arXiv:1910.13041v3 [eess.SP] 23 Mar 2021 hpes ihteepoe htcamt rvd decimeter provide to claim and that GNSS S10, phones dual-frequency these with 20, With G8, equipped Mate being chipsets. LG are Pro, 1 the Z6 Xperia Sony as 4, such [3]. satellites more GNSS of from and capable dual- signals Broadcom’s is More L5/E5 a and and with [5] L1/E1 equipped (BCM47755) receiving is chipset 8 GNSS Mi frequency Xiaomi [4]. chipset, conse- sma 8 GNSS first and dual-frequency Mi world’s a the smartphones with of equipped user’s introduction phone the general in on resulted quently apps s various positioning accurate to highly providing of the possibility the to [3]. due issue mostly security military use availabl and government and cost and limited high professional been to has location accuracy only provi of localization Unfortunately, high higher. number ever of the hig is media, for As accuracy demand few. social positioning significantly, a increases games, apps name navigation, service-based to as service, location in such exploit dining apps worldwide [2] active of 50% information are than More chipsets [1]. 2018 (GNSS) Navigati System Global with mor Satellite equipped report, smartphones mobility billion Ericson 4.8 the than to According smartphone. a phone. GNSS indoors, singe-frequency power a more with 28% and comparison outdoors, 37% in average consumes on chipset power energy GNSS th more dual-frequency the demonstrate a We compares with phone. and smartphone single-frequency a chipset t with GNSS equipp efficiency study To smartphone dual-frequency experimental a phone. a first of with the GNSS consumption study is power dual-frequency this the empirical characterizes knowledge, a an our of performing best the consumption by power gap paper, a this on knowledge is In chipset. this there GNSS bridge hand, dual-frequency smartph a other of with efficiency the energy equipped the On on gap apps. knowledge significant emerging researc positioni navigation of decimeter-level for academia development bro to and accuracy industry up location to excitement high provide unprecedentedly chipse to world’s Such GNSS accuracy. claimed the dual-frequency 8, is a Mi with which introduced equipped has smartphone Xiaomi first brand positionin o Mobile accurate based for higher. built demand are such, apps As more services. and location More smartphone. a on features ryEfficiency ergy oee,awdsra s fsatdvcsi expanding is devices smart of use widespread a However, on features used widely most the of one is service Location Terms Index Abstract Lcto evc soeo h otwdl used widely most the of one is service —Location Da-rqec NS oieCmuig En- Computing, Mobile GNSS, —Dual-frequency ulFeunyGS fSmartphone of GNSS Dual-Frequency hrceiigPwrCnupinof Consumption Power Characterizing eateto optrSine nvriyo epi,TN, Memphis, of University Science, Computer of Department .I I. NTRODUCTION irmKriadMonguWon Myounggyu and Karki Bikram { kri mwon bkarki, sever is g ervice and h ught ones sion ta at her hat we on rt- ng ed n e e - t } @memphis.edu otiuin fti ae r umrzda follows. the as Concretely, summarized are environment. paper indoor this an of in contributions power an outdoors, more chipset counterpa 28% GNSS its single-frequency with a with compared on equipped positions power a updating more with for 37% phone average consumes a chipset that environmen GNSS demonstrate different We dual-frequency efficiency. of energy effects the and the on indoors energy understand both the performed to were out outdoors Experiments ruling phones. the by componen of hardware/ phones other the by incurred of consumption modules performed GNSS powe positions the the updating measured for we exclusively particular, for consumption In vendor same study. the comparative from single-frequency 7, fair Note a and Xiaomi 8, phone, Mi GNSS Xiaomi phone, GNSS frequency qipdwt igefeunyGS hpe.Frthis For chipset. smartphone GNSS selected, a are dual- single-frequency phones with two a study, a comparison of with in first consumption equipped the phone power is GNSS this the knowledge, frequency characterizes our of that best work the dual-frequency To a chipset. with equipped GNSS smartphone a of consumption location high such achieve unexplored. largely to is consumed accuracy is power