ﬂexible hybrid electronics (FHE) systems [73].

Performance Parameters Printed Electronics Silicon ICs Charge carrier mobility (cm2 V−1 s−1) ~1 (organics) ~1000 Switching speed (MHz) ~1 ~5000 Operating voltage (V) ~10 ~1 Lifetime (yr) ~0.1 ~10 Substrate softness—elastic modulus-1 (Gpa−1) ~2.5–1 (plastics) ~180–1

TFT-based RFID tags have been developed by companies or by research groups. Some developed tags can communicate only with specially designed RFID readers with custom protocols (like 8, 12, or 16-bit tags), whereas other designs are capable of communi- cating with commercial NFC readers, which, nowadays, are also integrated into cellphones. Clearly, if simplified protocols that take into account the technology limitations of TFTs are employed, the system requirements for the chip turn out to be less stringent. The three main challenges for designing TFT-based NFC tags are the data rates of 106 kbit s−1, a 128-bit memory read-out, and a limited incident power at the tag, in the range of ~10 mW, from the smartphone [72]. Companies are putting great effort into finding a viable manufacturing process. PolyIC is the first company that demonstrated RFID tags produced by roll-to-roll printing. EVONIK develops unique oxide semiconductor materials, whereas IMEC and TNO recently made the OLAE chip with the largest memory (128 bits) and a rectifier working at ultrahigh frequencies. Other companies such as Dai Nippon Printing and OrganicID are working on printed electronics for realizing RFID. OrganicID’s proprietary technology, acquired by Weyerhauser, is supported by several U.S. and foreign patents and patent applications. This technology is used to generate a low-cost 13.56 MHz RFID tag for use in tracking items through manufacturing and shipping. However, the development of the technology is still in progress [74]. Despite significant efforts, printed circuits are still not on par with silicon ICs in terms of stability and performance. Hence, silicon ICs continue to dominate flexible hybrid electronics (FHE) implementations.

5. Applications Radio frequency identification has gained wide public interest, since RFID can be a valid alternative to traditional barcode technology and provides additional features with respect to other alternatives. Printed bar codes are typically read by a laser-based optical scanner that requires a direct line-of-sight to detect and extract information. On the contrary, RFID can be read even when the tag is concealed for either aesthetic or security reasons. Moreover, the low cost could boost the use of RFID tags as pervasive environmental sensors on an unprecedented scale [75]. RFID sensing applications are comprised of monitoring physical parameters, automatic product tamper detection, harmful agent detection, and noninvasive monitoring. Some applications require reading passive tags from a distance of a few centimeters, while others need to read active tags at several hundred meters. Every application poses different constraints and requirements, so a feasibility field study must be performed in the operative environment to choose the best frequency and tags. An overview of the most relevant fields of applications for RFID sensors is summarized in Figure9. Sensors 2021, 21, 3138 14 of 34

RFID Sensors Structural Localization Food Wereable & Healthcare Agriculture Automotive Health Space & activity Quality Implantable Monitoring monitoring

▪ Body ▪ Meat, fish, ▪ Soil moisture ▪ Automatic ▪ Metal and ▪ Temperature ▪ human ▪ human temperature vegetable monitoring production concrete monitoring movements movements monitoring freshness monitoring crack monitoring monitoring ▪ Precision ▪ CO2 ▪ Heart & ▪ Heart & ▪ Blood irrigation ▪ Security of monitoring breath breath glucose ▪ Expiration infants ▪ Structural frequency frequency monitoring date ▪ Agro-food damage ▪ Battery level monitoring monitoring monitoring supply chain ▪ Tire pressure detection monitoring ▪ Activity and monitoring sensors ▪ monitoring ▪ monitoring gesture ▪ Monitoring body areas body areas sensing ▪ Vineyard ▪ Vehicles of structural and vascular and vascular monitoring road movements prosthesis prosthesis ▪ Sleep distance disorders ▪ Cold chain ▪ Corrosion monitoring monitoring

Examples of practical applications

Figure 9. RFID sensors: application ﬁelds and examples of practical applications.

5.1. Healthcare The healthcare industry plays a pivotal role in many economies. For example, in the United States (U.S.), the budget is in the order of several trillions of dollars [76]. Ana- lysts forecast that the number of older people in the U.S. will increase by 135% between 2000 and 2050, and that the “population aged 85 and over—probably the group needing health and long-term care services more than any other—should increase by 350%” [77]. This implies a significant rise in the budget that states should allocate for healthcare. A possible mitigation effect of this trend can be provided by the massive employment of RFID technology, as it allows for preventing errors, saving costs, increasing security, and providing improved quality of life for patients. Cheap sensors (wearable, implanted, and environmental), integrated into the paradigm of the internet of things (IoT), have the potential to make real a personalized smart-health system, where the natural habitat of the person, their body, and the internet are collabo- rating to manage and increase overall medical knowledge. For example, by displacing wireless sensors inside the home, on clothes, and on personal items, it becomes possible to monitor the macroscopic behavior of the person, as well as to compile statistics, to identify precursors of dangerous behavioral anomalies, and finally to activate alarms or prompt for remote actions by appropriate assistance procedures. RFID systems may represent a strategic enabling component thanks to the energy autonomy of battery-less tags and their low cost, which make them compatible with extensive distribution and even disposable applications [78]. Assuming a large diffusion of cheap passive UHF RFID tags inside the environment, it is possible to infer information about human activity. For example, human body movements in close proximity to the tags may introduce scattering and shadowing effects, thus altering the communication link between a fixed reader and the tags [79]. Such changes in the signals received by a combination of wearable tags and ambient tags can be then used to monitor human activity. For example, the state of children and disabled and elderly people in domestic and hospital environments during the night can be monitored by installing a UHF RFID reader that continuously illuminates the bed and detects the presence of a patient in the bed, their body movements, accidental falls, or suspicious long periods of inactivity (which might be due to fainting, unconsciousness, or even death), as well as interactions with objects nearby (e.g., medicines or glasses). By also employing Sensors 2021, 21, 3138 15 of 34

temperature or humidity RFID sensors, even fever evolution and urine loss could be taken under control. Printed sensors utilize printed materials to transduce physical quantities such as temperature, light intensity, sound, force, or chemical reaction to electrical signals. In a printer thermistor, for example, the temperature variation leads to a change in the resistance of the printed active material. For these reasons, in the last decades a large number of research studies have been focused on the design of printed sensors for wearable health monitoring, which may allow for a continuous measurement of all vital signs such as blood pressure, pulse oxygenation, body temperature, and heart and respiration rates. The system is also suitable for providing reports and aggregated statistics, useful for the formulation of diagnosis and for the follow-up of therapies [78]. An example of monitoring system is shown in Figure 10a.

Figure 10. (a) Ambient intelligence system aimed at taking care of night sleep involving RFID tags placed over the body and in the surrounding environment. (b) Set-up for the classiﬁcation of arm and leg gestures by passive RFID and examples of RFID backscattered patterns. Reproduced from [78], © 2014 IEEE.

Moreover, by monitoring the ﬂuctuations of backscattered power from sensor-less passive tags placed over the body it is possible to collect information about the subject, such as standing or moving conditions, and, eventually, to estimate the frequency of periodic movements [80,81]. A measurement setup for gesture recognition based on multiple RFIDs is shown in Figure 10b.

(a) (b) Figure 10. (a) Ambient intelligence system aimed at taking care of night sleep involving RFID tags placed over the body and in the surrounding environment. (b) Set-up for the classification of arm and leg gestures by passive RFID and examples of RFID backscattered patterns. Reproduced from [78], © 2014 IEEE.

5.2. Food Intelligent labeling of food products to indicate and report their freshness and other conditions is one important possible application for RFID sensors. Indeed, the market de- mands new sensors for food quality and safety with battery-free operation and minimal sensor cost [48,82]. In these sensors, the electric field generated in the RFID sensor antenna is affected by the ambient environment, providing the opportunity for sensing. This environment may be in the form of a food sample within the electric field of the sensing region, or a sensing film deposited onto the sensor antenna. Examples of applications include the monitoring of milk freshness, fish freshness, and bacterial growth in a solution. Unlike other food freshness monitoring approaches that require a thin film battery for the operation of an RFID sensor and the fabrication of custom-made sensors, the passive RFID sensing approach combines the advantages of both battery-free and cost-effective sensor design, and offers response selectivity. For food spoilage or quality monitoring applications, most HF passive RFID sensors rely on measuring volatiles or analytes in food pack- Sensors 2021, 21, 3138 aging, whereas LC resonator sensors are based on measuring changes in dielectric 16per- of 34 mittivity [83,84]. As food ripens and spoils, its chemical composition changes, resulting in changes in its dielectric properties. This changes the coupled capacitance, and, in turn, the resonant frequencyfrequency of the the sensor. sensor. These These sensors sensors are are either either attached attached to food to food packaging packaging or placed or placed on onthe the surface surface of the of food the food itself. itself. The approach The approach has been has demonstrated been demonstrated using conformal using conformal adhe- adhesivesive LC resonators LC resonators attached attached to the tosurface the surface of a banana of a bananaskin or cheese skin or [85], cheese as a [3D85], printed as a 3D printedLC resonator LC resonator integrated integrated in a milk inpackage a milk cap package [30], and cap as [30 a], planar and as LC a planarresonator LC attached resonator attachedto the surface to the of surface a milk of package a milk package (Figure (Figure11). The 11 depth). The of depth interaction of interaction between between the food the foodand the and sensor the sensor (the penetration (the penetration depth) depth)depends depends on the operating on the operating frequency frequency of the sensor, of the sensor,electric electricconductivity, conductivity, and theand dielectric the dielectric polarization polarization loss of the loss food of the[48]. food [48].

(a) (b) (c)

Figure 11. (a,b) A wireless passive sensor demonstration of a “smart cap”, containing the 3D-printed LC-resonant circuit. The degradation of the liquid food inside the liquid package can cause changes in the dielectric constant and a shift in the resonance frequency of the LC circuity. (c) A wireless inductive reader is used to monitor the signals in real time. Reproduced from [86].

5.3. Agriculture The development of RFID systems for smart agriculture applications has attracted considerable research efforts in recent years [87]. The precision agriculture system, through the exploitation of RFID technology, allows farmers to maximize the yield by increasing efficiencies, productivity, and profitability, while minimizing unintended impacts on wildlife and the environment [88]. In fact, the real time information acquired by RFID sensors provides helpful data for farmers to adjust crop strategies at any time by taking into account the environmental condition alteration. Among smart agriculture paradigms, leaf sensing represents a new technology, which is used for the detection of plants’ health, such as water status [89,90]. An example of a low-cost and low-power system for leaf sensing using a plant backscatter sensor tag is presented in [90]. A schematic representation of the moisture monitoring RFID system and its practical implementation is reported in Figure 12. More in detail, a sensor measures the temperature differential between the leaf and the air, which is directly related to the plant water stress. A Morse code modulation was used for wireless communication with a low-cost receiver by backscattering RF signals from an 868 MHz carrier emitter. An example of an RFID soil moisture sensing system is reported in [91], which exploits two RFID tags on a pot with a low-pass filtered differential minimum response threshold (DMRT) to estimate the soil moisture level. The extensive measurement campaigns revealed high soil moisture detection accuracy, able to considerably improve the productivity. A passive UHF RFID tag for soil moisture and an environment temperature sensor for low cost agriculture applications was developed in [92]. The soil moisture sensor has been implemented by exploiting the capacitance variation of a passive inter digital capacitive, due to soil permittivity modification, while the environmental temperature was obtained by using an RFID chip with a temperature sensor. Sensors 2021, 21, x FOR PEER REVIEW 18 of 36

Figure 11. (a-b) A wireless passive sensor demonstration of a “smart cap”, containing the 3D-printed LC-resonant circuit. The degradation of the liquid food inside the liquid package can cause changes in the dielectric constant and a shift in the resonance frequency of the LC circuity. (c) A wireless inductive reader is used to monitor the signals in real time. Repro- duced from [86].

5.3. Agriculture The development of RFID systems for smart agriculture applications has attracted considerable research efforts in recent years [87]. The precision agriculture system, through the exploitation of RFID technology, allows farmers to maximize the yield by increasing efficiencies, productivity, and profitability, while minimizing unintended impacts on wildlife and the environment [88]. In fact, the real time information acquired by RFID sensors provides helpful data for farmers to adjust crop strategies at any time by taking into account the environmental condition alteration. Among smart agriculture paradigms, leaf sensing represents a new technology, which is used for the detection of plants’ health, such as water status [89,90]. An example of a low-cost and low-power system for leaf sensing using a plant backscatter sensor tag is presented in [90]. A schematic representation of the moisture monitoring RFID system and its practical implementation is reported in Figure 12. More in detail, a sensor measures the temperature differential between the leaf and the air, which is directly related to the plant water stress. A Morse code modulation was used for wireless communication with a low-cost receiver by backscattering RF signals from an 868 MHz carrier emitter. An example of an RFID soil moisture sensing system is reported in [91], which exploits two RFID tags on a pot with a low-pass filtered differential minimum response threshold (DMRT) to estimate the soil moisture level. The extensive measurement campaigns revealed high soil moisture detection accuracy, able to considerably improve the productivity. A passive UHF RFID tag for soil moisture and an environment temperature sensor for low cost agriculture applications was developed in [92]. The soil moisture sensor has been implemented by exploiting Sensors 2021, 21, 3138 the capacitance variation of a passive inter digital capacitive, due to soil permittivity mod-17 of 34 ification, while the environmental temperature was obtained by using an RFID chip with a temperature sensor.

(a) (b)

FigureFigure 12. SchemeScheme of ofthe theRFID RFID-based-based autonomous autonomous leaf-compatible leaf-compatible temperature temperature sensing sensingsystem proposed system proposed in [89], © 2019 in [89 ], IEEE, and experimental setup. (a) working principle of the RFID system. (b) A picture of the realized system. © 2019 IEEE, and experimental setup. (a) working principle of the RFID system. (b) A picture of the realized system.

5.4.5.4. Automotive Automotive RFIDRFID sensors sensors for for the the automotive automotive industry industry have have exhibited exhibited a significant a significant growth growth in the in the lastlast few few years, years, stimulated stimulated by by the the need need for for increasing increasing the the safety safety and and reliability reliability of vehicles of vehicles,, asas well well as as to to automate automate and and improve improve its its manufacturing manufacturing and and logistics logistics processes. processes. A recent A recent exampleexample is is g giveniven by by the the application application of of RFID RFID technology technology for for univocally univocally identifying identifying and and localizinglocalizing tires tires moving moving on on conveyor conveyor lines lines [93]. [93 ].The The UHF UHF RFID RFID system system is comprised is comprised of ofan an RFIDRFID reading reading gate gate,, located located above above a aconveyor conveyor belt belt,, that that interacts interacts with with passive passive RFID RFID tags tags hostedhosted on on the the moving moving tires. tires. RFID RFID tag tag sensors sensors can can he he zbe zbe embedded embedded in intires tires to toprovide provide Sensors 2021, 21, x FOR PEER REVIEW 19 of 36 informationinformation about about the the tire tires’s’ general condition,condition, suchsuch asas pressurepressure andand temperature. temperature. However, How- ever,a big a challenge big challenge for the for adoption the adoption of RFID of RFID sensors sensors in this in case this is case represented is represented by the by ability the to provide a robust and reliable implementation within the complex and harsh environment abiof ality factory to provid manufacturinge a robust and tires. reliable For implementation example, the detuning within the effects complex of commercial and harsh en- RFID vironmenttags in tires of is a presentedfactory manufacturing in [94]. Moreover, tires. For an exampleexample, of the flexible detuning and effects stretchable of commer- RFID tag cialantennae RFID tags for automotivein tires is presented tire applications in [94]. Moreover, are presented an example in [95 of]. flexible The tag and is manufactured stretchable RFIDwith conductivetag antenna textilese for automotive and embedded tire application into polymers are to presented improve itsin [95]. flexibility The tag and is bondingman- ufacturedwith the tire with rubber. conductive Other textiles examples and where embedded RFID sensorsinto polymer can be to useful improve for theits flexibility automotive andmarket bonding are represented with the tire by rubber. a license Other plate examples RF identification where RFID system sensors [96 ca],n intelligent be useful for park- theing automotive [97], as well market as child are detection represented in vehicles by a license [98– 102plate] Some RF identification examples of system RFID sensors[96], intelligentapplied in parking automotive [97], applicationsas well as child are showndetection in Figurein vehicles 13. [98-102] Some examples of RFID sensors applied in automotive applications are shown in Figure 13.

