<<

Frequencies for Communication (1)

twisted coax cable optical transmission pair

1 Mm 10 km 100 m 1 m 10 mm 100 µm 1 µm 300 Hz 30 kHz 3 MHz 300 MHz 30 GHz 3 THz 300 THz 2. Wireless Transmission VLF LF MF HF VHF UHF SHF EHF infraredvisible light UV

‰ VLF = Very Low Frequency UHF = Ultra High Frequency Frequencies and Signals ‰ LF = Low Frequency SHF = Super High Frequency ‰ MF = Medium Frequency EHF = Extra High Frequency and ‰ HF = High Frequency UV = Ultraviolet Light ‰ VHF = Very High Frequency

‰ Frequency and wave length: λ = c/f wave length λ, speed of light c ≅ 3x108m/s, frequency f

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 1 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 2

Frequencies for Mobile Communication (2) Frequencies and Regulations

‰ VHF-/UHF-ranges for mobile radio ‰ ITU-R holds auctions for new frequencies and manages – Simple, small antenna for cars frequency bands worldwide (WRC, World Radio Conferences) – Deterministic propagation characteristics, reliable connections Europe USA Japan Cellular GSM 450-457, 479- AMPS, TDMA, CDMA PDC Phones 486/460-467,489- 824-849, 810-826, 496, 890-915/935- 869-894 940-956, ‰ SHF and higher for directed radio links, satellite communication 960, TDMA, CDMA, GSM 1429-1465, 1710-1785/1805- 1850-1910, 1477-1513 – Small antenna, focusing 1880 1930-1990 – Large bandwidth available UMTS (FDD) 1920- 1980, 2110-2190 UMTS (TDD) 1900- 1920, 2020-2025 ‰ Wireless LANs use frequencies in UHF to SHF spectrum Cordless CT1+ 885-887, 930- PACS 1850-1910, 1930- PHS Phones 932 1990 1895-1918 – Some systems planned up to EHF CT2 PACS-UB 1910-1930 JCT 864-868 254-380 – Limitations due to absorption by water and oxygen molecules (resonance DECT 1880-1900 frequencies) Wireless IEEE 802.11 902-928 IEEE 802.11 LANs 2400-2483 IEEE 802.11 2471-2497 • Weather dependent fading, signal loss caused by heavy rainfall etc. HIPERLAN 2 2400-2483 5150-5250 5150-5350, 5470- 5150-5350, 5725-5825 5725 Others R F-C ontrol RF-Control RF-Control 27, 128, 418, 433, 315, 915 426, 868 868

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 3 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 4

Signals (1) Signals (2)

‰ Physical representation of data ‰ Different representations of signals: ‰ Function of time and location – Amplitude (amplitude domain) – Frequency spectrum (frequency domain) – Phase state diagram (amplitude M and phase ϕ in polar coordinates) ‰ Signal parameters: ϕ – Parameters representing the value of data A [V] A [V] Q = M sin ‰ Classification: – Continuous time/discrete time t[s] ϕ – Continuous values/discrete values I= M cos ϕ – Analog signal = continuous time and continuous values ϕ f [Hz] – Digital signal = discrete time and discrete values ‰ Composed signals transferred into frequency domain using ‰ Signal parameters of periodic signals: Fourier transformation – Period T, frequency f=1/T, amplitude A, and phase shift ϕ – Sine wave as special periodic signal for a carrier: ‰ Digital signals need: π ϕ – Infinite frequencies for perfect transmission s(t) = At sin(2 ft t + t) – Modulation with a carrier frequency for transmission (analog signal!)