Howe much [6][7][8][9]. how accuracy positioning high focus achieving phones on performed GNSS dual-frequency been utilizing have effectively on works processing Numerous h achieve accuracy. more to positioning consumption and power receiver rates higher in chipping resulting higher GNSS cy- dual-frequency dual- require consumption, duty chipsets power improved recent the reduce provide with to Although that cling equipped antennas are efficiency. phones low-cost these advanced energy for chipsets the GNSS frequency is phones horizon. the on ap is location-based research in navigation dawn and new a accuracy, positioning level • nti ae,w efr h rteprclsuyo power on study empirical first the perform we paper, this In dual-frequency these for challenges critical the of One • • rqec NScistfrbt noradoutdoor and dual- power indoor a the both with for on environments. equipped chipset study smartphone GNSS empirical frequency a first of the consumption perform We ioiRdiNt qipdwt single-frequency a with chipset. with equipped GNSS 7 compared average Note indoors, on Redmi power power Xiaomi more more 28% and 37% dual-frequency outdoors, consumes a with chipset 8 GNSS Mi Xiaomi that demonstrate We epeetaueu eeec nteeeg fcec of efficiency energy the on reference useful a present We ntdStates United i.e., urnl vial dual- available currently a ver, igh ing by ps ts ts rt d r , a dual-frequency GNSS phone to facilitate research on delay corrections. Receivers use the almanac data to search mobile computing and navigation that exploits those dual- satellites. Especially, receivers that support only a single GNSS frequency GNSS phones to achieve higher positioning frequency use the ionospheric delay data to correct the delay. accuracy. To transmit the whole almanac data, 25 navigation messages This paper is organized as follows. In Section II we review are needed, and it takes about 12.5 minutes to complete the the background on GNSS concentrating on dual-frequency transmission. In contrast to the almanac data, the ephemeris GNSS. We then explain the experimental settings and the data contain highly precise orbital parameters of satellites and methodology for measuring power consumption of both dual clock correction information. The ephemeris data is used to and single-frequency GNSS phones in Section III. The results calculate the positions of satellites precisely. Each satellite are analyzed in Section IV, and we conclude in Section V. broadcasts its own ephemeris data every 30 seconds. A smartphone is equipped with a GNSS/navigation chipset II. BACKGROUND to receive GNSS signals. It is kind of a blackbox that produces A. Global Navigation Satellite System (GNSS) the user position, velocity and time (PVT) as well as infor- mation about tracked satellites. Fig. 2 shows a block diagram GNSS is a system of satellites that provides time and of a typical GNSS receiver. The GNSS antenna is used to location information anywhere on or near the Earth when capture GNSS signals in L band. The RF front-end takes the an unblocked line of sight to four or more GNSS satellites RF signals as input from the antenna and performs down- is available [10]. There are two major GNSS systems that conversion to reduce the cost. And then, the analog to digital cover the entire world. Global Positioning System (GPS) is the converter (ADC) digitizes the signal. The baseband processing most widely used system developed by US. It has at least 24 module performs several signal processing tasks to acquire and GNSS satellites. Globalnaya Navigatsionnaya Sputnikovaya track the signals. The acquisition task determines satellites that Sistema (GLONASS) is a navigation system developed by are in view and can be tracked. The tracking stage is used Russian consisting of 31 GNSS satellites. There are two other to update dynamically the code delay and carrier frequency systems with global coverage that are under development: of the signal in order to track the signal correctly. The PVT BeiDou and Galileo. Beidou is a Chinese navigation system processing block combines the information from the baseband that has 22 satellites. While global coverage is not provided processing block to derive a solution (e.g., PVT). yet, it is already used in Asia-Pacific region. Galileo is the