(a) (b) (c)

FigureFigure 13.13. Possible employment of of as as RFID RFID system system for for preventing preventing child child abandonment abandonment inside inside cars cars (a), (a RFID), RFID on ona license a license plate (b), RFID inside tires (c). plate (b), RFID inside tires (c).

5.5.5.5. Structural Structural Health Health Monitoring Monitoring ThereThere are are several several applications applications which which require require security security measures measures provided provided by bystruc- struc- turaltural health health monitoring monitoring (SHM) (SHM),, such such as as the the monitoring monitoring of ofbridges, bridges, railways railways and and pipelines. pipelines. InIn these these applications, applications, deterioration deterioration of of and and damage damage to to these these structures structures might might occur occur during during theirtheir operational operational lifespan lifespan [103]. [103]. For For this this reason, reason, periodic periodic visual visual inspections inspections are are performed performed,, butbut they they are are proven proven to tobe beunreliable unreliable,, particularly particularly when when the structures the structures are hard are to hard access. to ac- Consequently, several non-destructive testing and evaluation (NDT & E) approaches, such as ultrasonic [104], pulsed eddy current (PEC) [105], and eddy current pulsed ther- mography (ECPT), were developed for monitoring defects in structures with good resolution, sensitivity, and reliability. Cabled large-scale sensor networks are very suitable for real time acquisition applications, because of their good performance. On the other hand, the use of these networks for manual data collection is rarely justified by the costs, installation difficulties, and maintenance [106]. Wireless sensor networks (WSNs) are cheaper and simpler to install by removing the electric wiring from traditional sensors. Spatial granularity is a crucial problem for possible upcoming applications. Battery-powered sensors are used for existing wireless sensing applications. Unfortunately, these sensors have reduced battery life and are more expensive than RFID sensors. Consequently, this restricts the granularity of their deployment and motivates low-cost, wireless, and passive sensor production for large-scale networks and applications for big data. Thanks to its low-cost, wireless, and “sensing-friendly” capabilities, radio frequency identification (RFID) technology can play a strategic role. The following paragraphs explain several examples of controlled parameters in structural health monitoring applications. Strain detection: to detect deformation or structural change that occurs in our surrounding infrastructure, strain sensors (gauges) are needed. In order to produce an effective strain sensor, researchers are looking for a material that, in response to a small applied strain, can display a significant structural change [107]. Crack detection: the development of fatigue cracks accounts for more than 50% of mechanical failures. The conventional crack sensing methods make use of lead wiring for data extraction, which is inefficient and costly for the positioning and maintenance of large lengths [108]. In [109], through the mutual-coupling between two patch antennas, it

Sensors 2021, 21, 3138 18 of 34

cess. Consequently, several non-destructive testing and evaluation (NDT & E) approaches, such as ultrasonic [104], pulsed eddy current (PEC) [105], and eddy current pulsed thermog- raphy (ECPT), were developed for monitoring defects in structures with good resolution, sensitivity, and reliability. Cabled large-scale sensor networks are very suitable for real time acquisition applications, because of their good performance. On the other hand, the use of these networks for manual data collection is rarely justified by the costs, installation difficulties, and maintenance [106]. Wireless sensor networks (WSNs) are cheaper and simpler to install by removing the electric wiring from traditional sensors. Spatial granularity is a crucial problem for possible upcoming applications. Battery-powered sensors are used for existing wireless sensing applications. Unfortunately, these sensors have reduced battery life and are more expensive than RFID sensors. Consequently, this restricts the granularity of their deployment and motivates low-cost, wireless, and passive sensor production for large-scale networks and applications for big data. Thanks to its low-cost, wireless, and “sensing-friendly” capabilities, radio frequency identification (RFID) technology can play a strategic role. The following paragraphs explain several examples of controlled Sensors 2021, 21, x FOR PEER REVIEWparameters in structural health monitoring applications. 20 of 36 Strain detection: to detect deformation or structural change that occurs in our surrounding infrastructure, strain sensors (gauges) are needed. In order to produce an effective strain sensor, researchers are looking for a material that, in response to a small applied has been demonstrated that the backscattered phase can work as a sensing variable at a strain, can display a significant structural change [107]. sub-mm resolution. A scheme of the crack detection system implemented with a couple Crack detection: the development of fatigue cracks accounts for more than 50% of of RFID tags is shown in Figure 14. In [110], a frequency signature-based chipless RFID is mechanical failures. The conventional crack sensing methods make use of lead wiring for presented for metal crack detection and characterization operating in the ultra-wideband data extraction, which is inefficient and costly for the positioning and maintenance of large frequency. lengths [108]. In [109], through the mutual-coupling between two patch antennas, it has Corrosion detection: in the form of stress corrosion cracking (SCC) in susceptible metal been demonstrated that the backscattered phase can work as a sensing variable at a sub-mm components, the interaction of a corrosive environment and tensile stress (e.g., directly resolution. A scheme of the crack detection system implemented with a couple of RFID applied stresses or in the form of residual stresses) will cause failure. A thin film of oxides tags is shown in Figure 14. In [110], a frequency signature-based chipless RFID is presented occurs in the early stages of corrosion and induces changes in the conductivity, permittiv- for metal crack detection and characterization operating in the ultra-wideband frequency. ity and permeability of the metallic surface [111].

(a) (b) Figure 14. Principle of (a) an RFID sensor tag couplet for the detection of cracks on concrete struc- Figure 14. Principle of (a) an RFID sensor tag couplet for the detection of cracks on concrete tures [109], © 2015 IEEE; (b) an RFID chipless RFID sensor system for crack detection and charac- structures [109], © 2015 IEEE; (b) an RFID chipless RFID sensor system for crack detection and terization on a metallic structure. characterization on a metallic structure. 5.6. Space Corrosion detection: in the form of stress corrosion cracking (SCC) in susceptible metal components,Sensor RFIDs the interactionhave also found of a corrosive application environments on space andplatforms. tensile Wire stress harness (e.g., directlyes rep- ap- resentplied a stresses considerable or in thepart form of the of overall residual mass stresses); therefore will, eliminating cause failure. the Awiring thin ﬁlmof the of sen- oxides sorsoccurs is of in the the utmost early stagesimportance of corrosion for reducing and induces the total changes mass of in the the vehicle conductivity,, and thu permittiv-s the fabricationity and permeability and launch ofcosts. the Moreover, metallic surface a wireless [111 ].system requires low-maintenance costs and offers an increase in reliability that is crucial, especially for long-term missions that can5.6. last Space years. An example of this is represented by the adoption of a RAIN RFID tag chip by theSensor U.S. National RFIDs have Space also Agency found (NASA) applications [112] on. The space tag platforms.mounts a Monza Wire harnesses X-8K Dura repre- chipsent that a considerable can monitor part temperature, of the overall CO2 mass;, and therefore, battery eliminating levels. A different the wiring approach of the sensors for achievingis of the utmostthe sensing importance function for is reducingprovided theby SAW total massRF tags of [113]. the vehicle, In this and case, thus two the phase fabrica-- matchedtion and SAW launch RF costs.tags are Moreover, coupled with a wireless a Van Atta system antenna requires array. low-maintenance A combination of costs code and diversity and time diversity has allowed for producing a set of 16 sensors that operate offers an increase in reliability that is crucial, especially for long-term missions that can last simultaneously in the field of view of the wireless sensor system. On the other hand, the Van Atta array provides a completely passive capability for tracking the probing direction and, at the same time, guarantees the beam-steering capability for properly retransmitting the encoded information. This configuration exhibits a remarkable read range, it can be employed for monitoring temperature and pressure, and it is robust enough to be deployed in a harsh environment. It is worth noting that, to reduce the weight of a satellite, the RFID reader itself is required to be compact in size and lightweight, as with the one proposed in [114] working within the 5725–5850 MHz frequency range.

5.7. Localization and Activity Monitoring Even if, in these applications, RFIDs were not used as transducers for sensing purposes, RFID tags are still employed to extract useful information which exceeds the simple tracking of objects. RFID sensor networks are a promising approach for indoor activity monitoring [115–117]. A variety of experiments have been performed for tracking and tracing, indoors and outdoors. For example, in transportation systems, including buses, subways, and trains, the City of London implemented a system for outdoor data collection. The main goal was to use payments and access cards equipped with this connectivity technology to track users of the transport network. The level of usage, number of travelers,

Sensors 2021, 21, 3138 19 of 34

years. An example of this is represented by the adoption of a RAIN RFID tag chip by the U.S. National Space Agency (NASA) [112]. The tag mounts a Monza X-8K Dura chip that can monitor temperature, CO2, and battery levels. A different approach for achieving the sensing function is provided by SAW RF tags [113]. In this case, two phase-matched SAW RF tags are coupled with a Van Atta antenna array. A combination of code diversity and time diversity has allowed for producing a set of 16 sensors that operate simultaneously in the ﬁeld of view of the wireless sensor system. On the other hand, the Van Atta array provides a completely passive capability for tracking the probing direction and, at the same time, guarantees the beam-steering capability for properly retransmitting the encoded information. This conﬁguration exhibits a remarkable read range, it can be employed for monitoring temperature and pressure, and it is robust enough to be deployed in a harsh environment. It is worth noting that, to reduce the weight of a satellite, the RFID reader itself is required to be compact in size and lightweight, as with the one proposed in [114] working within the 5725–5850 MHz frequency range.

5.7. Localization and Activity Monitoring Even if, in these applications, RFIDs were not used as transducers for sensing purposes, RFID tags are still employed to extract useful information which exceeds the simple tracking of objects. RFID sensor networks are a promising approach for indoor activity monitoring [115–117]. A variety of experiments have been performed for tracking and tracing, indoors and outdoors. For example, in transportation systems, including buses, subways, and trains, the City of London implemented a system for outdoor data collection. The main goal was to use payments and access cards equipped with this connectivity technology to track users of the transport network. The level of usage, number of travelers, and public transport habits in the city, and knowledge of the stations, breakpoints, origin, and destination of network travel streams were calculated with the information collected. Indoor technologies have also been developed to consider citizens’ movement flows, such as museum visits. The collected data allowed for the identification of the visiting habits for the various areas and the identification of atypical behaviors [118]. Some of the mentioned drawbacks of a global positioning system (GPS) are addressed by RFID technology, as it does not require too much user cooperation, and reduces energy costs. However, its use for tracking does not provide precise positions, and the temporal position marks are restricted, like other wireless networks, by the locations of the signal reader antennas and the analysis of the readings’ received signal strength indicator (RSSI). The resolution of the trajectories followed by the users is less accurate than the resolution of a GPS. In addition, the outdoor aim of this technology is not comparable to GPS coverage. On the other hand, phase-based RFID localization methods allow for a better localization accuracy, since they are more robust to multipath propagation than the RSSI-based approaches. As an example, phase-based techniques have been used to localize moving tagged items on conveyor belts [119], robots [120,121], and drones [122]. Examples of robots and drones equipped with RFID readers for localization purposes are reported in Figure 15. Sensors 2021, 21, x FOR PEER REVIEW 21 of 36

and public transport habits in the city, and knowledge of the stations, breakpoints, origin, and destination of network travel streams were calculated with the information collected. Indoor technologies have also been developed to consider citizens′ movement flows, such as museum visits. The collected data allowed for the identification of the visiting habits for the various areas and the identification of atypical behaviors [118]. Some of the mentioned drawbacks of a global positioning system (GPS) are addressed by RFID technology, as it does not require too much user cooperation, and reduces energy costs. However, its use for tracking does not provide precise positions, and the temporal position marks are restricted, like other wireless networks, by the locations of the signal reader antennas and the analysis of the readings′ received signal strength indicator (RSSI). The resolution of the trajectories followed by the users is less accurate than the resolution of a GPS. In addition, the outdoor aim of this technology is not comparable to GPS coverage. On the other hand, phase-based RFID localization methods allow for a better localization accuracy, since they are more robust to multipath propagation than the RSSI-based approaches. As an example, phase-based techniques have been used to localize moving Sensors 2021, 21, 3138 tagged items on conveyor belts [119], robots [120,121], and drones [122]. Examples of20 ro- of 34 bots and drones equipped with RFID readers for localization purposes are reported in Figure 15.

(a) (b)

FigureFigure 15. 15.( a(a)) Robotic Robotic wheeled wheeled walker walker equipped equipped with with the the Impinj Impinj Speedway Speedway Revolution Revolution R420 R420 UHF-RFID UHF-RFID reader, reader, the the reader reader an- antenna, and optical markers used for the measurement campaign in [120], © 2021 IEEE; (b) Colibrì I3-A drone by IDS, tenna, and optical markers used for the measurement campaign in [120], © 2021 IEEE; (b) Colibrì I3-A drone by IDS, equipped with a GNSS module and the UHF-RFID reader and antenna used in [122], © 2019 IEEE. equipped with a GNSS module and the UHF-RFID reader and antenna used in [122], © 2019 IEEE.

5.8.5.8. Wearable and Implantable RFID RFID Sensors Sensors One challenge is the design of wearable antennas that hav havee high immunity to inter- interac- tionsactions with with the the human human body, body which, which may may notably notably change change the the radiation radiation diagram diagram and and degrade de- antennagrade antenna efﬁciency. efficiency. In active In active and semi-active and semi-active architectures, architectures, as in as the in case the ofcase body-centric of body- communicationcentric communication systems, systems, the overall the overall radiation radiation performance performance is enhanced is enhanced by additionalby addi- battery-assistedtional battery-assisted electronics. electronics. In the In case the case of passive of passive tags tags instead, instead, where where the the energy energy to to pro- duceproduce the responsethe response comes comes from from a remote a remote unit, unit, the antennathe antenna design design is much is much more more challenging. chal- Embroiderylenging. Embroidery techniques techniques have been have experimented been experimented with to with fully tointegrate fully integrate RFID RFID tags insidetags clothes.inside clothes. E-textiles, E-textiles, conductive conductive threads, threads, and embroidery and embroidery techniques techniques proved proved to be suitableto be suitable for the design of garment-integrated tags [123,124]. In order to enable on-body for the design of garment-integrated tags [123,124]. In order to enable on-body wireless wireless connectivity, this new technology will include multi-functional everyday cloth- connectivity, this new technology will include multi-functional everyday clothing inte- ing integrated with wearable antennas, sensors, and power harvesting devices. The con- grated with wearable antennas, sensors, and power harvesting devices. The conductivity ductivity of the embroidered tag antennas mainly depends on the orientation of the sewn of the embroidered tag antennas mainly depends on the orientation of the sewn lines in lines in relation to the current flow direction in the antenna system. Moreover, the con- relation to the current ﬂow direction in the antenna system. Moreover, the conductivity of ductivity of the embroidered structure is also determined by the electrical properties of the embroidered structure is also determined by the electrical properties of the conductive the conductive thread. The thread used has a silver weight of 55 g/10,000 m, which assures thread. The thread used has a silver weight of 55 g/10,000 m, which assures good conductivity. Moreover, the conductivity and performance of the embroidered tag antennas is affected by the spacing between the sewn lines. An example of embroidered tags is shown in Figure 16a.