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 5 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 6

1 Fourier Representation of Periodic Signals Antennas: Isotropic Radiator

‰ Radiation and reception of electromagnetic waves: – Coupling of wires to space for radio transmission ∞ ∞ = 1 + π + π ‰ Isotropic radiator: g(t) c ∑an sin(2 nft) ∑bn cos(2 nft) 2 n=1 n=1 – Equal radiation in all directions (three dimensional) – Only a theoretical reference antenna ‰ Real antennas always have directive effects: 1 1 – Vertically and/or horizontally ‰ Radiation pattern: – Measurement of radiation around an antenna 0 0 tt z y z Ideal periodic signal Real composition yxIdeal isotropic radiator (based on harmonics) x

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 7 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 8

Signal Propagation Ranges Signal Propagation

‰ Transmission range: ‰ Propagation in free space always like light (straight line) – Communication possible ‰ Receiving power proportional to 1/d² – Low error rate – d = distance between sender and receiver ‰ Receiving power additionally influenced by: ‰ Detection range: – Fading (frequency dependent) – Detection of the signal – Shadowing (Abschattung) possible Sender – Reflection (Reflexion) at large obstacles – No communication – Refraction (Brechung) depending on the density of a medium possible Transmission – Scattering (Streuung) at small obstacles Distance – Diffraction (Beugung) at edges ‰ Interference range: Detection – Signal may not be Interference detected – Signal adds to the background noise shadowing reflection refraction scattering diffraction

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 9 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 10

Real World Examples Multi-path Propagation

‰ Signal can take many different paths between sender and receiver due to: – Reflection, scattering, and diffraction LOS pulses Multi-path pulses

Signal at sender Signal at receiver

‰ Time dispersion: signal is dispersed over time Î Interference with “neighbor” symbols, Inter Symbol Interference (ISI) ‰ The signal reaches a receiver directly and phase shifted Î Distorted signal depending on the phases of the different parts

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 11 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 12

2 Effects of Mobility Multiplexing

‰ Channel characteristics change over time and location: Channels ki – Signal paths change permanently ‰ Multiplexing in 4 dimensions: k k k k k k – Different delay variations of different signal parts 1 2 3 4 5 6 –Space (si) – Different phases of signal parts –Time (t) c Î Quick changes in the power received (short term fading) – Frequency (f) t c – Code (c) t s Power Long term 1 f ‰ Additional changes in: fading ‰ Goal: s 2 f – Distance to sender – Multiple use of a shared medium c – Obstacles further away t Î Slow changes in the average power received (long term fading) t ‰ Important: guard spaces needed! Short term fading s 3 f

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 13 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 14

Frequency Multiplexing Time Multiplexing

‰ Separation of the entire spectrum into smaller frequency bands ‰ A channel gets the whole spectrum for a certain amount of time ‰ A channel gets a certain band of the spectrum for the full time ‰ Advantages: ‰ Advantages: – Only one carrier in the – No dynamic coordination k k k k k k medium at any time 1 2 3 4 5 6 k1 k2 k3 k4 k5 k6 necessary – Throughput high even c – Works also for analog signals for many users f c f ‰ Drawbacks: ‰ Drawbacks: – Waste of bandwidth – Precise if the traffic is synchronization distributed unevenly necessary – Common – Inflexible t –Guard spaces t clock

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 15 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 16

Time and Frequency Multiplexing Code Multiplexing

‰ Combination of both methods ‰ Each channel has a unique code k k k k k k ‰ A channel gets a certain frequency band for a certain 1 2 3 4 5 6 amount of time ‰ All channels use the same spectrum –Example: GSM at the same time c

k1 k2 k3 k4 k5 k6 ‰ Advantages: ‰ Advantages: – Bandwidth efficient – Better protection against tapping c – No coordination and synchronization necessary – Protection against frequency f – Good protection against interference and selective interference tapping f – Higher data rates compared ‰ Drawbacks: to code multiplex – Lower user data rates – More complex signal regeneration ‰ But: t ‰ Implemented using spread spectrum t – Precise coordination required technology

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 17 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 18

3 Modulation Modulation and

‰ Digital modulation: Analog – Digital data is translated into an analog signal (baseband) Base-band – ASK, FSK, PSK - main focus in this chapter Digital Data Signal – Differences in spectral efficiency, power efficiency, robustness Digital Analog ‰ Analog modulation: 101101001 Modulation Modulation Radio Transmitter – Shifts center frequency of baseband signal up to the radio carrier Radio ‰ Motivation: Carrier – Smaller antennas (e.g., λ/4) – Frequency Division Multiplexing Analog – Medium characteristics Base-band Digital Signal Data ‰ Basic schemes: Analog Synchronization – (AM) Demodulation Decision 101101001 Radio Receiver – (FM) radio – (PM) carrier

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 19 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 20