European navigation system consisting of 18 satellites. Full Antenna global coverage by Galileo is expected in 2020. Front End Raw Data and Navigation Message RF 0010 RF Analog to Baseband Pre-Amplifier Digital PVT GNSS satellites transmit radio signals over two or more Front-end Processing Processing Filtering Converter frequencies in L band, i.e., the operating frequency range of

1-2 GHz. Fig. 1 shows the frequencies used by different GNSS Input/Output systems. Radio signals transmitted from GNSS satellites carry Local ranging codes and navigation data which are used to calculate Oscillator User Interface the coordinates of satellites and the distance between a satellite and a receiver. The binary phase shift keying (BPSK) [11] is Fig. 2. Block diagram of a typical GNSS receiver. used to modulate these signals. B. Dual Frequency GNSS GPS GLONASS Beidou Galileo E1 Dual-frequency GNSS receives two different radio signals L5 B2I L1 C/A at different frequencies from each satellite to provide more B2a L2C E6 L3 accurate positioning. Most of currently available devices uti- E5a E5b L2 B3 B1I L1 lize a single narrow band (L1/E1) with low sampling rates. Recently, the mass market introduce products that support dual wide band (upper L band and partial lower L band) 1237MHz 1254 MHz 1610 MHz 1593 MHz

1176.45 MHz with high sampling rates. However, due to the high cost, 1268.52 MHz 1227.60 MHz 1278.75 MHz 1207.140 MHz 1575.42 MHz 1561.098 MHz use of these dual-frequency GNSS devices has been limited Fig. 1. GNSS frequencies for different navigation systems. to professional and governmental users. In 2017, Broadcom introduce the first low-cost dual-frequency GNSS chipset, A GNSS navigation message conveys various information BCM47755, specifically designed for smartphones [5]. In such as the position and velocity of satellites, clock, satellite 2018, u-blox launch their dual-frequency GNSS chipset, F9 orbit shape, etc. The navigation messages are transmitted at chip [13], and STMicroelectronics introduce the Teseo receiver a slower rate than the ranging codes. For example, receiving that supports L1/L2 or L1/L5 frequencies [14]. [15] and a whole navigation message takes 30 seconds for GPS [12]. Qualcomm [16] also start production of their dual-frequency The message consists of two types of data: Almanac and GNSS chipsets in 2018. Ephemeris. The almanac data contain the coarse orbital pa- With the growth of the mass market for dual-frequency rameters of all satellites and information about ionospheric GNSS chipsets, the first smartphone, , equipped with a dual-frequency GNSS chipset, Broadcom’s BCM47755, III. SYSTEM SETUP is introduced in 2018. This smartphone supports two frequen- A. System Settings cies (L1+L5) and is capable of tracking and processing GPS L1 C/A, GPS L5, GLONASS L1, Galileo E5a and QZSS L5, We select a currently available dual-frequency GNSS phone, Galileo (GAL) E1, BeiDou (BDS) B1, GLONASS L1, and Xiaomi Mi 8, and its single-frequency counterpart, Xiaomi QZSS L1 signals. Redmi Note 7 for this experimental study. Xiaomi Mi 8 is These smartphones equipped with a dual-frequency GNSS equipped with a dual-frequency GNSS chipset, Broadcom chipset enjoy significant advantages. While enhanced posi- BCM47755, and Xiaomi Redmi Note 7 has a single-frequency tioning accuracy by directly estimating the ionosphere delay GNSS integrated into its 660 proces- is the most significant benefit, the dual-frequency GNSS sor. Both phones have similar hardware specs as they are improves robustness against jamming and provides access to from the same vendor. They are equipped with Snapdragon advanced satellite navigation technologies such as PPP [3] and processors, 16M color capacitive touchscreens, Adreno GPUs, RTK [17], which are currently available for only specialized and a similar set of sensors. We installed the same OS, receivers. 9 (API level 28) on both phones. To cut off the effect of Assisted GPS [20] (i.e., a technique that utilizes More precisely, errors from different sources influence the the information from cell towers for faster position update), positioning accuracy such as imprecise information received the sim cards of both phones were removed and Wi-Fi was from a satellite (e.g., on-board clock, ephemeris, etc.), at- disabled. An app was created that requests for a position mosphere effects (e.g., ionosphere and troposphere), receiver update every second. All other apps including background noise, and multipath effect [18]. Among these various sources processes were all disabled, and the brightness level of screen of errors, the ionosphere causes the biggest delay [2]. The was kept to minimum. ionosphere is a layer of the Earth’s atmosphere extending from 60km up to 2,000km that contains a high concentration of free electrons and ions that can reflect radio waves. The impact of the layer is measured based on Total Electron Content (TEC) Data Logger which is defined as the number of electrons in a tube of a 1m2 cross section between two points, i.e., a receiver and a satellite. Monsoon Power Monitor

Thus, the contribution of the ionosphere can be written based Xiaomi Mi 8 on TEC as the following.