Figure 16. (a) Embroidered dipoles with pattern I and pattern II. The application of UHF RFID technology for monitoring. Reproduced from [124], © 2013 IEEE. (b) The in-stent restenosis inside a carotid stent and a prototype of a STENTag. Reproduced from [78], © 2014 IEEE. Sensors 2021, 21, 3138 21 of 34

The maximum read range which is currently achievable with an RFID transmitter compliant with the regional power emission constraints is almost 5–6 m. Current link performances are already enough for tracking a person equipped with two tags over front/rear torso or over the arms within a regular size room [78]. In entertainment, healthcare, and medical applications, a wearable tag supplied with passive accelerometers can be applied on the arms for the tracking of human motion. In some common sleep disorders, such as restless leg syndrome and periodic limb movements, wearable tags with inertial switches have been shown to detect limb movements. Tags applied to the chest may also be useful to detect breath. Generally, wireless motion tracking systems may help to produce data to support diagnosis and track a patient’s behavior discreetly within a structure, and generate notifications about unusual activities, such as when the patient falls down or remains motionless for long periods [125]. RFID technology has been also demonstrated to be potentially useful for taking care of the human health-state internally by labeling body prostheses, sutures, stents, or orthopedic fixing. Each item could be monitored in real time or on demand by the IoT infrastructure for the ambitious goal of monitoring biophysical processes in evolution, such as tissue regrowth and prosthesis displacement. In this case, tags are inserted in the prosthesis, converting them into multi-functional devices capable of generating information in addition to providing the original medical functionality. In the design of implantable tags, the key challenge is to create a convenient communication link by using a reader power that complies with power emission regulation. Nowadays, the creation of an RFID connection with subcutaneous implants up to 0.1–0.5 m from the reader is now feasible, with promising applications for tracking some specific areas of the body, and vascular protheses. In order to automatically scan the health-state of a prosthesis without the direct involvement of the patient, a direct link to a reader-equipped door located at a distance of 0.5 m may be sufficient. On the other hand, deeper implants such as implants within the stomach remain a real challenge, even over the long term, for passive RFID systems under current power regulations [78]. A vascular stent implanted into an artery to restore natural blood flow after an- gioplasty is an example of near-subcutaneous protheses sufficient for RFID integration, as shown in Figure 16b. As presented in [126], the RFID tag features have been strongly incorporated into the stent geometry itself to act as a self-sensor by exploiting the depen- dency of the input impedance of the tag and back-radiation on the dielectric properties of the tissues in the immediate vicinity of the so-obtained STENTag. This makes it possible to detect the process of in-stent restenosis (ISR), e.g., diffuse proliferation of neointima early after implantation, or new atherosclerotic plaque, even at a distance of time, which might cause new occlusion of the vessel.

6. Commercial RFID Sensors RFID technology has become extremely popular in recent years for both localization and sensing applications. This technology is widely used within the IoT paradigm, where the demand for low-power and low-cost wireless devices is increasing. From this perspective, RFID sensors respond to different needs of the IoT, as they are power efficient, small and easy to use. The use of innovative materials and manufacturing techniques, combined with increasingly advanced data collection techniques make RFID sensors even more appealing for IoT applications where sensing capabilities are required. Moreover, each RFID sensor has its own identification number, which makes it unique, making the collection of data from the sensor unambiguous. Finally, compared to other sensing and identification techniques, RFIDs are extremely advantageous, as they can be used in harsh environments, do not require a line-of-sight connection, and allow real-time collection of data from multiple sensors simultaneously. For these reasons, RFID sensors have raised great interest, not only in the world of scientific research, but also from industries. In fact, there are many companies that have heavily invested in RFID technology, such as TI, NXP, and ON Semicondutors. Sensors 2021, 21, 3138 22 of 34

The main manufacturers of ready-to-use RFID sensors are reported in Table6, as well as some examples of RFID sensors that can be purchased on the market. Some commercial sensor tags conﬁgurations are reported in Table7. Very often, these solutions can integrate with other sensors through an I2C interface, as in the case of TI RF430FRL152H. An interesting solution is proposed by NXP with the NTAG® SmartSensor. These are single-chip solutions that combine NFC connectivity with autonomous sensing, data processing, and logging via the I2C interface. The data collected by NTAG SmartSensor are uploaded into the cloud via NFC using a smartphone and a dedicated app. The NTAG SmartSensors are temperature-calibrated, which is an important feature for tracking temperature-sensitive products such as vaccines or wines. By using the I2C interface, the NTAG SmartSensor can also monitor conditions such as humidity, tilting, shocks, and vibrations. Another application of NXP NTAG is smart agriculture (Figure 17). The NTAG® SmartSensor Figure 17a is placed in the plant, as shown in Figure 17b. Another example of a commercial RFID sensor is HYGRO-FENIX-RM of Farsens (Donostia-San Sebastian, Spain), which is an electronic product code (EPC) class 1 generation 2 (C1G2) RFID tag based on Farsens’ battery-less sensor technology. The tag contains an ambient temperature sensor and a relative humidity sensor, which are compatible with commercial UHF RFID readers (EPC C1G2). The battery-less resistance meter can communicate over 5 m with a 2W ERP setup. Farsens also produces other passive sensor tags, such as the EVAL01-Kineo-RM, which is a UHF RFID battery-free orientation sensor tag. It features an LIS3DH 3-axis accelerometer from ST Microelectronics with a range between ±4 g and an accuracy of ±40 mg. Other passive sensors are proposed by Farsens to sense parameters such as RF ﬁeld (EVAL02-Photon-R) and ambient light intensity (EVAL01-Spectre-RM).

Table 6. Principal RFID sensors manufacturers. The websites have been accessed on 1 March 2021.

Company Name Website Sensor AMS https://ams.com/wireless-sensor-tags-interfaces Temperature/integration with external sensors Axzon (RFmicron) https://axzon.com/sensors/ Temperature, moisture Temperature, humidity, force/strain, Pressure, http://www.farsens.com/en/products/battery- Farsens LED/RF field detection, Switch/relay free-rfid-sensors/ monitoring, orientation, light, magnetic field Infratab https://www.infratab.com/rf-sensor-solutions food freshness https://www.melexis.com/en/products/ Melexis Temperature/integration with external sensors communicate/nfc-rfid-products https://www.nxp.com/products/rfid-nfc/nfc-hf/ NXP Temperature ntag/ntag-smartsensor:NTAG-SMART-SENSOR https://www.onsemi.com/products/sensors/ ON Semiconductor Temperature, moisture battery-free-wireless-sensor-tags https: Phase IV Temperature, moisture, pressure, strain //www.phaseivengr.com/products/rfid-sensors/ https://www.powercastco.com/products/rfid- PowerCast Temperature, humidity, light sensor-tags/ Temperature, humidity, water detection, Motion, RadioForce https://www.radioforce.net/en/products/sensors vibrations, pressure Silent Sensors https://www.silentsensors.com/ Tire conditions https://www.smartrac-group.com/sensor-inlays- Smartrac Temperature, moisture and-tags.html https://www.ti.com/wireless-connectivity/other- Texas Instruments Temperature + SPI/I2C external sensor wireless-technologies/overview/nfc-rfid.html PST Sensors https://www.pstsensors.com/wireless.htm Temperature Sensors 2021, 21, 3138 23 of 34

Table 7. Examples of commercial RFID sensors.

Tags Sensing RFID Unit Manufacturer Sensor Model Accuracy Type Size [mm3] Function Band Price ON SPS1M001FOM Moisture UHF N/A passive 165.7 × 20 × 5 $7.3 Semiconductor ON SPS1T001PET Temperature UHF ±0.5◦ @ 30◦ passive 101.60 × 31.75 × 5 $3.99 Semiconductor Axzon RFM2110 Moisture UHF N/A passive 79.5 × 19.2 × 2.1 $5.7 (RFmicron) EVAL01-Hygro- Temperature ±1 ◦C [0 ◦C–60 ◦C] Farsens Fenix-RM- and relative UHF ±3.5% rH [20% passive 137 × 16 × 10 $58 DKWB humidity rH–80% rH] Orientation range ±4 g EVAL02-Kineo- Farsens (3-axis UHF acceleration accuracy passive 137 × 17 × 10 $42 RM-DKSWB accelerometer) ±40 mg EVAL02-Photon- RF ﬁeld Farsens UHF ON/OFF passive 160 × 27.5 × 10 $42 R detection Optical spectrum: 300 nm EVAL01-Spectre- UHF to 1000 nm Farsens Ambient light passive/BAP 137 × 16 ×10 $60 RM Range: 1.2 nW/cm2 to 10 mW/cm2 Compression load range: 0 EVAL01-Zygos- 95 × 95 × 45 Farsens Force/strain UHF to 5 kg, Compression load passive/BAP $48 RM 137 × 16 × 10 accuracy: ±0.05 kg Chip Sensing RFID Unit Manufacturer Sensor Model Accuracy Type size [mm3] Function Band Price Default range: −20 ◦C to Temperature + 55 ◦C AMS AS39513 external HF passive/BAP Chip: 2.5 × 2.5 × ∆ $1.54 ±0.5 ◦C over −20 ◦C to resistive sensor 10 ◦C Melexis MLX90129 Temperature HF ±2.5 ◦C with calibration passive/BAP Chip 6.4 × 6.4 × 1.1 ~$3 NTAG® Temperature/ ±0.5 ◦C from −40 ◦C–0 ◦C NXP HF/UHF BAP Chip: 4 × 4 × 0.85 $1.5/3 SmartSensor moisture ±0.5 ◦C from 40 ◦C–85 ◦C Temperature + Texas RF430FRL152H SPI/I2C HF N/A passive/BAP Chip: 4 × 4 × 0.75 $3.8 Instruments external sensor

(a)(b) Figure 17. NTAG® SmartSensor for smart agriculture: (a) pictorial representation of the NTAG®;(b) application of the sensor in the plant. It is also worth mentioning the use of RFID sensors for the moisture intrusion detection system for vehicle assembly lines. Water leakage has been one of the main problems in the auto industry for years. Indeed, water leaks might cause mold growth and damage to the expensive electronic components of motor vehicles, thus reducing the quality of the fabrication process. Traditional manual inspection methods rely on visual inspection. Consequently, this approach does not detect leaks located in inaccessible areas, or small Sensors 2021, 21, 3138 24 of 34

leaks. ON Semiconductor, Inc. (Phoenix, AZ, U.S.A.) and RFMicron, Inc. (Austin, TX, USA) have solved this problem with a two-fold leak detection solution [127]. The first part utilizes thin and low-profile moisture sensors that are placed in the vehicle without affecting or displacing other components or trim pieces (Figure 18a). The second part is a system to read these sensors and then aggregate that sensor data to determine where leaks are located (Figure 18b). Battery-free wireless moisture sensors and a highly capable processing device mounted directly onto moving assembly lines are included in the RFM5126 water leak detection system. A wireless communication between the antennas mounted on a system portal and the sensors located on the vehicles is established. When the vehicles drive through the portal, the system collects the data from the sensors. In order to decide whether and where any leaks might be present, the device processes the sensors’ data. The identification of the water leaks allows for a reduction in repair costs and saves time, so that rework teams do not spend their time searching for leaks.

(a)(b) Figure 18. RFID sensors for the moisture intrusion detection system: (a) Moisture sensors placed directly on the vehicles’ metal chassis; (b) Interrogation setup.

7. RFID Readers RFID systems can operate at different frequency bands: low-frequency (LF, 125–134 KHz), high-frequency (HF, 13.56 MHz), ultra-high-frequency (UHF, 866–928 MHz), and microwave (MW, 2.45 GHz and 5.8 GHz), as summarized in Table8. The speciﬁc operating frequency is chosen on the basis of the application requirements. For example, LF and HF are typically used for short-range applications (e.g., of a few centimeters), since the wireless power transmission is based on the inductive coupling between two coils. The magnetic ﬁeld received by the tag coil induces a current which activates the chip. The chip impedance is then varied based on the EPC stored inside the chip. Such a load variation induces a modulation of the backscattered signal. The latter is then collected at the reader side and decoded. On the other hand, UHF RFID systems are typically used for long-range applications, since they are based on electromagnetic coupling and wave propagation. The backscattered electromagnetic signal is then received by the reader antenna and converted to an electrical signal. If the received power is higher than the reader sensitivity, then the interrogated tag can be properly detected, and the unique ID (EPC) can be stored. Sensors 2021, 21, 3138 25 of 34

Table 8. RFID operating frequencies and standards.

LF HF UHF Microwave 865–868 MHz (ETSI) Freq. Range 125–134 KHz 13.56 MHz 2.45–5.8 GHz 902–928 MHz (FCC) Read Range 10 cm 1 m 0–10 m 1 m Market share 74% 17% 6% 3% Coupling Magnetic Magnetic Electromagnetic Electromagnetic 18,000-3.1, 15,693, 14,443 Existing standards 11,784/85, 14,223 EPC C0, C1, C1G2, 18,000-6 18,000-4 A, B, and C Small item management, Transportation vehi- Smart card, ticketing, Transportation vehicle ID, supply chain, cle ID, access/security, Applications animal tagging, access/security, large item anti-theft, library, item management, access, laundry management, supply chain transportation supply chain

A RFID reader is powered by a battery or from an external power source, and gen- erates an interrogation signal centered at the speciﬁc operating frequency. The signal is then emitted by the RFID reader antenna after an electrical-to-electromagnetic conversion. The carrier is then received by tag antennas placed at a certain distance from the reader antenna. If the power received by the transponder is higher than the chip sensitivity, then the chip is powered up and sends the unique ID stored in it. Speciﬁcally, the interrogation signal is backscattered by the tag after a load modulation, which causes a little shift in the operating frequency, as schematically represented in Figure 19.

Signal Power

Carrier signal emitted by the reader antenna PC ΔP power attenuation of Backscattered signal – the backscattered signal Modulation product by with respect to the load modulation interrogation signal

PC – ΔP

f0 – Δf f0 f0 + Δf Frequency

Figure 19. Load modulation creates two sidebands around the transmission frequency f0 of the reader. The actual information is carried in the sidebands of the two subcarrier sidebands, which are themselves created by the modulation of the subcarrier.

RFID readers can be classiﬁed into three main typologies: - Fixed or desktop reader. In applications where the available space is not limited, ﬁxed readers can be used, which guarantee the best performance. The typical size is smaller than 30 cm × 30 cm and the cost is approximately in the range of 800–1200 USD. Some commercial readers are listed in Table9, also including the operating frequency and main performance parameters. In general, readers can be Sensors 2021, 21, 3138 26 of 34

SensorsSensorsSensors 2021 20212021,, ,21 2121,, ,xx x FORFOR FOR PEERPEER PEER REVIEWREVIEW REVIEW 282828 ofofof 363636 connected to up to four different external antennas, and a time division approach is used to select the specific radiating element. - Integrated readers. Several commercial solutions are integrated systems, where both industries. However, differently from fixed readers, the antenna cannot be changed, industrindustrindustrthe readeriesiesies.. . However, However,However, and the differently differentlydifferently antenna are from fromfrom embedded fixed fixedfixed readers, readers,readers, in the the thethe same antenna antennaantenna device ca caca (Tablennotnnot be be 10 changed, changed,). This repre- thus the RFID system cannot be fully designed on the basis of the operative scenario. thusthussents the the aRFIDRFID compact system system solution cannot cannot forbe be fully mostfully designed designed of the applications on on the the basis basis in of of the the the retail operative operative or pharmaceutical scenario. scenario. In HF RFID commercial integrated readers, the radiating element is represented by a InIn HF industries.HF RFID RFID commercial commercial However, integrated integrated differently readers, readers, from fixed the the radiating readers,radiating theelement element antenna is is represented represented cannot be by changed,by a a coil, properly tuned at 13.56 MHz. On the other hand, in UHF RFID readers, different coilcoilthus,, , properly properly the RFID tuned tuned system at at 13.56 13.56 cannot MHz. MHz. be On On fully the the designedother other hand, hand, on in in the UHF UHF basis RFID RFID of the rea rea operativeders,ders, different different scenario. typologies of antennas are integrated to guarantee reliable detection when the tag is typologiestypologiesIn HF RFID of of antennas antennas commercial are are integrated integrated to to readers, guarantee guarantee the reliable radiatingreliable detection detection element when when is represented the the tag tag is is by a placedplaced bothboth onon thethe readerreader casecase andand atat aa longerlonger distancedistance [130].[130]. IntegratedIntegrated readerreader’s’s typ-typ- placedcoil, both properly on the tuned reader at case 13.56 and MHz. at a longer On the distance other hand, [130]. in Integrated UHF RFID reader readers,’s’s typ- differ- icalical costcost isis inin thethe rangerange ofof 700700––12001200 USD.USD. icalicalent costcost typologies isis inin thethe rangerange of antennas ofof 700700– are1200 integrated USD. to guarantee reliable detection when the tag -- Portable/handheldPortable/handheld oror wearwearableable readersreaders.. HandheldHandheld readersreaders representrepresent aa lightweightlightweight - Portable/handheldis placed both on or the wear readerable casereaders and.. Handheld atHandheld a longer readersreaders distance representrepresent [128]. Integrated aa lightweightlightweight reader’s andand easy easy--toto--carrycarry solution, solution, which which can can be be used used by by a a human human operator operator for for scanning scanning andtypical easy-to cost-carry is in solution, the range which of 700–1200 can be USD. used by a human operator for scanning taggedtagged items items placed placed in in different different positions positions ( (TableTable 11).11). In In this this framework, framework, portable portable - taggedPortable/handheld items placed in or different wearable positions readers (.TableHandheld 11). In readers this framework, represent a portable lightweight readerreader antennasantennas mustmust bebe integratedintegrated inin aa compactcompact space,space, thusthus theythey areare requiredrequired toto bebe readerand easy-to-carryantennas must solution, be integrated which in can a becompact used by space, a human thus operator they are forrequired scanning to be tagged lightweight,lightweight, lowlow profileprofile,, andand compactcompact inin size.size. Consequently,Consequently, thethe readread rangerange isis limitedlimited lightweight,lightweight,items placed lowlow in profileprofile different,, and positions compact (Table in size. 11 Consequently,). In this framework, the read portable range is reader limited anten- toto upup toto aa fewfew meters.meters. Furthermore,Furthermore, thethe proximityproximity ofof thethe humanhuman handhand representsrepresents anan to upnas to must a few be meters. integrated Furthermore, in a compact the proximity space, thus of they the human are required hand represents to be lightweight, an importantimportant aspectaspect toto taketake intointo accountaccount duriduringng thethe antennaantenna designdesign process,process, becausebecause thethe importantimportantlow profile, aspectaspect and toto taketake compact intointo accountaccount in size. duriduri Consequently,ng the antenna the readdesign range process, is limited because to the up to a dielectricdielectric propertiesproperties ofof humanhuman tissuestissues maymay significantlysignificantly affectaffect thethe antennaantenna’s’s’s perfor-perfor- dielectricfew meters. properties Furthermore, of human the tissues proximity may significantly of the human affect hand the represents antenna’s an perfor- important mance.mance. Handheld Handheld readers readers’’ ’ typical typicaltypical cost costcost is isis in inin the thethe range rangerange of ofof 200 200200––900900 USD. USD. mance.aspect Handheld to take intoreaders account’ typical during cost is the in antennathe range design of 200– process,900 USD. because the dielec- ItItIt is isistric worth worthworth properties mentioning mentioningmentioning of human that thatthat softwaresoftwaresoftware tissues may defined defineddefined significantly radio radioradio (SDR)(SDR)(SDR) affect readers readersreaders the antenna’s have havehave been beenbeen performance. also alsoalso de- de-de- signedsignedsigned andHandheldandand describeddescribeddescribed readers’ ininin thethethe typical scientificscientificscientific cost literatureliteratureliterature is in the range[131,132],[131,132],[131,132], of 200–900 butbutbut theytheythey USD. areareare notnotnot commerciallycommerciallycommercially availableavailable yet. yet. Table 9. Examples of commercial fixed RFID readers. The websites have been accessed on 1 March 2021. TableTable 9. 9. Examples Examples of of commercial commercialcommercial fixed fixedfixed RFID RFIDRFID readers readersreaders.. . The The websites websites have have been been accessed accessed on on 1 11 March March 2 2021.021.021. MODEL Number of MODELMODEL Picture Operating Frequency Band NumberNumber of of An- An- Website Link (Manufacturer) PicturePicture OperatingOperating Frequency Frequency Band Band Antenna Ports WebsiteWebsite Link Link (Manufacturer)(Manufacturer)(Manufacturer) tennatennatenna Ports PortsPorts