Digital Modulation Advanced Frequency Shift Keying

‰ Modulation of digital signals known ‰ Bandwidth needed for FSK depends on the distance between 101 as Shift Keying the carrier frequencies ‰ Special pre-computation avoids sudden phase shifts: ‰ Amplitude Shift Keying (ASK): t Î MSK (Minimum Shift Keying) – Very simple ‰ Bit-separated into even and odd bits, the duration of each bit is – Low bandwidth requirements 101 doubled – Very susceptible to interference ‰ Depending on the bit values (even, odd) the higher or lower t frequency, original or inverted is chosen ‰ Frequency Shift Keying (FSK): – Needs larger bandwidth ‰ The frequency of one carrier is twice the frequency of the other 101 ‰ Equivalent to offset QPSK ‰ Even higher bandwidth efficiency using a Gaussian low-pass

‰ Phase Shift Keying (PSK): t filter: –More complex Î GMSK (Gaussian MSK), used in GSM – Robust against interference

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 21 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 22

Advanced Phase Shift Keying Quadrature Amplitude Modulation

‰ BPSK (Binary Phase Shift Keying): Q ‰ Quadrature Amplitude Modulation (QAM): QPSK – Bit value 0: sine wave (0° shift) – Combines amplitude and phase modulation 10 11 – Bit value 1: inverted sine wave (180° shift) I Q ‰ It is possible to code n bits using one symbol: 1 0 – Very simple PSK –2n discrete levels, n=2 identical to QPSK – Low spectral efficiency I – Bit error rate increases with n, but less errors compared to comparable PSK –Robust,e.g., used in satellite systems BPSK schemes 00 01 ‰ QPSK (Quadrature Phase Shift Keying): A – 2 bits coded as one symbol Q 0010 Example: 16-QAM (4 bits = 1 symbol) 000 – Symbol determines shift of sine wave Symbols 0011 and 0001 have the same phase, 0011 1 – Needs less bandwidth compared to BPSK t 0000 but different amplitude. 0000 and 1000 have – More complex (regular synchronization, pilot signal) I different phase, but same amplitude. 11 10 00 01 1000 Î used in standard 9600 bit/s ‰ Often also transmission of relative, not absolute phase shift: DQPSK - Differential QPSK (American IS-136, Japanese PHS)

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 23 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 24

4 Hierarchical Modulation Spread Spectrum Technology

‰ DVB-T (Digital Video Broadcast Terrestrial) modulates two ‰ Problem of radio transmission: separate data streams onto a single DVB-T stream – Frequency-dependent fading can wipe out narrow band signals for duration of the interference ‰ High Priority (HP) embedded within a Low Priority (LP) stream ‰ Solution: – Spread the narrow band signal into a broad band signal using a special code ‰ Multi-carrier system, about 2000 or 8000 carriers Interference ‰ QPSK (2 first bits), 16 QAM, 64 QAM Power Power Signal Q Spread ‰ Example: 64 QAM Signal Spread – Good reception: resolve the entire Detection at Interference Receiver 64 QAM constellation f f 10 – Poor reception, mobile reception: Protection against narrow-band interference I resolve only QPSK portion ‰ Side effects: – 6 bit per QAM symbol, 2 most – Coexistence of several signals without dynamic coordination significant determine QPSK 00 – Tap-proof – HP service coded in QPSK (2 bit), LP uses remaining 4 bit 000010 010101 ‰ Alternatives: Direct Sequence, Frequency Hopping

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 25 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 26

Spreading and Frequency Selective Fading DSSS (Direct Sequence Spread Spectrum) (1)

Channel Narrow-band Channels ‰ XOR of the signal with pseudo-random number (chipping Quality sequence): 1. 6 signals in FDM – Many chips per bit (e.g., 128) result in higher bandwidth of the signal 2 2. Quality changes over time 1 5 6 t 3 3. Spot view b 4 ‰ Advantages: f 4. 1, 2, 5, 6 ok, 3, 4 not ok User Data 5. Apply SS onto 6 signals! – Reduces frequency selective 01XOR Narrow-band Signal Guard Space fading t – In cellular networks: c Chipping Channel • Base stations can use the Quality Spread Spectrum Channels Sequence same frequency range 2 011010 1 011010 1 = 2 2 1. 6 signals utilize full f • Several base stations can 2 Resulting 2 2. No f planning required detect and recover the signal 1 Signal 3. Separation by codes per signals • Soft hand-over 011010 1 100101 0 f 4. CDM, unique per signal ‰ Drawbacks: t : Bit Period Spread 5. “Extension” of f b – Precise power control necessary t : Chip Period Spectrum c – Spreading factor s = tb/tc -> s*w bandwidth requirements © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 27 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 28