40.3 · T EC I = . (1) p f 2

Here f is the carrier frequency, and 40.3 is the TEC parameter Fig. 3. Experimental setup. which depends on the location of receiver, the intensity of solar activity, and the hour of day. A dual-frequency GNSS chipset We used the Monsoon power monitor to measure power can eliminate the impact of the ionosphere based on a pseu- consumption [21]. We deployed the system both indoors and dorange ρ1 calculated on frequency f1 and a pseudorange ρ2 outdoors. Fig. 3 displays the system settings for outdoor calculated on frequency f2 as follows. Details on calculating deployment. The probes of the power monitor were connected a pseudorange can be found in [19]. to the battery terminals of smartphone, providing current to the phone. A laptop was connected to the Monsoon power 2 2 ∗ f1 ρ1 − f2 ρ2 monitor through USB to measure the current drawn and the ρ = 2 2 . (2) f1 − f2 voltage at a rate of 5KHz in real time. We also confirm that both phones receive signals from a Here ρ∗ the pseudorange without the ionosphere effect. sufficient number of visible satellites. Figs. 4 and 5 display the There are challenges, however, to achieve enhanced accu- number of visible satellites over time after the app is started racy. A dual-frequency GNSS impacts the design of the re- outdoors and indoors, respectively. It shows that the number of ceiver, i.e., the antenna, RF front-end, and baseband processing visible satellites quickly increases when the app is started, and blocks should be replicated to handle the additional frequency, then the phones see about 18-22 satellites outdoors and about leading to higher power consumption. Specifically, improved 14-18 satellites indoors. We observe that a sufficient number positioning accuracy is available with antennas with improved of satellites were available in both environments, although duty cycling for reducing power consumption. Utilization of the number of visible satellites was smaller in the indoor dual bands requires higher chipping rates and more processing environment. power resulting in degraded energy efficiency. This paper aims to characterize such extra power consumption for a B. Methodology dual-frequency GNSS phone by comparing that of a single- In this section, we present details on measuring the power frequency GNSS phone. consumption of the phones for updating positions. After GPS 25 5000

20 4000

15 3000

10 2000

Power (mW) App Start TTFF Number of Satellites 5 1000 Mi 8 Redmi Note 7 Position Update Baseline 0 0 0 100 200 300 400 500 0 50 100 150 200 Time (sec) Time (sec)

Fig. 4. The number of visible satellites outdoors. Fig. 6. Power consumption of Xiaomi Mi 8.

20 2500

2000 15

1500

10 1000

Power (mW) App Start

Mi 8 TTFF

Number of Satellites 5 Redmi Note 7 500 Position Update Baseline

0 0 0 100 200 300 400 500 0 50 100 150 200 Time (sec) Time (sec)

Fig. 5. The number of visible satellites indoors. Fig. 7. Power consumption of Xiaomi Redmi Note 7. is switched on, it takes some time to complete the first position service. fix. This is called the time to first fix (TTFF) [22]. There are three different scenarios for TTFF. If GPS has been turned off 1400 for a long time and/or has moved a long distance, GPS does not have the almanac, ephemeris, time and position information.In 1200 this case, which is called the cold start, TTFF can be very large 1000 which may take several minutes. When only the ephemeris 800 data is not available, which is called the warm start, TTFF can be significantly reduced as short as 30 seconds. If all the 600 Power (mW) data are available, which is called the hot start, TTFF becomes 400 minimal taking only 0.5 to 20 seconds. Figs. 6 and 7 show the power consumption of Mi 8 and 200 Redmi Note 7. Both phones consume power independent of 0 0 50 100 150 200 the GPS activity before the app is started. This is the baseline Time (sec) power consumption. Once the app is started, consumed power quickly increases for a short moment to load the app. And Fig. 8. Background power consumption of Xiaomi Redmi Note 7. then, both phones consume relatively higher energy for TTFF compared to regular position update. The results show that A challenge is to measure only the consumed power TTFF for both phones were different. In fact, it is known that for updating positions, excluding other sources of power different GNSS chipsets have varying TTFF. Once the first consumption such as background kernel processes, sensors, position is fixed, both phones use power to update position. In network modules, and screen. A tricky part is that these this experiment, we focus on measuring power consumption phones consume different amounts of power for these non- for this regular position update, which accounts for the major GPS activities. Fortunately, we found that the baseline power part of power consumption for many apps based on location consumption of the two phones was relatively stable when we disable all background processes, disconnect network 1800 services such as Wi-Fi, and minimize the screen brightness 1600 to minimum as shown in Fig. 8. Given the stable baseline Mi 8 power consumption, we simply subtract the baseline power 1400 Redmi Note 7 consumption from measured power consumption and obtain 1200 the “pure” power consumption used for updating positions. 1000 More precisely, we set a one second interval within which 800 a position update is performed, measure accumulated power Power (mW) consumption during this period, and then subtract it by the 600 base line power consumption, obtaining consumed power for 400 a single position update. 200 0 0.25 0.5 0.75 1 IV. RESULTS Time (sec)

A. Outdoor Experiments Fig. 11. Comparison of the power consumption of Xaiomi Mi 8 and Xiaomi Redmi Note 7 for a one second interval.