https://www.caenr-https://www.caenr- QuattroQuattroQuattro-R4321P‐‐‐R4321PR4321P 865–868865865––868868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) https://www.caenrﬁd.com/en/ 4 44 fid.com/en/pro-fid.com/en/pro-fid.com/en/pro- (CAEN(CAEN(CAEN(CAEN RFID RFIDRFID RFID s.r.l.) s.r.l.)s.r.l.) s.r.l.) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) products/quattro-r4321p/ ducts/quattroducts/quattro---r4321p/r4321p/r4321p/

https://www.im-https://www.im- ImpinjImpinjImpinjImpinj R700 R700R700 R700 reader readerreader reader 865–868865865––868868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) https://www.impinj.com/ 4 44 pinj.com/products/rea-pinj.com/products/rea- (Impinj)(Impinj)(Impinj)(Impinj) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) products/readers/impinj-r700 ders/impinjders/impinj---r700r700r700

https://www.im-https://www.im- https://www.im-https: ImpinjImpinjImpinjImpinj Speedway SpeedwaySpeedway Speedway 865–868865865––868868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) 4 44 //www.impinj.com/products/pinj.com/products/rea-pinj.com/products/rea- (Impinj)(Impinj)(Impinj)(Impinj) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) readers/impinj-speedwayders/impinjders/impinj---speedwayspeedwayspeedway https://www.zebra.cohttps://www.zebra.co 865–868 MHz (ETSI UHF) https://www.zebra.com/it/it/m/it/it/pro- 865–868865865––868868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) m/it/it/pro-m/it/it/pro- FX7500FX7500FX7500 (ZEBRA) (ZEBRA) (ZEBRA) 4 44 products/rfid/rfid-readers/fx7 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) ducts/rfid/rfidducts/rfid/rfid---rea-rea-rea- 500.html ders/fx7500.htmlders/fx7500.html https://www.zebra.cohttps://www.zebra.co https://www.zebra.com/it/it/ 865–868865865––868868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) m/it/it/pro-m/it/it/pro- FX9600FX9600FX9600 (ZEBRA) (ZEBRA) (ZEBRA) 4 44 products/rfid/rfid-readers/fx9 902–928902902––928928 MHz MHzMHz (FCC (FCC(FCC UHF) UHF)UHF) ducts/rfid/rfidducts/rfid/rfid--rea-rea- 902–928 MHz (FCC UHF) ducts/rfid/rfid600.html -readers/fx9600.htmlders/fx9600.html

TableTable 10. 10. Examples Examples of of commercial commercial integrated integrated RFID RFID readers readers.. . The The websites websites have have been been accessed accessed on on 1 11 March March 2021. 2021. It is worth mentioning that software deﬁned radio (SDR) readers have been also MODELMODEL designed and describedOperatingOperating in Frequency Frequency the scientiﬁc Nominal literatureNominal Read Read [129, 130 ], but they are not commercially PicturePicture WebsiteWebsite Link Link (Manufacturer)(Manufacturer)(Manufacturer) available yet. BandBand RangeRange Most commercial readers are not able to directly provide information on the data https://www.tertiumtechno-https://www.tertiumtechno- IceKeyIceKeyIceKey HF HFHF (Tertium (Tertium(Tertium collected from the sensor tag. Typically, only a few parameters are provided on the 13.5613.56 MHz MHz (HF) (HF) 1010 cm cmcm logy.com/products/desktoplogy.com/products/desktoplogy.com/products/desktop--- Technology)Technology) graphical user interface, such as: hf/hf/ Sensors 2021, 21, x FOR PEER REVIEW 28 of 36

industries. However, differently from fixed readers, the antenna cannot be changed, thus the RFID system cannot be fully designed on the basis of the operative scenario. In HF RFID commercial integrated readers, the radiating element is represented by a coil, properly tuned at 13.56 MHz. On the other hand, in UHF RFID readers, different typologies of antennas are integrated to guarantee reliable detection when the tag is placed both on the reader case and at a longer distance [130]. Integrated reader’s typical cost is in the range of 700–1200 USD. - Portable/handheld or wearable readers. Handheld readers represent a lightweight and easy-to-carry solution, which can be used by a human operator for scanning tagged items placed in different positions (Table 11). In this framework, portable reader antennas must be integrated in a compact space, thus they are required to be lightweight, low profile, and compact in size. Consequently, the read range is limited to up to a few meters. Furthermore, the proximity of the human hand represents an important aspect to take into account during the antenna design process, because the dielectric properties of human tissues may significantly affect the antenna’s performance. Handheld readers’ typical cost is in the range of 200–900 USD. It is worth mentioning that software defined radio (SDR) readers have been also designed and described in the scientific literature [131,132], but they are not commercially available yet.

Table 9. Examples of commercial fixed RFID readers. The websites have been accessed on 1 March 2021.

MODEL Number of An- Picture Operating Frequency Band Website Link (Manufacturer) tenna Ports

https://www.caenr- Quattro‐R4321P 865–868 MHz (ETSI UHF) 4 fid.com/en/pro- (CAEN RFID s.r.l.) 902–928 MHz (FCC UHF) Sensors 2021, 21, 3138 ducts/quattro-r4321p/27 of 34

https://www.im- Impinj R700 reader 865–868 MHz (ETSI UHF) 4 pinj.com/products/rea- (Impinj) • 902–928 MHz (FCC UHF) Tag EPC, ders/impinj-r700 • RSSI (received signal strength indicator) of each detected tag, giving an estimation of the quality of the detection signal, https://www.im- Impinj Speedway 865–868 MHz (ETSI UHF) • Antenna port. Since fixed readers can support up4 to four externalpinj.com/products/rea- antennas, informa- (Impinj) 902–928 MHz (FCC UHF) tion on which antenna detected a tag is fundamental, ders/impinj-speedway • Time stamps of the first and last reads of each tag ID. https://www.zebra.co All of this information can be automatically sent to a management system/software, 865–868 MHz (ETSI UHF) m/it/it/pro- FX7500 (ZEBRA) or collected into a logfile (e.g., in .txt format) for post4 processing operations. Not all 902–928 MHz (FCC UHF) ducts/rfid/rfid-rea- commercial RFID readers provide information on the phase of the received backscattered ders/fx7500.html signal. This parameter is proportional to the distance between the reader antenna and the tag, and it can be used, for example, in phase-based localizationhttps://www.zebra.co systems and angle-of- 865–868 MHz (ETSI UHF) m/it/it/pro- FX9600 (ZEBRA) arrival (AoA) estimations if multiple antennas are used4 at the reader side. Among other commercial readers902– that928 provideMHz (FCC phase UHF) information, the Impinj Speedwayducts/rfid/rfid Revolution-rea- R420 UHF-RFID reader can be mentioned. ders/fx9600.html

TableTable 10. 10. ExamplesExamples of ofcommercial commercial integrated integrated RFID RFID readers readers.. The The websites websites have have been been accessed accessed on 1on March 1 March 2021. 2021.

MODELMODEL Operating Frequency NominalNominal Read Picture Operating Frequency Band WebsiteWebsite Link Link (Manufacturer)(Manufacturer) Band ReadRange Range Sensors 2021, 21, x FOR PEER REVIEW 29 of 36