DSSS (Direct Sequence Spread Spectrum) (2) FHSS (Frequency Hopping Spread Spectrum) (1)

Spread ‰ Discrete changes of carrier frequency: Transmit Spectrum Example: – Available bandwidth of small channels separated as in FDM User Data Signal Signal IEEE 802.11 X Modulator – Sequence of frequency changes determined via pseudo random number (TDM) Chip: 10110111000 Chipping Radio (Barker Code) ‰ Two versions: 1 MHz user data – Fast Hopping: Sequence Cs Carrier fr generate 11 MHz several frequencies per user bit Transmitter signal in 2.4 GHz – Slow Hopping: frequency several user bits per frequency Low-pass Correlator ‰ Advantages: Received Filtered Sampled – Frequency selective fading and interference limited to short period Signal Products Signal Sums Data – Simple implementation Demodulator X Integrator Decision – Uses only small portion of spectrum at any time (dwell time: Verweildauer) ‰ Drawbacks: Radio Chipping – Not as robust as DSSS Carrier fr Sequence C Receiver s – Simpler to detect (no knowledge of a chipping sequence as in DSSS required!)

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 29 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 30

5 FHSS (Frequency Hopping Spread Spectrum) (2) Cell Structure

‰ Implements space division multiplex: t b – Base station covers a certain transmission area (cell) User Data ‰ Mobile stations communicate only via the base station 01 011 t ‰ Advantages of cell structures: f td – Higher capacity, higher number of users f3 Slow – Less transmission power needed f2 Hopping – More robust, decentralized f1 (3 bits/hop), – Base station deals with interference, transmission area etc. locally e.g., GSM f td t ‰ Problems: – Fixed network needed for the base stations f3 Fast – Hand-over (changing from one cell to another) necessary f2 Hopping – Interference with other cells f1 (3 hops/bit), e.g., Bluetooth, ‰ Cell sizes from some 100 m in cities to, e.g., 35 km on the t 1600 f changes/s country side (GSM): t : Bit Period t : Dwell Time into 79 carrier f b d – Even less for higher frequencies

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 31 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 32

Frequency Planning (1) Frequency Planning (2)

‰ Frequency reuse only with a certain distance between the base f f f stations 3 3 3 f2 f2 f1 f1 f1 f3 f3 f 3 Cell Cluster f2 f3 f7 3 f2 f2 f2 f f f5 f2 5 2 f1 f1 f f f f4 f6 f5 ‰ Standard model using 7 frequencies: 4 6 5 f3 f3 f3 f1 f4 f1 f4 f3 f7 f1 f3 f7 f1 f2 f2 f3 f6 f5 f2

‰ Fixed frequency assignment: 7 Cell Cluster – Certain frequencies are assigned to a certain cell – Problem: different traffic load in different cells f f f2 f 2 f 2 f 1 f h 1 f 1 f ‰ Dynamic frequency assignment: 3 2 3 h2 3 h1 h1 3 Cell Cluster g2 h3 g2 h3 g2 – Base station chooses frequencies depending on the frequencies already used in g1 g1 g1 g3 g3 g with 3 Sector Antennas neighbor cells 3 – More capacity in cells with more traffic – Assignment can also be based on interference measurements

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 33 © 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 34

Cell Breathing

‰ CDM (Code Division Multiplexing) systems: – Cell size depends on current load • Low load: larger areas • High load: smaller areas

‰ Reason: Additional traffic appears as noise to other users: – Increasing noise due to increasing number of participants – If the noise level is too high users, located at cell boundaries, drop out of cells

“Noisy area”

© 2005 Burkhard Stiller and Jochen Schiller FU Berlin M2 – 35

6