2000 1

0.8 1500

0.6

1000 CDF Power (mW) 0.4

Mi8 0.2 Redmi Note 7 500

0 2 4 6 8 10 0 Time (sec) 100 200 300 400 500 Consumed Energy (mJ) Fig. 9. Power consumption of Xiaomi Mi 8 for position update. Fig. 12. Cumulative distribution function graph of the consumed energy of Xiaomi Mi 8 and Xiaomi Redmi Note 7 for updating positions.

1400 We then calculate the “pure” power consumption for posi- 1200 tion update by subtracting with the baseline power consump- tion. We repeat a 5 minute measurement 5 times for each 1000 smartphone. Fig. 12 displays the cumulative distribution func-

800 tion (CDF) plots of power consumption for position update of both smartphones. The results demonstrate that the average

Power (mW) 600 power consumption for Mi 8 and Redmi Note 7 was 318mJ (± 32mJ) and 232mJ (± 20mJ), respectively, indicating that 400 Mi 8 with the dual-frequency GNSS chipset consumes 37% 200 more power for position update in comparison with the single- 0 2 4 6 8 10 frequency GNSS of Redmi Note 7. Considering that a typical Time (sec) phone battery has about 29,000 joules, without considering Fig. 10. Power consumption of Xiaomi Redmi Note 7 for position update. power consumption from any other hardware/software com- ponents, only the GPS for position update will deplete the Figs. 9 and 10 display the power consumption of Mi 8 and battery after about 25 hours and 35 hours for Mi 8 and Redmi Redmi Note 7 for updating positions, respectively. As shown, Note 7, respectively. It is interesting to note that a single- a peak is observed every second a request for a position update frequency GNSS phone would last 10 hours longer only due is sent to the phones. The results also show that Mi 8 with to the difference in the GNSS chipset. a dual-frequency GNSS chipset consumes more power than Redmi Note 7 with a single-frequency GNSS chipset. Fig. 11 B. Indoor Experiments more clearly shows the difference in power consumption as We measure the power consumption of Mi 8 and Redmi we align the graphs exactly with the time when a request for Note 7 used for updating positions in an indoor environment, position update is sent. i.e., inside an apartment. Fig. 13 shows the results. As shown, with a single-frequency counterpart from the same vendor. We 350 Outdoor demonstrated that the dual-frequency phone consumed 37% 300 Indoor more power on average for position update compared with the single-frequency phone outdoors, and 28% indoors. We expect 250 that the results will be a useful reference for academia and 200 industry in developing mobile applications exploiting location

150 service based on dual-frequency GNSS. REFERENCES 100

Consumed Energy (mJ) [1] Ericsson, “Ericsson Mobility Report,” http://www.ericsson.com/ 50 ericsson/, accessed: 2019-09-01. [2] E. Kaplan and C. Hegarty, Understanding GPS: principles and applica- 0 tions. Artech house, 2005. Mi8 Redmi Note7 [3] U. Robustelli, V. Baiocchi, and G. Pugliano, “Assessment of dual frequency GNSS observations from a Xiaomi Mi 8 Android smartphone and positioning performance analysis,” Electronics, vol. 8, no. 1, p. 91, Fig. 13. Power consumption of Xiaomi Mi 8 and Xiaomi Redmi Note 7 in 2019. indoor and outdoor environments. [4] Xiaomi, “Xiaomi Mi 8,” https://www.mi.com/global/mi8, 2018, ac- cessed: 2019-9-1. [5] Broadcom, “BCM47755,” https://www.broadcom.com/products/ 1 wireless/gnss-gps-socs/bcm47755, 2018, accessed: 2019-9-1. Redmi Note 7 (indoor) [6] Q. Wu, M. Sun, C. Zhou, and P. Zhang, “Precise point positioning using 0.8 dual-frequency gnss observations on smartphone,” Sensors, vol. 19, no. 9, p. 2189, 2019. Redmi Note 7 [7] A. Elmezayen and A. El-Rabbany, “Precise point positioning using (outdoor) 0.6 world’s first dual-frequency gps/galileo smartphone,” Sensors, vol. 19, Mi 8 (indoor) no. 11, p. 2593, 2019.