SensorsSensors 2021 2021,, 21, 21,, xx, xFORFOR FOR PEERPEER PEER REVIEWREVIEW REVIEW https://www.tertiumtechno-2929 of of 36 36 SensorsIceKey 2021IceKey HF, 21 (Tertium, x HF FOR PEER REVIEW https://www.tertiumtechnology.29 of 36 13.5613.56 MHz (HF)(HF) 10 10 cm cm logy.com/products/desktop- (TertiumTechnology) Technology) com/products/desktop-hf/ Atlantis Desktop hf/ http://www.sensorid.it/pro- Atlantis Desktop HF/NFCAtlantisAtlantisAtlantis Desktop(Tertium Desktop Desktop 13.56 MHz (HF) 10 cm Atlantis Desktop http://www.sensorid.it/pro-ducts/atlantishttp://www.sensorid.it/pro-http://www.sensorid.it/-desktop.html HF/NFCHF/NFCTechnology)HF/NFC (Tertium (Tertium 13.5613.5613.56 MHz MHz MHz (HF)(HF) (HF) 10 10 cm cmcm http://www.sensorid.it/pro- HF/NFC (Tertium 13.56 MHz (HF) 10 cm ducts/atlantisproducts/atlantis-desktop.html-desktop.html (TertiumTechnology) Technology) 865–868 MHz (ETSI ducts/atlantisducts/atlantis-desktop.html-desktop.html Technology)Technology) https://www.tertiumtechno- IceKey UHF (Tertium 865–868UHF) MHz (ETSI 865865––868868 MHz MHz (ETSI (ETSI 1.2 m https://www.tertiumtechno-logy.com/products/desktop- IceKeyTechnology) UHF (Tertium 902–928UHF) MHz (FCC https://www.tertiumtechno-https://www.tertiumtechno- IceKeyIceKey IceKeyUHF UHF (Tertium UHF(Tertium 865–868 MHzUHF)UHF) (ETSI UHF) 1.2 m logy.com/products/desktophttps://www.tertiumtechnology.uhf/ - Technology) 902–928UHF) MHz (FCC 1.1.1.222 m mm logy.com/products/desktoplogy.com/products/desktop-- (TertiumTechnology)Technology) Technology) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) com/products/desktop-uhf/uhf/ 865–868UHF) MHz (ETSI uhf/uhf/ Slate‐R1260E/U UHF)UHF) https://www.caenr- 865–868UHF) MHz (ETSI Short- to me- USBSlate Desktop‐R1260E/U RAIN 865865––868868 MHz MHz (ETSI (ETSI fid.com/en/products/r1260https://www.caenr- - SlateSlateSlate-R1260E/U‐R1260E/U‐R1260E/U 902–928UHF) MHz (FCC SdiumhortShort-- -torange me- to https://www.caenr-https://www.caenr- USBRFID Desktop (CAEN RAINRFID) 865–868 MHzUHF)UHF) (ETSI UHF) SShorthort- -to to me- me- fid.com/en/products/r1260https://www.caenrfid.com/en/slate/ - USBUSBUSB Desktop Desktop Desktop RAIN RAIN RAIN 902–928UHF) MHz (FCC diummedium--range fid.com/en/products/r1260fid.com/en/products/r1260-- RFID (CAEN RFID) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) diumdium-range-range products/r1260-slate/slate/ RFIDRFIDRFID (CAEN (CAEN (CAEN RFID) RFID) RFID) 865–868UHF) MHz (ETSI range slate/slate/ UHF)UHF) http://www.sensorid.it/pro- Discovery Desktop 865865–868–868UHF) MHz MHz (ETSI (ETSI 865–868 MHz (ETSI 50 cm http://www.sensorid.it/pro-http://www.sensorid.it/pro-ducts/discovery-desk- DiscoveryDiscoveryDiscovery(SENSOR Desktop Desktop Desktop ID) 865–868902–928 MHzUHF)UHF) MHz (ETSI (FCC UHF) http://www.sensorid.it/pro-http://www.sensorid.it/ Discovery Desktop UHF) 5050 c cmm ducts/discoverytop.html -desk- (SENSOR(SENSOR ID) ID) 902–928902–928 MHzUHF) MHz (FCC (FCC UHF) 5500 c mcm products/discovery-desktop.htmlducts/discoveryducts/discovery-desk--desk- (SENSOR(SENSOR ID) ID) 902902––928928 MHz MHz (FCC (FCC top.html 865–868UHF) MHz (ETSI top.htmltop.html UHF)UHF) DiscoveryDiscovery Gate Gate PI PI 865–868865865–868–868 MHzUHF) MHz MHz (ETSI (ETSI (ETSI UHF) http://www.sensorid.it/pro-http://www.sensorid.it/ 865–868 MHz (ETSI 88 m m Discovery(SENSOR(SENSOR Gate ID) ID) PI 902–928902–928 MHzUHF) MHz (FCC (FCC UHF) ducts/discoveryhttp://www.sensorid.it/pro-products/discovery-gate-pi.html-gate-pi.html DiscoveryDiscovery Gate Gate PI PI UHF)UHF) 8 m http://www.sensorid.it/pro-http://www.sensorid.it/pro- (SENSOR ID) 902–928UHF) MHz (FCC 88 m m ducts/discovery-gate-pi.html (SENSOR(SENSOR ID) ID) 902902––928928 MHz MHz (FCC (FCC ducts/discoveryducts/discovery-gate-gate-pi.html-pi.html UHF) UHF)UHF) MostSometimes, commercial software readers modifications are not able and to directly add-ons provide are needed information at the reader on the side data to read collectedspecificMost from registerscommercial the sensor of the readers tag. RFID Typically, chip,are not thus able only providing to a fewdirectly parameters values provide from are information the provided sensors on (e.g., on the the temperature graph- data MostMost commercial commercial readers readers are are not not able able to to directly directly provide provide information information on on the the data data collectedicalor user humidity). interface,from the For sensor such example, as: tag. Typically, Farsens only demo a few software parameters patch are is availableprovided on online the graph- for testing collectedcollected from from the the sensor sensor tag. tag. Typically, Typically, only only a a few few parameters parameters are are provided provided on on the the graph- graph- icalbattery-free user interface, sensor such tags. as: However, a PC version of such software is only compatible with ical•ical user Taguser interface,EPC, interface, such such as: as: few UHF RFID readers, such as the Impinj Speedway (RX20), ThingMagic M6, Alien ALR- • • TagRSSITag EPC, (EPC,received signal strength indicator) of each detected tag, giving an estimation of • 9900,Tag Zebra EPC, FX9500, or Nordic ID Sampo S1/2. • • RSSItheRSSI quality ( received(received of the signal signal detection strength strength signal, indicator indicator ) )of of ea eachch detected detected tag, tag, giving giving an an estimation estimation of of • RSSIOn the(received other hand,signal chipless strength RFID indicator readers) of areeach not detected yet available tag, giving on the an market. estimation However, of • theAntennathe quality quality port. of of the theSince detection detection fixed readerssignal, signal, can support up to four external antennas, infor- theythe have quality been of proposed the detection in the signal, scientific literature, and some configurations are listed in • • AntennamationAntenna on port. port.which Since Since antenna fixed fixed detectedreaders readers cana can tag support supportis fundamental, up up to to four four external external antennas, antennas, infor- infor- • TableAntenna 12. In general,port. Since a chipless fixed readers RFID readercan support is a type up to of four radar external that can antennas, be designed infor- with • mationTimemation stamps on on which which of the antenna antenna first and detected detected last reads a atag tag of is eachis fundamental, fundamental, tag ID. eithermation a frequency on which domain antenna or adetected time domain a tag approachis fundamental, [131]. The frequency domain strategy • • TimeAllTime of stamps thisstamps information of of the the first first can and and be last lastautomatically reads reads of of each each sent tag tag to ID. ID.a management system/software, • expectsTime the stamps transmission of the first of and a harmonic, last reads whichof each variestag ID. in frequency within the operative or collectedAll of this into information a logfile (e.g., can in be .txt automatically format) for postsent processingto a management operations. system/software Not all com-, bandwidthAllAll of of this this of information information the tag. The can can reader be be automatically automatically architectures sent sent used to to a ina management management the literature system/software system/software are based on either, , ormercial collected RFID into readers a logfile provide (e.g., informationin .txt format) on for the post phase processing of the received operations. backscattered Not all com- sig- oror collected collected into into a a logfile logfile (e.g., (e.g., in in . txt.txt format) format) for for post post processing processing operations. operations. Not Not all all com- com- mercialnal. This RFID parameter readers is provide proportional information to the ondistance the phase between of the the received reader backscattered antenna and sig-the mercialmercial RFID RFID readers readers provide provide information information on on the the phase phase of of the the received received backscattered backscattered sig- sig- nal.tag, Thisand itparameter can be used is proportional, for example to, inthe phase distance-based between localization the reader systems antenna and angle and -theof- nal.nal. This This parameter parameter is is proportional proportional to to the the distance distance between between the the reader reader antenna antenna and and the the tag,arrival and (AoA) it can estimations be used, for if examplemultiple, antennasin phase -arebased used localization at the reader systems side. andAmong angle other-of- tag,tag, and and it it can can be be used used, ,for for example example, ,in in phase phase-based-based localization localization systems systems and and angle angle-of-of-- arrivalcommercial (AoA) readers estimations that provideif multiple phase antennas information, are used the at Impinjthe reader Speedway side. Among Revolution other arrivalarrival (AoA) (AoA) estimations estimations if if multiple multiple antennas antennas are are u usedsed at at the the reader reader side. side. Among Among other other commercialR420 UHF-RFID readers reader that can provide be mentioned phase information,. the Impinj Speedway Revolution commercialcommercial readers readers that that provide provide phase phase information, information, the the Impinj Impinj Speedway Speedway Revolution Revolution R420 Sometimes,UHF-RFID readersoftware can modifications be mentioned and. add-ons are needed at the reader side to read R420R420 UHF UHF-RFID-RFID reader reader can can be be mentioned mentioned. . specificSometimes, registers software of the RFID modifications chip, thus providingand add-ons values are needed from the at sensorsthe reader (e.g. side, tempera- to read Sometimes,Sometimes, software software modifications modifications and and add add-ons-ons are are needed needed at at the the reader reader side side to to read read specificture or humidity).registers of For the example,RFID chip Farsens, thus providing demo software values patch from isthe available sensors online(e.g., tempera- for test- specificspecific regist registersers of of the the RFID RFID chip chip, ,thus thus providing providing values values from from the the sensors sensors (e.g. (e.g., ,tempera- tempera- tureing batteryor humidity).-free sensor For example, tags. However, Farsens a demo PC version software of suchpatch software is available is only online compatible for test- tureture or or humidity). humidity). For For example, example, Farsens Farsens demo demo software software patch patch is is available available online online for for test- test- ingwith battery few UHF-free RFID sensor readers tags. ,However, such as the a ImpinjPC version Speedway of such (RX20), software ThingMagic is only compatible M6, Alien inging battery battery-free-free sensor sensor tags. tags. However, However, a a PC PC version version of of such such software software is is only only compatible compatible withALR -few9900, UHF Zebra RFID FX9500 readers, or, Nordicsuch as IDthe Sampo Impinj S1/2. Speedway (RX20), ThingMagic M6, Alien withwith few few UHF UHF RFID RFID readers readers, ,such such as as the the Impinj Impinj Speedway Speedway (RX20), (RX20), ThingMagic ThingMagic M6, M6, Alien Alien ALR-9900, Zebra FX9500, or Nordic ID Sampo S1/2. ALRALR-9900,-9900, Zebra Zebra FX9500 FX9500, ,or or Nordic Nordic ID ID Sampo Sampo S1/2. S1/2. Table 11. Examples of commercial handheld RFID readers. The websites have been accessed on 1 March, 2021. Table 11. Examples of commercial handheld RFID readers. The websites have been accessed on 1 March, 2021. MODEL Table(Manufactu-Table 11. 11. Examples Examples of of commercial commercial handheld handheldOperating RFID RFID readers. readers. Frequency The The websites websites have Nominalhave been been accessed accessedRead on on 1 1 March, March, 2021. 2021. Picture Website Link MODELMODEL (Manufactu-rer) (Manufactu- OperatingOperatingBand Frequency Frequency NominalNominalRange Read Read MODEL (Manufactu- PicturePicture Operating Frequency Nominal Read WebsiteWebsite Link Link rer)rer) Picture BandBand RangeRange Website Link rer) Band Range https://www.tertium- BLUEBERRY HS HF 13.56 MHz (HF) 6 cm https://www.tertium-technology.com/pro- (TertiumBLUEBERRY Technology) HS HF https://www.tertium-https://www.tertium- BLUEBERRYBLUEBERRY HS HS HF HF 13.56 MHz (HF) 6 cm technology.com/pro-ducts/desktop-hf/ (Tertium Technology) 13.5613.56 MHz MHz (HF) (HF) 66 cm cm technology.com/pro-technology.com/pro- (Tertium(Tertium Technology) Technology) ducts/desktop-hf/ ducts/desktopducts/desktop-hf/-hf/

Sensors 2021, 21, x FOR PEER REVIEW 29 of 36

Atlantis Desktop http://www.sensorid.it/pro- HF/NFC (Tertium 13.56 MHz (HF) 10 cm ducts/atlantis-desktop.html Technology) 865–868 MHz (ETSI https://www.tertiumtechno- IceKey UHF (Tertium UHF) 1.2 m logy.com/products/desktop- Technology) 902–928 MHz (FCC uhf/ UHF) 865–868 MHz (ETSI Slate‐R1260E/U https://www.caenr- UHF) Short- to me- USB Desktop RAIN fid.com/en/products/r1260- 902–928 MHz (FCC dium-range RFID (CAEN RFID) slate/ UHF) 865–868 MHz (ETSI http://www.sensorid.it/pro- Discovery Desktop UHF) 50 cm ducts/discovery-desk- (SENSOR ID) 902–928 MHz (FCC top.html UHF) 865–868 MHz (ETSI Discovery Gate PI UHF) http://www.sensorid.it/pro- 8 m (SENSOR ID) 902–928 MHz (FCC ducts/discovery-gate-pi.html UHF)

Most commercial readers are not able to directly provide information on the data collected from the sensor tag. Typically, only a few parameters are provided on the graphical user interface, such as: • Tag EPC, • RSSI (received signal strength indicator) of each detected tag, giving an estimation of the quality of the detection signal, • Antenna port. Since fixed readers can support up to four external antennas, information on which antenna detected a tag is fundamental, • Time stamps of the first and last reads of each tag ID.

Sensors 2021, 21, 3138 All of this information can be automatically sent to a management system/software28, of 34 or collected into a logfile (e.g., in .txt format) for post processing operations. Not all commercial RFID readers provide information on the phase of the received backscattered signal. This parameter is proportional to the distance between the reader antenna and the tag,the and stepped-frequency it can be used, for continuous example, wave in phase (SFCW)-based or localization FMCW concepts. systems Both and conﬁgurations angle-of- arrivalare equipped (AoA) estimations with a voltage-controlled if multiple antennas oscillator are used (VCO) at the in reader transmission, side. Among whose other center commercialfrequency readersis settled that by provide a control phase unit [ information,131]. A directional the Impinj coupler Speedway at the VCO Revolution output is R420used UHF as an-RFID input reader for the can demodulator. be mentioned A. reader founded on the time domain methodology transmitsSometimes, a sub-nanosecond software modifications pulse toward and add the- tag,ons are and needed measures at the the reader backscattered side to read signal specificfrom the regist tag.ers This of the technique RFID chip is, used thus inproviding impulse values radio UWBfrom the (IR-UWB) sensors technology,(e.g., tempera- where turethe or reader humidity). is composed For example, of a pulse Farsens generator demo software and a time patch domain is available receiver. online The transmittedfor test- ingpulse battery should-free besensor compliant tags. However, with the UWBa PC version regulation of such masks. software The power is only spectral compatible density with(PSD) few over UHF megahertz RFID readers is calculated, such as the over Impinj a period Speedway of time (RX20), much ThingMagic higher than M6, 1 ns Alien (1 ms for ALRthe- FCC,9900, Zebra 1 ms–1 FX9500µs for, theor Nordic ETSI) [ 131ID Sampo]. S1/2.

TableTable 11. 11. ExamplesExamples ofof commercial commercial handheld handheld RFID RFID readers. readers. The The websites websites have have been been accessed accessed on 1 on March, 1 March 2021. 2021.

MODELMODEL (Manufactu- Operating Frequency NominalNominal Read Read PicturePicture Operating Frequency Band WebsiteWebsite Link Link (Manufacturer)rer) Band RangeRange

https://www.tertium- SensorsBLUEBERRYBLUEBERRY 2021, 21, x FORHS HS HFPEER HF REVIEW https://www.tertiumtechnology.30 of 36 Sensors Sensors 20212021,, 2121,, xx FORFOR PEERPEER REVIEWREVIEW 13.5613.56 MHz MHz (HF) (HF) 6 cm6 cm technology.com/pro-3030 ofof 3636 (TertiumSensorsSensors(Tertium 2021 2021 Technology), ,21Technology) 21, ,x x FOR FOR PEER PEER REVIEW REVIEW com/products/desktop-hf/3030 ofof 3636 ducts/desktop-hf/

13.5613.56 MHz MHz (NFC) (NFC) https://www.chain- C71C71 UHF UHF RFID RFID Reader 13.5613.56 MHzMHz (NFC)(NFC) 2–42–4 cm cm (NFC) (NFC) https://www.chainway.net/https://www.chain-https://www.chain- C71C71 UHFUHF RFIDRFID 865–868865–86813.5613.56 MHz MHz MHz MHz (ETSI (ETSI (NFC) (NFC) UHF) UHF) 22––44 cmcm (NFC)(NFC) https://www.chain-https://www.chain-way.net/Pro- C71C71(CHAINWAY) UHF UHF RFID RFID 865865––868868 MHzMHz (ETSI(ETSI UHF)UHF) 3 m22 (UHF––44 cm cm RFID) (NFC) (NFC) Products/Info/41way.net/Pro-way.net/Pro- Reader (CHAINWAY) 902–928865865––868868 MHz MHz MHz (FCC (ETSI (ETSI UHF) UHF) UHF) 3 m (UHF RFID) way.net/Pro-way.net/Pro- ReaderReader (CHAINWAY) (CHAINWAY) 902865–928–868 MHz MHz (FCC (ETSI UHF) UHF) 3 3m m (UHF (UHF RFID) RFID) ducts/Info/41way.net/Pro- ReaderReader (CHAINWAY) (CHAINWAY) 902902–928–928 MHz MHz (FCC (FCC UHF) UHF) 33 m m (UHF (UHF RFID) RFID) ducts/Info/41ducts/Info/41 902902––928928 MHz MHz (FCC (FCC UHF) UHF) ducts/Info/41ducts/Info/41

https://www.tertium- BLUEBERRY HS UHF 865–868 MHz (ETSI UHF) https://www.tertium-https://www.tertium-https: BLUEBERRYBLUEBERRYBLUEBERRY HSHS HS UHFUHF UHF 865–868865865––868868 MHz MHzMHz (ETSI (ETSI(ETSI UHF) UHF)UHF) 30 cm technology.com/pro-https://www.tertium-https://www.tertium- (TertiumBLUEBERRYBLUEBERRY Technology) HS HS UHF UHF 902865865––928–868868 MHzMHz MHz (FCC(ETSI (ETSI UHF) UHF) UHF) 303030 cm cmcm //www.tertiumtechnology.com/technology.com/pro-technology.com/pro- (Tertium(Tertium(Tertium Technology)Technology) Technology) 902–928902902––928928 MHz MHzMHz (FCC (FCC(FCC UHF) UHF)UHF) 3030 cm cm dotto/blueberrytechnology.com/pro-technology.com/pro--hs-uhf/ (Tertium(Tertium Technology) Technology) 902902––928928 MHz MHz (FCC (FCC UHF) UHF) dotto/blueberryprodotto/blueberry-hs-uhf/dotto/blueberry--hshs--uhf/uhf/ dotto/blueberrydotto/blueberry--hshs--uhf/uhf/ dotto/blueberry-hs-uhf/

http://www.senso- Discovery Mobile 865–868 MHz (ETSI UHF) http://www.senso-http://www.senso-http: DiscoveryDiscoveryDiscovery Mobile MobileMobile UHF 865–868865865––868868 MHz MHzMHz (ETSI (ETSI(ETSI UHF) UHF)UHF) 50 cm rid.it/products/dis-http://www.senso-http://www.senso- UHFDiscoveryDiscovery (SENSOR Mobile Mobile ID) 902÷928865865––868868 MHzMHz MHz (FCC(ETSI (ETSI UHF) UHF) UHF) 505050 cm cmcm //www.sensorid.it/products/rid.it/products/dis-rid.it/products/dis- UHF (SENSOR ID) 902÷928÷ MHz (FCC UHF) 5050 cm cm rid.it/products/dis-rid.it/products/dis- UHFUHF(SENSOR (SENSOR (SENSOR ID) ID) ID) 902902÷928902÷928928 MHzMHz MHz (FCC (FCC (FCC UHF) UHF) UHF) 50 cm coveryrid.it/products/dis--mobile-uhf.html UHFUHF (SENSOR (SENSOR ID) ID) 902÷928902÷928 MHz MHz (FCC (FCC UHF) UHF) coverydiscovery-mobile-uhf.htmlcovery-mobile-mobile-uhf.html-uhf.html coverycovery--mobilemobile--uhf.htmluhf.html qID‐R1240IE/IU qIDqIDqID-R1240IE/IU‐‐R1240IE/IUR1240IE/IU https://www.caenr- qIDqID‐‐R1240IE/IUR1240IE/IU https://www.caenr- WearableWearableqID‐R1240IE/IU Bluetooth Bluetooth 865–868 MHz (ETSI UHF) https://www.caenr-https://www.caenr- WearableWearable Bluetooth Bluetooth 865–868865865–868–868 MHz MHz MHz (ETSI (ETSI (ETSI UHF) UHF) UHF) 1.5 m https://www.caenrfid.com/en/https://www.caenr-https://www.caenr-fid.com/en/pro- WearableRAINWearable RFID/BAR-RAIN Bluetooth Bluetooth 902865865––928–868868 MHzMHz MHz (FCC(ETSI (ETSI UHF) UHF) UHF) 1.51.51.5 m mm fid.com/en/pro-fid.com/en/pro- RAINRAIN RFID/BAR-RFID/BAR- 902902––928928 MHzMHz (FCC(FCC UHF)UHF) 1.51.5 m m ducts/r1240ieiufid.com/en/pro-fid.com/en/pro--qid/ RAINRAIN RFID/BAR- RFID/BAR- 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) products/r1240ieiu-qid/ducts/r1240ieiu-qid/ CODERAINRFID/BARCODE (CAEN RFID/BAR- RFID) 902–928 MHz (FCC UHF) ducts/r1240ieiuducts/r1240ieiu-qid/-qid/ CODECODE (CAEN (CAEN RFID) RFID) ducts/r1240ieiuducts/r1240ieiu--qid/qid/ CODECODEqIDmini(CAEN (CAEN (CAEN‐R1170I RFID) RFID) RFID) qIDminiqIDmini‐‐R1170IR1170I https://www.caenr- KeyfobqIDminiqIDmini Bluetooth‐‐R1170IR1170I 865–868 MHz (ETSI UHF) https://www.caenr-https://www.caenr- KeyfobKeyfobqIDmini-R1170 BluetoothBluetooth 865865––868868 MHzMHz (ETSI(ETSI UHF)UHF) 90 cm https://www.caenr-https://www.caenr-fid.com/en/pro- RAINKeyfobKeyfobIKeyfob RFID Bluetooth Bluetooth Bluetooth (CAEN 865–868902865865––928–868868 MHz MHzMHz MHz (ETSI (FCC(ETSI (ETSI UHF) UHF) UHF) UHF) 9090 cmcm https://www.caenrfid.com/en/fid.com/en/pro-fid.com/en/pro- RAINRAIN RFIDRFID (CAEN(CAEN 902902––928928 MHzMHz (FCC(FCC UHF)UHF) 9090 cm90 cm cm ducts/r1170ifid.com/en/pro-fid.com/en/pro--qidmini/ RAINRAINRAIN RFID RFID RFID (CAEN (CAEN 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) products/r1170i-qidmini/ducts/r1170i-qidmini/ RAINRFID) RFID (CAEN 902–928 MHz (FCC UHF) ducts/r1170iducts/r1170i-qidmini/-qidmini/ (CAENRFID)RFID) RFID) ducts/r1170iducts/r1170i--qidmini/qidmini/ RFID)RFID) https://www.zebra.com/i Culla RFID UHF 865–868 MHz (ETSI UHF) https://www.zebra.com/ihttps://www.zebra.com/i CullaCulla RFIDRFID UHFUHF 865865––868868 MHzMHz (ETSI(ETSI UHF)UHF) 6 m https://www.zebra.com/it/it/https://www.zebra.com/ihttps://www.zebra.com/it/it/products/rfid/rfid- RFD2000CullaCullaCulla RFID RFID RFID (ZEBRA) UHF UHF 865–868902865865––928–868868 MHz MHzMHz MHz (ETSI (FCC(ETSI (ETSI UHF) UHF) UHF) UHF) 66 mm t/it/products/rfid/rfidt/it/products/rfid/rfid-- RFD2000RFD2000 (ZEBRA)(ZEBRA) 902902––928928 MHzMHz (FCC(FCC UHF)UHF) 6 m66 m m products/rfid/rfid-handhelds.t/it/products/rfid/rfidt/it/products/rfid/rfidhandhelds.html -- RFD2000RFD2000RFD2000 (ZEBRA) (ZEBRA) 902–928902902––928928 MHz MHz MHz (FCC (FCC (FCC UHF) UHF) UHF) handhelds.htmlhandhelds.html handhelds.htmlhandhelds.htmlhtml

On the other hand, chipless RFID readers are not yet available on the market. How- OnOn thethe otherother hand,hand, chiplesschipless RFIDRFID readerreaderss areare notnot yetyet availableavailable onon thethe market.market. How-How- Tableever, 12. Performance OntheyOn the the have otherother been and hand,hand,cost proposed chiplesscomparisonchipless in RFIDRFID the of chiplessscientific readerreaderssRFID are areliterature, notnot tag yet readersyet availableavailableand [ 132some]. onon configurations thethe market.market. How-How- are ever,ever, theythey havehave beenbeen proposedproposed inin thethe scientificscientific literature,literature, andand somesome configurationsconfigurations areare listedever,ever, in theythey Table havehave 12 . been Inbeen general, proposedproposed a chipless inin thethe RFID scientificscientific reader literature,literature, is a type ofandand radar somesome that configurationsconfigurations can be designed areare Reference Operatinglistedlisted Bandwidth inin TableTable 1212.. InIn general,general, Reader aa chipless Systemchipless RFIDRFID readerreader Cost (Approx.) isis aa typetype ofof radarradar Background thatthat cancan bebe Calibration designeddesigned withlistedlisted either in in Table Table a frequency 12 12. .In In general, general, domain a a chipless chiplessor a time RFID RFID domain reader reader approach is is a a type type [133]. of of radar radar The that that frequency can can be be designed designeddomain withwith eithereither aa frequencyfrequency domaindomain oror aa timetime domaindomain approachapproach [133].[133]. TheThe frequencyfrequency domaindomain [43] 5–20.5strategywithwith GHzeithereither expects a a frequency frequency the tran smissiondomain VNA-baseddomain or orof aa timeharmonic,time domain domain which >$3000 approach approach varies [133]. [133].in frequency The The frequency frequency within yes domainthe domain op- strategystrategy expectsexpects thethe trantransmissionsmission ofof aa harmonic,harmonic, whichwhich variesvaries inin frequencyfrequency withinwithin thethe op-op- [132] 2–2.5erativestrategystrategy GHz bandwidth expects expects the the of tran tranthe smissiontag.smission VCO The readerof of a a harmonic, harmonic, architectures which which $120 used varies varies in in thein frequency frequency literature within withinare no based the the op-on op- erativeerative bandwidthbandwidth ofof thethe tag.tag. TheThe readerreader architecturesarchitectures usedused inin thethe literatureliterature areare basedbased onon [133]eithererative 4–8erative GHz the bandwidth bandwidth stepped-frequency of of the the tag. IR-UWBtag. continuous The The reader reader wavearchitectures architectures (SFCW) <$650 usedusedor FMCW in in thethe concepts.literature literature Bothare noare based based config- on on [134] 3.1–10.6eithereither the GHzthe steppedstepped--frequencyfrequency IR-UWB continuouscontinuous wavewave (SFCW)(SFCW) <$2500 oror FMCWFMCW concepts.concepts. BothBoth no config-config- urationseithereither thethe are steppedstepped equipped--frequencyfrequency with a voltage continuouscontinuous-controlled wavewave (SFCW)(SFCW)oscillator oror (VCO) FMCWFMCW in concepts. concepts.transmission, BothBoth whose config-config- [135] 2.4–3.4urationsurations GHz areare equippedequipped withwith a VCOa voltagevoltage--controlledcontrolled oscillatoroscillator <$1000 (VCO)(VCO) inin transmission,transmission, yes whosewhose centerurationsurations frequency areare equippedequipped is settled withwith by aa voltagecontrolvoltage- -unitcontrolledcontrolled [133]. A oscillatoroscillator directional (VCO)(VCO) coupler inin transmission,transmission, at the VCO output whosewhose centercenter frequencyfrequency isis settledsettled byby aa controlcontrol unitunit [133].[133]. AA directionaldirectional couplercoupler atat thethe VCOVCO outputoutput iscentercenter used frequency asfrequency an input is is forsettled settled the demodulator.by by a a control control unit unit A reader[133]. [133]. A Afounded directional directional on thecoupler coupler time at atdomain the the VCO VCO method- output output isis usedused asas anan inputinput forfor thethe demodulator.demodulator. AA readerreader foundedfounded onon thethe timetime domaindomain method-method- ologyisis usedused transmi asas anan ts inputinput a sub forfor-nanosecond thethe demodulator.demodulator. pulse toward AA readerreader the foundedfounded tag, and on onmeasures thethe timetime the domaindomain backscattered method-method- ologyology transmitransmitsts aa subsub--nanosecondnanosecond pulsepulse towardtoward thethe tagtag,, andand measuresmeasures thethe backscatteredbackscattered signalologyology from transmitransmi thetsts tag. aa subsub This--nanosecondnanosecond technique pulseispulse used towardtoward in impulse thethe tagtag radio, , andand UWB measuresmeasures (IR-UWB) thethe backscatteredbackscattered technology, signalsignal fromfrom thethe tag.tag. ThisThis techniquetechnique isis usedused inin impulseimpulse radioradio UWBUWB (IR(IR--UWB)UWB) technology,technology, wheresignalsignal the fromfrom reader thethe tag. tag.is composed ThisThis techniquetechnique of a pulse isis usedused generator inin impulseimpulse and a radioradio time UWBdomainUWB (IR(IR receiver.--UWB)UWB) technology, technology,The trans- wherewhere thethe readerreader isis composedcomposed ofof aa pulsepulse generatorgenerator andand aa timetime domaindomain receiver.receiver. TheThe trans-trans- mittedwherewhere pulthe the sereader reader should is is composed composedbe compliant of of a a pulsewith pulse thegenerator generator UWB regulation and and a a time time masks. domain domain The receiver. receiver. power The The spectral trans- trans- mittedmitted pulpulsese shouldshould bebe compliantcompliant withwith thethe UWBUWB regulationregulation masks.masks. TheThe powerpower spectralspectral densitymittedmitted pul(PSD)pulsese shouldovershould megahertz bebe compliantcompliant is calculated withwith thethe over UWBUWB a period regulationregulation of time masks.masks. much TheThe higher powerpower than spectralspectral 1 ns densitydensity (PSD)(PSD) overover megahertzmegahertz isis calculatedcalculated overover aa periodperiod ofof timetime muchmuch higherhigher thanthan 11 nsns (1densitydensity ms for (PSD) (PSD)the FCC, overover 1 megahertzmsmegahertz–1 μs for is isthe calculatedcalculated ETSI) [133]. overover aa periodperiod ofof timetime muchmuch higherhigher thanthan 11 nsns (1(1 msms forfor thethe FCC,FCC, 11 msms––11 μsμs forfor thethe ETSI)ETSI) [133].[133]. (1(1 ms ms for for the the FCC, FCC, 1 1 ms ms––11 μs μs for for the the ETSI) ETSI) [133]. [133]. Table 12. Performance and cost comparison of chipless RFID tag readers [134]. TableTable 12.12. PerformancePerformance andand costcost comparisoncomparison ofof chiplesschipless RFIDRFID tagtag readersreaders [134].[134]. TableTable 12. 12. Performance Performance and and cost cost comparison comparison of of chipless chipless RFID RFID tag tag readers readers [134]. [134]. Reference Operating bandwidth Reader System Cost (approx.) Background calibration ReferenceReference OperatingOperating bandwidthbandwidth ReaderReader SystemSystem CostCost (approx.)(approx.) BackgroundBackground calibrationcalibration ReferenceReference[43] OperatingOperating5–20.5 bandwidthGHzbandwidth ReaderReaderVNA-based System System CostCost> $3000 (approx.) (approx.) BackgroundBackgroundyes calibrationcalibration [43][43] 55––20.520.5 GHzGHz VNAVNA--basedbased >> $3000$3000 yesyes [134][43][43] 255––2.520.520.5 GHz GHz GHz VNAVNAVCO--basedbased >>$120 $3000 $3000 noyesyes [134][134] 22––2.52.5 GHzGHz VCOVCO $120$120 nono [135][134][134] 242––2.582.5 GHz GHz GHz IR-VCOUWBVCO <$650$120$120 nonono [135][135] 44––88 GHzGHz IRIR--UWBUWB <$650<$650 nono [136][135][135] 3.1–4410.6––88 GHz GHz GHz IRIRIR--UWB-UWBUWB <$2500<$650<$650 nonono [136][136] 3.13.1––10.610.6 GHzGHz IRIR--UWBUWB <$2500<$2500 nono [137][136][136] 2.43.13.1––3.410.610.6 GHz GHz GHz IRIRVCO--UWBUWB <$1000<$2500<$2500 yesnono [137][137] 2.42.4––3.43.4 GHzGHz VCOVCO <$1000<$1000 yesyes [137][137] 2.42.4––3.43.4 GHz GHz VCOVCO <$1000<$1000 yesyes 8. Conclusions 88.. ConclusionConclusionss 88. .Conclusion Conclusionss

Sensors 2021, 21, 3138 29 of 34

8. Conclusions A comprehensive overview of RFID sensing technology has been presented. The advantages of RFID sensors with respect to classical battery-equipped sensor nodes are highlighted. Then, a detailed classiﬁcation of RFID sensors is provided, and the different architectures are described and discussed. The main areas of application of RFID sensors have been presented. Finally, some commercial RFID sensing solutions are summarized, as well as the readers available on the market.

Author Contributions: Conceptualization, F.C.; writing—original draft preparation, F.C., S.G., M.B., A.M. and F.A.D.; review and editing, F.C., S.G., M.B., A.M., F.A.D. and G.M.; supervision, G.M. All authors have read and agreed to the published version of the manuscript. Funding: This work was partially supported by the Italian Ministry of Education and Research (MIUR) in the framework of the CrossLab project (Departments of Excellence) and PRIN2017 GREEN TAGS “Chipless radio frequency identification (RFID) for GREEN TAGging and Sens- ing” (2017NT5W7Z). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: In this section, you can acknowledge any support given which is not covered by the author contribution or funding sections. This may include administrative and technical support, or donations in kind (e.g., materials used for experiments). Conflicts of Interest: The authors declare no conflict of interest.

References 1. RFID Forecasts, Players and Opportunities 2018–2028: IDTechEx. Available online: https://www.idtechex.com/en/research- report/rfid-forecasts-players-and-opportunities-2018-2028/642 (accessed on 27 August 2020). 2. RFID Sensors 2018–2028: Forecasts, Players, Opportunities: IDTechEx. Available online: https://www.idtechex.com/en/ research-report/rfid-sensors-2018-2028-forecasts-players-opportunities/556 (accessed on 27 August 2020). 3. Potyrailo, R.A.; Morris, W.G.; Sivavec, T.; Tomlinson, H.W.; Klensmeden, S.; Lindh, K. RFID Sensors Based on Ubiquitous Passive 13.56-MHz RFID Tags and Complex Impedance Detection. Wirel. Commun. Mob. Comput. 2009, 9, 1318–1330. [CrossRef] 4. De Donno, D.; Stefanizzi, M.L.; Catarinucci, L.; Mainetti, L.; Patrono, L.; Tarricone, L. Integrating Passive UHF RFID Tags with WSN Nodes: Challenges and Opportunities. J. Commun. Softw. Syst. 2014, 10, 99–106. [CrossRef] 5. Cui, L.; Zhang, Z.; Gao, N.; Meng, Z.; Li, Z. Radio Frequency Identification and Sensing Techniques and Their Applications—A Review of the State-of-the-Art. Sensors 2019, 19, 4012. [CrossRef][PubMed] 6. AG, I.T. Transceivers—Infineon Technologies. Available online: https://www.infineon.com/cms/en/product/transceivers/ (accessed on 3 September 2020). 7. RF Integrated Transmitters, Receivers, & Transceivers|Analog Devices. Available online: https://www.analog.com/en/ products/rf-microwave/integrated-transceivers-transmitters-receivers.html (accessed on 20 March 2021). 8. Rokhsaz, S.; Young, B.D.; Younis, A.; Paulos, J.J.; Misra, A.; Cooke, B.; Hassoun, M. Radio Frequency Identification (RFID) Tag(s) and Sensor(s). U.S. Patent 9748632 B2, 29 August 2017. 9. Philipose, M.; Smith, J.R.; Jiang, B.; Mamishev, A.; Roy, S.; Sundara-Rajan, K. Battery-Free Wireless Identification and Sensing. IEEE Pervasive Comput. 2005, 4, 37–45. [CrossRef] 10. Pardo, D.; Vaz, A.; Gil, S.; Gomez, J.; Ubarretxena, A.; Puente, D.; Morales-Ramos, R.; Garcia-Alonso, A.; Berenguer, R. Design Criteria for Full Passive Long Range UHF RFID Sensor for Human Body Temperature Monitoring. In Proceedings of the 2007 IEEE International Conference on RFID, Grapevine, TX, USA, 26–28 March 2007; pp. 141–148. 11. MIT Engineers Configure RFID Tags to Work as Sensors. Available online: https://news.mit.edu/2018/mit-engineers-configure- rfid-tags-to-work-as-sensors-0613 (accessed on 24 August 2020). 12. Active RFID vs. Passive RFID: What’s the Difference? Available online: https://www.atlasrfidstore.com/rfid-insider/active-rfid- vs-passive-rfid (accessed on 21 September 2020). 13. RFID EDUCATION—Kontinental Establishment. Available online: http://www.ce.com.pk/rfid-solution/ (accessed on 9 February 2021). 14. RFID Tag—Battery Assisted Passive Tags—Corerfid. Available online: https://www.corerfid.com/rfid-technology/rfid-tracking/ (accessed on 9 February 2021). 15. Monza X-2K Dura Datasheet. Available online: https://support.impinj.com/hc/en-us/articles/202756848-Monza-X-2K-Dura- Datasheet (accessed on 15 March 2021). Sensors 2021, 21, 3138 30 of 34