CDF [8] B. Chen, C. Gao, Y. Liu, and P. Sun, “Real-time precise point positioning 0.4 Mi 8 (outdoor) with a xiaomi mi 8 android smartphone,” Sensors, vol. 19, no. 12, p. 2835, 2019. 0.2 [9] Z. Niu, P. Nie, L. Tao, J. Sun, and B. Zhu, “Rtk with the assistance of an imu-based pedestrian navigation algorithm for smartphones,” Sensors, vol. 19, no. 14, p. 3228, 2019. 0 [10] B. Hofmann-Wellenhof, H. Lichtenegger, and E. Wasle, GNSS–global 0 10 20 30 40 50 navigation satellite systems: GPS, GLONASS, Galileo, and more. Carrier-to-noise-density ratio (dBHz) Springer Science & Business Media, 2007. [11] F. Gardner, “A BPSK/QPSK timing-error detector for sampled re- Fig. 14. Signal strength of GNSS signals (i.e., carrier-to-noise-density) for ceivers,” IEEE Transactions on Communications, vol. 34, no. 5, pp. Mi 8 and Redmi Note 7 for indoor and outdoor environments. 423–429, 1986. [12] D. Wells, N. Beck, A. Kleusberg, E. J. Krakiwsky, G. Lachapelle, R. B. Langley, K.-p. Schwarz, J. M. Tranquilla, P. Vanicek, and the average power consumption of Mi 8 indoors and outdoors D. Delikaraoglou, “Guide to gps positioning,” in Canadian GPS Assoc, 1987. was 321mJ (± 28) and 318mJ (± 32), respectively. The [13] U-blox, “F9,” https://www.u-blox.com/en/publication/beyond-stories/u- average power consumption for Redmi Note 7 for indoors and blox-f9-taking-gnss-precision-next-level, 2018, accessed: 2019-9-1. outdoors was 251mJ (± 23) and 232mJ (± 20), respectively. [14] STM, “Teseo,” https://www.broadcom.com/products/wireless/gnss-gps- socs/bcm47755, 2018, accessed: 2019-9-1. The results indicate that more power is consumed in an [15] Intel, “GNSS2x00,” https://www.intel.com/content/dam/www/public/ indoor environment for both smartphones. Specifically, Mi 8 us/en/documents/product-briefs/wireless-gnss-2x00-brief.pdf, 2018, ac- consumed 1.2% more power, and Redmi Note 7 consumed 8% cessed: 2019-9-1. [16] Qualcomm, “Snapdragon 855,” https://www.qualcomm.com/products/ more power in an indoor environment. Overall, in an indoor snapdragon-855-mobile-platform, 2018, accessed: 2019-9-1. environment, Mi 8 consumed 28% more power on average [17] P. Dabove and V. Di Pietra, “Single-baseline rtk positioning using dual- compared with Redmi Note 7. The reason for the higher power frequency gnss receivers inside smartphones,” Sensors, vol. 19, no. 19, p. 4302, 2019. consumption in an indoor environment can be attributed to [18] J. Sanz Subirana, J. Juan Zornoza, and M. Hern´andez-Pajares, “GNSS smaller signal strength of received GNSS signals in an indoor data processing. Vol. i: Fundamentals and algorithms,” Noordwijk: environment [23] as shown in Fig. 14. The average signal European Space Agency Communications, 2013. [19] P. D. Groves, Principles of GNSS, inertial, and multisensor integrated strength for Mi 8 in an indoor environment decreased by 24%, navigation systems. Artech house, 2013. and that for Redmi Note 7 decreased by 10.5%. Interestingly, [20] F. S. T. Van Diggelen, A-gps: Assisted gps, gnss, and sbas. Artech it was found that the dual-frequency GNSS chipset of Mi 8 House, 2009. [21] Monsoon, “Power Monitor,” https://www.msoon.com/, 2018, accessed: was less susceptible to weak signal strength compared with 2019-9-1. the single-frequency GNSS chipset of Redmi Note 7. [22] M. Paonni, M. Anghileri, S. Wallner, J.-A. Avila-Rodriguez, and B. Eiss- feller, “Performance assessment of gnss signals in terms of time to first fix for cold, warm and hot start,” Proc. of the ION ITM, pp. 25–27, V. CONCLUSION 2010. We presented the first empirical study on the power con- [23] A. T. Lo’ai, A. Basalamah, R. Mehmood, and H. Tawalbeh, “Greener and smarter phones for future cities: Characterizing the impact of gps sumption of a dual-frequency GNSS smartphone. The mea- signal strength on power consumption,” IEEE Access, vol. 4, pp. 858– sured power for a dual-frequency GNSS phone was compared 868, 2016.