16. Hande, A.; Bridgelall, R.; Bhatia, D. Energy Harvesting for Active RF Sensors and ID Tags. In Energy Harvesting Technologies; Priya, S., Inman, D.J., Eds.; Springer: Boston, MA, USA, 2009; pp. 459–492, ISBN 978-0-387-76464-1. 17. Self-Powered Smart Active RFID Tag Integrated with Wearable Hybrid Nanogenerator. Nano Energy 2019, 64, 103911. [CrossRef] 18. Vaz, A.; Ubarretxena, A.; Zalbide, I.; Pardo, D.; Solar, H.; Garcia-Alonso, A.; Berenguer, R. Full Passive UHF Tag with a Temperature Sensor Suitable for Human Body Temperature Monitoring. IEEE Trans. Circuits Syst. II Express Briefs 2010, 57, 95–99. [CrossRef] 19. Catarinucci, L.; Colella, R.; Esposito, A.; Tarricone, L.; Zappatore, M. RFID Sensor-Tags Feeding a Context-Aware Rule-Based Healthcare Monitoring System. J. Med. Syst. 2012, 36, 3435–3449. [CrossRef] 20. Shenghua, Z.; Nanjian, W. A Novel Ultra Low Power Temperature Sensor for UHF RFID Tag Chip. In Proceedings of the 2007 IEEE Asian Solid-State Circuits Conference, Jeju, Korea, 12–14 November 2007; pp. 464–467. 21. Sosa, C. How to Use Power Scaling to Maximize Power Savings in a SAR ADC System. Available online: https://www.ti.com/ lit/an/slyt754/slyt754.pdf (accessed on 9 February 2021). 22. Yeager, D.J.; Powledge, P.S.; Prasad, R.; Wetherall, D.; Smith, J.R. Wirelessly-Charged UHF Tags for Sensor Data Collection. In Proceedings of the 2008 IEEE International Conference on RFID, Las Vegas, NV, USA, 16–17 April 2008; pp. 320–327. 23. Marrocco, G. Pervasive Electromagnetics: Sensing Paradigms by Passive RFID Technology. IEEE Wirel. Commun. 2010, 17, 10–17. [CrossRef] 24. Kim, S.; Kawahara, Y.; Georgiadis, A.; Collado, A.; Tentzeris, M.M. Low-Cost Inkjet-Printed Fully Passive RFID Tags for Calibration-Free Capacitive/Haptic Sensor Applications. IEEE Sens. J. 2015, 15, 3135–3145. [CrossRef] 25. AutoTuneTM Technology White Paper. Available online: https://support.impinj.com/hc/en-us/articles/202356796-AutoTune- Technology-White-Paper (accessed on 28 January 2021). 26. RFM3300-D Magnus®-S3 M3D Passive Sensor, IC. Available online: https://axzon.com/rfm3300-d-magnus-s3-m3d-passive- sensor-ic/ (accessed on 9 February 2021). 27. Sensor Patch|Avery Dennison|RFID. Available online: https://rfid.averydennison.com/content/rfid/na/en/home/product- finder/sensor-patch.html (accessed on 28 January 2021). 28. Wegleiter, H.; Schweighofer, B.; Deinhammer, C.; Holler, G.; Fulmek, P. Automatic Antenna Tuning Unit to Improve RFID System Performance. IEEE Trans. Instrum. Meas. 2011, 60, 2797–2803. [CrossRef] 29. Caccami, M.C.; Marrocco, G. Electromagnetic Modeling of Self-Tuning RFID Sensor Antennas in Linear and Nonlinear Regimes. IEEE Trans. Antennas Propag. 2018, 66, 2779–2787. [CrossRef] 30. Rasilainen, K.; Viikari, V.V. Transponder Designs for Harmonic Radar Applications. Available online: https://www.hindawi. com/journals/ijap/2015/565734/ (accessed on 2 September 2020). 31. Rasilainen, K.; Ilvonen, J.; Lehtovuori, A.; Hannula, J.-M.; Viikari, V. On Design and Evaluation of Harmonic Transponders. IEEE Trans. Antennas Propag. 2015, 63, 15–23. [CrossRef] 32. Alimenti, F.; Roselli, L. Theory of Zero-Power RFID Sensors Based on Harmonic Generation and Orthogonally Polarized Antennas. Prog. Electromagn. Res. 2013, 134, 337–357. [CrossRef] 33. Dey, S.; Saha, J.K.; Karmakar, N.C. Smart Sensing: Chipless RFID Solutions for the Internet of Everything. IEEE Microw. Mag. 2015, 16, 26–39. [CrossRef] 34. Genovesi, S.; Costa, F.; Borgese, M.; Dicandia, A.F.; Manara, G.; Tedjini, S.; Perret, E. Enhanced Chipless RFID Tags for Sensors Design. In Proceedings of the 2016 IEEE International Symposium on Antennas and Propagation (APSURSI), Fajardo, PR, USA, 26 June–1 July 2016; pp. 1275–1276. 35. Vena, A.; Perret, E.; Kaddour, D.; Baron, T. Toward a Reliable Chipless RFID Humidity Sensor Tag Based on Silicon Nanowires. IEEE Trans. Microw. Theory Tech. 2016, 64, 2977–2985. [CrossRef] 36. Amin, E.M.; Saha, J.K.; Karmakar, N.C. Smart Sensing Materials for Low-Cost Chipless RFID Sensor. IEEE Sens. J. 2014, 14, 2198–2207. [CrossRef] 37. Genovesi, S.; Costa, F.; Borgese, M.; Dicandia, F.A.; Monorchio, A.; Manara, G. Chipless RFID Sensor for Rotation Monitor- ing. In Proceedings of the 2017 IEEE International Conference on RFID Technology Application (RFID-TA), Warsaw, Poland, 20–22 September 2017; pp. 233–236. 38. Ando, B.; Baglio, S. Inkjet-Printed Sensors: A Useful Approach for Low Cost, Rapid Prototyping. IEEE Instrum. Meas. Mag. 2011, 14, 36–40. [CrossRef] 39. Lázaro, A.; Villarino, R.; Costa, F.; Genovesi, S.; Gentile, A.; Buoncristiani, L.; Girbau, D. Chipless Dielectric Constant Sensor for Structural Health Testing. IEEE Sens. J. 2018, 18, 5576–5585. [CrossRef] 40. Ramos, A.; Girbau, D.; Lazaro, A.; Villarino, R. Wireless Concrete Mixture Composition Sensor Based on Time-Coded UWB RFID. IEEE Microw. Wirel. Compon. Lett. 2015, 25, 681–683. [CrossRef] 41. Lazaro, A.; Ramos, A.; Girbau, D.; Villarino, R. Chipless UWB RFID Tag Detection Using Continuous Wavelet Transform. IEEE Antennas Wirel. Propag. Lett. 2011, 10, 520–523. [CrossRef] 42. Harrop, P.; Das, R. Printed and Chipless RFID Forecasts, Technologies & Players 2009–2029. Available online: https://www. giiresearch.com/report/ix243391-rfid-forecasts-players-opportunities-2012-2022.html (accessed on 9 February 2021). 43. Blischak, A.T.; Manteghi, M. Embedded Singularity Chipless RFID Tags. IEEE Trans. Antennas Propag. 2011, 59, 3961–3968. [CrossRef] Sensors 2021, 21, 3138 31 of 34

44. Costa, F.; Borgese, M.; Gentile, A.; Buoncristiani, L.; Genovesi, S.; Dicandia, F.A.; Bianchi, D.; Monorchio, A.; Manara, G. Robust Reading Approach for Moving Chipless RFID Tags by Using ISAR Processing. IEEE Trans. Microw. Theory Tech. 2018, 66, 2442–2451. [CrossRef] 45. Gupta, S.; Nikfal, B.; Caloz, C. Chipless RFID System Based on Group Delay Engineered Dispersive Delay Structures. IEEE Antennas Wirel. Propag. Lett. 2011, 10, 1366–1368. [CrossRef] 46. Preradovic, S.; Balbin, I.; Karmakar, N.C.; Swiegers, G.F. Multiresonator-Based Chipless RFID System for Low-Cost Item Tracking. IEEE Trans. Microw. Theory Tech. 2009, 57, 1411–1419. [CrossRef] 47. Vena, A.; Perret, E.; Tedjini, S. Chipless RFID Tag Using Hybrid Coding Technique. IEEE Trans. Microw. Theory Tech. 2011, 59, 3356–3364. [CrossRef] 48. Raju, R.; Bridges, G.E.; Bhadra, S. Wireless Passive Sensors for Food Quality Monitoring: Improving the Safety of Food Products. IEEE Antennas Propag. Mag. 2020, 76–89. [CrossRef] 49. Fletcher, R.R.; Gershenfeld, N.A. Remotely Interrogated Temperature Sensors Based on Magnetic Materials. IEEE Trans. Magn. 2000, 36, 2794–2795. [CrossRef] 50. Buff, W.; Klett, S.; Rusko, M.; Ehrenpfordt, J.; Goroli, M. Passive Remote Sensing for Temperature and Pressure Using SAW Resonator Devices. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 1998, 45, 1388–1392. [CrossRef] 51. Schueler, M.; Mandel, C.; Puentes, M.; Jakoby, R. Metamaterial Inspired Microwave Sensors. IEEE Microw. Mag. 2012, 13, 57–68. [CrossRef] 52. Feng, Y.; Xie, L.; Chen, Q.; Zheng, L.-R. Low-Cost Printed Chipless RFID Humidity Sensor Tag for Intelligent Packaging. IEEE Sens. J. 2015, 15, 3201–3208. [CrossRef] 53. Javed, N.; Habib, A.; Amin, Y.; Loo, J.; Akram, A.; Tenhunen, H. Directly Printable Moisture Sensor Tag for Intelligent Packaging. IEEE Sens. J. 2016, 16, 6147–6148. [CrossRef] 54. Borgese, M.; Dicandia, F.; Costa, F.; Genovesi, S.; Manara, G. An Inkjet Printed Chipless RFID Sensor for Wireless Humidity Monitoring. IEEE Sens. J. 2017, 17, 4699–4707. [CrossRef] 55. Karmakar, N.C.; Amin, E.M.; Saha, J.K. Chipless RFID Sensors; John Wiley & Sons: Hoboken, NJ, USA, 2016; ISBN 978-1-118-93600-9. 56. Mc Gee, K.; Anandarajah, P.; Collins, D. A Review of Chipless Remote Sensing Solutions Based on RFID Technology. Sensors 2019, 19, 4829. [CrossRef] 57. Costa, F.; Gentile, A.; Genovesi, S.; Buoncristiani, L.; Lazaro, A.; Villarino, R.; Girbau, D. A Depolarizing Chipless RF Label for Dielectric Permittivity Sensing. IEEE Microw. Wirel. Compon. Lett. 2018, 28, 371–373. [CrossRef] 58. Vena, A.; Perret, E.; Tedjni, S. A Depolarizing Chipless RFID Tag for Robust Detection and Its FCC Compliant UWB Reading Sys- tem. IEEE Trans. Microw. Theory Tech. 2013, 61, 2982–2994. [CrossRef] 59. Genovesi, S.; Costa, F.; Dicandia, F.A.; Borgese, M.; Manara, G. Orientation-Insensitive and Normalization-Free Reading Chipless RFID System Based on Circular Polarization Interrogation. IEEE Trans. Antennas Propag. 2020, 68, 2370–2378. [CrossRef] 60. Costa, F.; Genovesi, S.; Monorchio, A. Normalization-Free Chipless RFIDs by Using Dual-Polarized Interrogation. IEEE Trans. Microw. Theory Tech. 2016, 64, 310–318. [CrossRef] 61. Borgese, M.; Genovesi, S.; Manara, G.; Costa, F. Radar Cross Section of Chipless RFID Tags and BER Performance. IEEE Trans. Antennas Propag. 2020, 1. [CrossRef] 62. Aliasgari, J.; Forouzandeh, M.; Karmakar, N. Chipless RFID Readers for Frequency-Coded Tags: Time-Domain or Frequency- Domain? IEEE J. Radio Freq. Identif. 2020, 4, 146–158. [CrossRef] 63. Garbati, M.; Perret, E.; Siragusa, R. Chipless RFID Reader Design for Ultra-Wideband Technology: Design, Realization and Characteriza- tion; Elsevier: Amsterdam, The Netherlands, 2018; ISBN 978-0-08-102761-5. 64. Plessky, V.P.; Reindl, L.M. Review on SAW RFID Tags. IEEE Trans. Ultrason. Ferroelectr. Freq. Control. 2010, 57, 654–668. [CrossRef] 65. Pohl, A.; Ostermayer, G.; Reindl, L.; Seifert, F. Monitoring the Tire Pressure at Cars Using Passive SAW Sensors. In Proceedings of the 1997 IEEE Ultrasonics Symposium Proceedings. An International Symposium (Cat. No.97CH36118), Toronto, ON, Canada, 5–8 October 1997; Volume 1, pp. 471–474. 66. Seifert, F.; Bulst, W.-E.; Ruppel, C. Mechanical Sensors Based on Surface Acoustic Waves. Sens. Actuators Phys. 1994, 44, 231–239. [CrossRef] 67. Pohl, A. A Review of Wireless SAW Sensors. IEEE Trans. Ultrason. Ferroelectr. Freq. Control. 2000, 47, 317–332. [CrossRef] [PubMed] 68. Puccio, D.; Malocha, D.C.; Gallagher, D.; Hines, J. SAW Sensors Using Orthogonal Frequency Coding. In Proceedings of the 2004 IEEE International Frequency Control Symposium and Exposition, Montreal, QC, Canada, 23–27 August 2004; pp. 307–310. 69. Das, R.; Harrop, P. RFID Forecasts, Players and Opportunities 2009–2019. Available online: https://www.idtechex.com/en/ research-report/printed-and-chipless-rﬁd-forecasts-technologies-and-players-2009-2019/225 (accessed on 9 February 2021). 70. Cantatore, E.; Geuns, T.C.T.; Gelinck, G.H.; van Veenendaal, E.; Gruijthuijsen, A.F.A.; Schrijnemakers, L.; Drews, S.; de Leeuw, D.M. A 13.56-MHz RFID System Based on Organic Transponders. IEEE J. Solid-State Circuits 2007, 42, 84–92. [CrossRef] 71. Conti, S.; Pimpolari, L.; Calabrese, G.; Worsley, R.; Majee, S.; Polyushkin, D.K.; Paur, M.; Pace, S.; Keum, D.H.; Fabbri, F.; et al. Low- Voltage 2D Materials-Based Printed Field-Effect Transistors for Integrated Digital and Analog Electronics on Paper. Nat. Commun. 2020, 11, 3566. [CrossRef] 72. Myny, K. The Development of Flexible Integrated Circuits Based on Thin-Film Transistors. Nat. Electron. 2018, 1, 30–39. [CrossRef] Sensors 2021, 21, 3138 32 of 34

73. Khan, Y.; Thielens, A.; Muin, S.; Ting, J.; Baumbauer, C.; Arias, A.C. A New Frontier of Printed Electronics: Flexible Hybrid Elec- tronics. Adv. Mater. 2020, 32, 1905279. [CrossRef] 74. Towards Electronic Product Coding with RFID Tags Based on Hybrid Organic-Oxide Complementary Thin-Film Technol- ogy|ORICLA Project|FP7|CORDIS|European Commission. Available online: https://cordis.europa.eu/project/id/247798 (accessed on 14 September 2020). 75. Want, R. Enabling Ubiquitous Sensing with RFID. Computer 2004, 37, 84–86. [CrossRef] 76. Payton, F.C.; Pare, G.; Le Rouge, C.M.; Reddy, M. Health Care IT: Process, People, Patients and Interdisciplinary Considerations. J. Assoc. Inf. Syst. 2011, 12, 4. [CrossRef] 77. Wiener, J.M.; Tilly, J. Population Ageing in the United States of America: Implications for Public Programmes. Int. J. Epidemiol. 2002, 31, 776–781. [CrossRef] 78. Amendola, S.; Lodato, R.; Manzari, S.; Occhiuzzi, C.; Marrocco, G. RFID Technology for IoT-Based Personal Healthcare in Smart Spaces. IEEE Internet Things J. 2014, 1, 144–152. [CrossRef] 79. TASA: Tag-Free Activity Sensing Using RFID Tag Arrays—IEEE Journals & Magazine. Available online: https://ieeexplore. ieee.org/abstract/document/5477420?casa_token=ecj9UpnmNgYAAAAA:KVKxKGYdeBlHI0NO8_Z1tEOVRWL6CTldHC4 1eEGVvZqwudQVDhfZ83B_dUKe3Awzcx2Q6OH4fIE (accessed on 26 September 2020). 80. Krigslund, R.; Dosen, S.; Popovski, P.; Dideriksen, J.L.; Pedersen, G.F.; Farina, D. A Novel Technology for Motion Capture Using Passive UHF RFID Tags. IEEE Trans. Biomed. Eng. 2013, 60, 1453–1457. [CrossRef] 81. Manzari, S.; Occhiuzzi, C.; Marrocco, G. Feasibility of Body-Centric Systems Using Passive Textile RFID Tags. IEEE Antennas Propag. Mag. 2012, 54, 49–62. [CrossRef] 82. Potyrailo, R.A.; Nagraj, N.; Tang, Z.; Mondello, F.J.; Surman, C.; Morris, W. Battery-Free Radio Frequency Identification (RFID) Sensors for Food Quality and Safety. J. Agric. Food Chem. 2012, 60, 8535–8543. [CrossRef] 83. Fletcher, R.R. Low-Cost Electromagnetic Tagging: Design and Implementation. Ph.D. Thesis, Massachusetts Institute of Technol- ogy, Cambridge, MA, USA, 2002. 84. TagSense, RFID, Sensors, Chipless, Consulting, Development, Vendor. Available online: http://www.tagsense.com/ingles/ products/product_chipless.html (accessed on 23 September 2020). 85. Tao, H.; Brenckle, M.A.; Yang, M.; Zhang, J.; Liu, M.; Siebert, S.M.; Averitt, R.D.; Mannoor, M.S.; McAlpine, M.C.; Rogers, J.A.; et al. Silk-Based Conformal, Adhesive, Edible Food Sensors. Adv. Mater. 2012, 24, 1067–1072. [CrossRef] 86. Wu, S.-Y.; Yang, C.; Hsu, W.; Lin, L. 3D-Printed Microelectronics for Integrated Circuitry and Passive Wireless Sensors. Microsyst. Nanoeng. 2015, 1, 1–9. [CrossRef] 87. Ruiz-Garcia, L.; Lunadei, L. The Role of RFID in Agriculture: Applications, Limitations and Challenges. Comput. Electron. Agric. 2011, 79, 42–50. [CrossRef] 88. García, L.; Parra, L.; Jimenez, J.M.; Lloret, J.; Lorenz, P. IoT-Based Smart Irrigation Systems: An Overview on the Recent Trends on Sensors and IoT Systems for Irrigation in Precision Agriculture. Sensors 2020, 20, 1042. [CrossRef] 89. Palazzi, V.; Gelati, F.; Vaglioni, U.; Alimenti, F.; Mezzanotte, P.; Roselli, L. Leaf-Compatible Autonomous RFID-Based Wireless Temperature Sensors for Precision Agriculture. In Proceedings of the 2019 IEEE Topical Conference on Wireless Sensors and Sensor Networks (WiSNet), Orlando, FL, USA, 20–23 January 2019; pp. 1–4. 90. Daskalakis, S.N.; Goussetis, G.; Assimonis, S.D.; Tentzeris, M.M.; Georgiadis, A. A UW Backscatter-Morse-Leaf Sensor for Low-Power Agricultural Wireless Sensor Networks. IEEE Sens. J. 2018, 18, 7889–7898. [CrossRef] 91. Wang, J.; Chang, L.; Aggarwal, S.; Abari, O.; Keshav, S. Soil Moisture Sensing with Commodity RFID Systems. In Proceedings of the 18th International Conference on Mobile Systems, Applications, and Services, Toronto, ON, Canada, 15–19 June 2020; Association for Computing Machinery: New York, NY, USA, 2020; pp. 273–285. 92. Da Fonseca, N.S.S.M.; Freire, R.C.S.; Batista, A.; Fontgalland, G.; Tedjini, S. A Passive Capacitive Soil Moisture and Environment Temperature UHF RFID Based Sensor for Low Cost Agricultural Applications. In Proceedings of the 2017 SBMO/IEEE MTT-S International Microwave and Optoelectronics Conference (IMOC), Aguas de Lindoia, Brazil, 27–30 August 2017; pp. 1–4. 93. Caccami, M.C.; Amendola, S.; Occhiuzzi, C. Method and System for Reading RFID Tags Embedded into Tires on Convey- ors. In Proceedings of the 2019 IEEE International Conference on RFID Technology and Applications (RFID-TA), Pisa, Italy, 25–27 September 2019; pp. 141–144. 94. Grosinger, J.; Mayer, L.W.; Mecklenbrauker, C.F.; Scholtz, A.L. Input Impedance Measurement of a Dipole Antenna Mounted on a Car Tire. In Proceedings of the 2009 International Symposium on Antennas and Propagation (ISAP 2009), Bangkok, Thailand, 20–23 October 2009. 95. Shao, S.; Kiourti, A.; Burkholder, R.; Volakis, J. Broadband and Flexible Textile RFID Tags for Tires. In Proceedings of the 2014 IEEE Antennas and Propagation Society International Symposium (APSURSI), Memphis, TN, USA, 6–11 July 2014; p. 1507. 96. Liang, Z.; Ouyang, J.; Yang, F.; Zhou, L. Design of License Plate RFID Tag Antenna Using Characteristic Mode Pattern Synthesis. IEEE Trans. Antennas Propag. 2017, 65, 4964–4970. [CrossRef] 97. Mobashsher, A.T.; Pretorius, A.J.; Abbosh, A.M. Low-Profile Vertical Polarized Slotted Antenna for On-Road RFID-Enabled Intelligent Parking. IEEE Trans. Antennas Propag. 2020, 68, 527–532. [CrossRef] 98. Lee, J.K.; Spach, D.R. Wireless System to Detect Presence of Child in a Baby Car Seat. U.S. Patent 773,322,8 B2, 8 June 2010. 99. Leschke, A.; Weinert, F.; Obermeier, M.; Kubica, S.; Bonaiuto, V. Method for Classification of Frontal Collision Events in Passenger Cars Based on Measurement of Local Component-Specific Decelerations. Int. J. Automot. Technol. 2020, 21, 785–794. [CrossRef] Sensors 2021, 21, 3138 33 of 34

100. Available online: Https://Shop.Semiperdo.Com/Prodotto/Mymi-Dispositivo-Anti-Abbandono-Bimbi-in-Auto/ (accessed on 16 March 2021). 101. Available online: Https://Www.Rfid-Wiot-Search.Com/Toennjes-Less-Bureaucracy-More-Security-Vehicles-Equipped-with- Rfid-1508 (accessed on 16 March 2021). 102. Clever Tyres—How Tyres, TPMS and RFID Are Combining. Available online: https://www.tyrepress.com/2014/03/clever- tyres-how-tyres-tpms-and-rfid-are-combining/ (accessed on 30 January 2021). 103. Zhang, J.; Tian, G.Y.; Marindra, A.M.J.; Sunny, A.I.; Zhao, A.B. A Review of Passive RFID Tag Antenna-Based Sensors and Systems for Structural Health Monitoring Applications. Sensors 2017, 17, 265. [CrossRef] 104. Hernandez-Valle, F.; Clough, A.R.; Edwards, R.S. Stress Corrosion Cracking Detection Using Non-Contact Ultrasonic Techniques. Corros. Sci. 2014, 78, 335–342. [CrossRef] 105. Tian, G.Y.; He, Y.; Adewale, I.; Simm, A. Research on Spectral Response of Pulsed Eddy Current and NDE Applications. Sens. Actuators Phys. 2013, 189, 313–320. [CrossRef] 106. Cook, B.S.; Vyas, R.; Kim, S.; Thai, T.; Le, T.; Traille, A.; Aubert, H.; Tentzeris, M.M. RFID-Based Sensors for Zero-Power Autonomous Wireless Sensor Networks. IEEE Sens. J. 2014, 14, 2419–2431. [CrossRef] 107. Bae, S.-H.; Lee, Y.; Sharma, B.K.; Lee, H.-J.; Kim, J.-H.; Ahn, J.-H. Graphene-Based Transparent Strain Sensor. Carbon 2013, 51, 236–242. [CrossRef] 108. Kalansuriya, P.; Bhattacharyya, R.; Sarma, S. RFID Tag Antenna-Based Sensing for Pervasive Surface Crack Detection. IEEE Sens. J. 2013, 13, 1564–1570. [CrossRef] 109. Caizzone, S.; DiGiampaolo, E. Wireless Passive RFID Crack Width Sensor for Structural Health Monitoring. IEEE Sens. J. 2015, 15, 6767–6774. [CrossRef] 110. Marindra, A.M.J.; Tian, G.Y. Chipless RFID Sensor Tag for Metal Crack Detection and Characterization. IEEE Trans. Microw. Theory Tech. 2018, 66, 2452–2462. [CrossRef] 111. Gotoh, Y.; Hirano, H.; Nakano, M.; Fujiwara, K.; Takahashi, N. Electromagnetic Nondestructive Testing of Rust Region in Steel. IEEE Trans. Magn. 2005, 41, 3616–3618. [CrossRef] 112. Available online: https://www.Impinj.Com/Library/Customer-Stories/Nasa-Scientists-Use-Rain-Rfid-to-Monitor-Spac (accessed on 1 March 2021). 113. Hines, J.H.; Solie, L.P.; Tucker, D.Y.G.; Hines, A.T. Individually Identifiable Surface Acoustic Wave Sensors, Tags and Systems. U.S. Patent 9,537,677, 3 January 2017. 114. Qi, C.; Griffin, J.D.; Durgin, G.D. Low-Power and Compact Microwave RFID Reader for Sensing Applications in Space. In Proceed- ings of the 2018 IEEE International Conference on RFID Technology Application (RFID-TA), Macau, China, 26–28 September 2018; pp. 1–6. 115. Bandy, W.R.; Peeters, J.P. Radio Frequency Identification (RFID) Based Sensor Networks. U.S. Patent 749,225,4 B2, 17 August 2006. 116. Buettner, M.; Prasad, R.; Philipose, M.; Wetherall, D. Recognizing Daily Activities with RFID-Based Sensors. In Proceedings of the 11th International Conference on Ubiquitous Computing, New York, NY, USA, 19–20 September 2009; Association for Computing Machinery: New York, NY, USA, 2009; pp. 51–60. 117. Motroni, A.; Buffi, A.; Nepa, P. A Survey on Indoor Vehicle Localization through RFID Technology. IEEE Access 2021, 9, 17921–17942. [CrossRef] 118. Mora-Mora, H.; Gilart-Iglesias, V.; Gil, D.; Sirvent-Llamas, A. A Computational Architecture Based on RFID Sensors for Traceability in Smart Cities. Sensors 2015, 15, 13591–13626. [CrossRef][PubMed] 119. Buffi, A.; Nepa, P.; Lombardini, F. A Phase-Based Technique for Localization of UHF-RFID Tags Moving on a Conveyor Belt: Performance Analysis and Test-Case Measurements. IEEE Sens. J. 2015, 15, 387–396. [CrossRef] 120. Bernardini, F.; Buffi, A.; Fontanelli, D.; Macii, D.; Magnago, V.; Marracci, M.; Motroni, A.; Nepa, P.; Tellini, B. Robot-Based Indoor Positioning of UHF-RFID Tags: The SAR Method With Multiple Trajectories. IEEE Trans. Instrum. Meas. 2021, 70, 1–15. [CrossRef] 121. Magnago, V.; Palopoli, L.; Buffi, A.; Tellini, B.; Motroni, A.; Nepa, P.; Macii, D.; Fontanelli, D. Ranging-Free UHF-RFID Robot Positioning Through Phase Measurements of Passive Tags. IEEE Trans. Instrum. Meas. 2020, 69, 2408–2418. [CrossRef] 122. Buffi, A.; Motroni, A.; Nepa, P.; Tellini, B.; Cioni, R. A SAR-Based Measurement Method for Passive-Tag Positioning with a Flying UHF-RFID Reader. IEEE Trans. Instrum. Meas. 2019, 68, 845–853. [CrossRef] 123. Kennedy, T.F.; Fink, P.W.; Chu, A.W.; Champagne, N.J.; Lin, G.Y.; Khayat, M.A. Body-Worn E-Textile Antennas: The Good, the Low-Mass, and the Conformal. IEEE Trans. Antennas Propag. 2009, 57, 910–918. [CrossRef] 124. Moradi, E.; Koski, K.; Ukkonen, L.; Rahmat-Samii, Y.; Björninen, T.; Sydänheimo, L. Embroidered RFID Tags in Body-Centric Communication. In Proceedings of the 2013 International Workshop on Antenna Technology (iWAT), Karlsruhe, Germany, 4–6 March 2013; pp. 367–370. 125. Occhiuzzi, C.; Cippitelli, S.; Marrocco, G. Modeling, Design and Experimentation of Wearable RFID Sensor Tag. IEEE Trans. Antennas Propag. 2010, 58, 2490–2498. [CrossRef] 126. Occhiuzzi, C.; Contri, G.; Marrocco, G. Design of Implanted RFID Tags for Passive Sensing of Human Body: The STENTag. IEEE Trans. Antennas Propag. 2012, 60, 3146–3154. [CrossRef] 127. White Papers. Available online: https://www.onsemi.com/support/design-resources/white-papers?rpn=SPSXM001FOM (accessed on 9 September 2020). Sensors 2021, 21, 3138 34 of 34

128. Michel, A.; Nepa, P.; Qing, X.; Chen, Z.N. Considering High-Performance Near-Field Reader Antennas: Comparisons of Proposed Antenna Layouts for Ultrahigh-Frequency Near-Field Radio-Frequency Identiﬁcation. IEEE Antennas Propag. Mag. 2018, 60, 14–26. [CrossRef] 129. Catarinucci, L.; De Donno, D.; Colella, R.; Ricciato, F.; Tarricone, L. A Cost-Effective SDR Platform for Performance Characteriza- tion of RFID Tags. IEEE Trans. Instrum. Meas. 2012, 61, 903–911. [CrossRef] 130. Solic, P.; Blazevic, Z.; Skiljo, M.; Patrono, L.; Colella, R.; Rodrigues, J.J. Gen2 RFID as IoT Enabler: Characterization and Performance Improvement. IEEE Wirel. Commun. 2017, 24, 33–39. [CrossRef] 131. Garbati, M.; Perret, E.; Siragusa, R.; Halope, C. Ultrawideband Chipless RFID: Reader Technology From SFCW to IR-UWB. IEEE Microw. Mag. 2019, 20, 74–88. [CrossRef] 132. Boggioni, L.; Monti, L.; Terranova, S.; Costa, F.; Genovesi, S.; Manara, G. Low-Cost Portable Reader for Frequency Domain Chipless Tags: Architecture and Experimental Results on Depolarizing Tags. Electronics 2019, 8, 35. [CrossRef] 133. Koswatta, R.V.; Karmakar, N.C. A Novel Reader Architecture Based on UWB Chirp Signal Interrogation for Multiresonator-Based Chipless RFID Tag Reading. IEEE Trans. Microw. Theory Tech. 2012, 60, 2925–2933. [CrossRef] 134. Garbati, M.; Siragusa, R.; Perret, E.; Halopé, C. Impact of an IR-UWB Reading Approach on Chipless RFID Tag. IEEE Microw. Wirel. Compon. Lett. 2017, 27, 678–680. [CrossRef] 135. Buchanan, N.B.; Fusco, V. Single VCO Chipless RFID Near-Field Reader. Electron. Lett. 2016, 52, 1958–1960. [CrossRef]