Spread Spectrum

Introduction to Spread SpectrumProblems such as capacity limits, propagation effects, synchronization occur with wireless systemsSpread spectrum modulation spreads out the modulated signal bandwidth so it is much greater than the message bandwidthIndependent code spreads signal at transmitter and despreads signal at receiver

MultiplexingMultiplexing in 4 dimensionsspace (si)time (t)frequency (f)code (c)

Goal: multiple use of a shared medium

Important: guard spaces needed!s2s3s1ftck2k3k4k5k6k1ftcftcchannels ki

Frequency multiplexSeparation of spectrum into smaller frequency bandsChannel gets band of the spectrum for the whole timeAdvantages:no dynamic coordination neededworks also for analog signalsDisadvantages:waste of bandwidth if traffic distributed unevenlyinflexibleguard spaces

k3k4k5k6ftc

Time multiplexChannel gets the whole spectrum for a certain amount of timeAdvantages:only one carrier in the medium at any timethroughput high even for many usersDisadvantages:precise synchronization necessaryftck2k3k4k5k6k1

Time and frequency multiplexA channel gets a certain frequency band for a certain amount of time (e.g. GSM)Advantages:better protection against tappingprotection against frequency selective interferencehigher data rates compared to code multiplexPrecise coordination requiredftck2k3k4k5k6k1

Code multiplexEach channel has unique codeAll channels use same spectrum at same timeAdvantages:bandwidth efficientno coordination and synchronizationgood protection against interferenceDisadvantages:lower user data ratesmore complex signal regenerationImplemented using spread spectrum technologyk2k3k4k5k6k1ftc

Spread Spectrum TechnologyProblem of radio transmission: frequency dependent fading can wipe out narrow band signals for duration of the interferenceSolution: spread the narrow band signal into a broad band signal using a special codedetection atreceiverinterferencespread signalsignalspreadinterferenceffpowerpower

Spread Spectrum TechnologySide effects:coexistence of several signals without dynamic coordinationtap-proofAlternatives: Direct Sequence (DS/SS), Frequency Hopping (FH/SS)Spread spectrum increases BW of message signal by a factor N, Processing Gain

Effects of spreading and interferencePfi)Pfii)senderPfiii)Pfiv)receiverfv)user signalbroadband interferencenarrowband interferenceP

Spreading and frequency selective fadingfrequencychannel quality123456Narrowband signalguard space22222 frequencychannel quality1spread spectrumnarrowband channelsspread spectrum channels

DSSS (Direct Sequence Spread Spectrum) IXOR the signal with pseudonoise (PN) sequence (chipping sequence)Advantagesreduces frequency selective fadingin cellular networks base stations can use the same frequency rangeseveral base stations can detect and recover the signalBut, needs precise power controluser datachipping sequenceresultingsignal0101101010100111XOR01100101101001=TbTc

DSSS (Direct Sequence Spread Spectrum) IIuser datam(t)chippingsequence, c(t)XmodulatorradiocarrierSpread spectrumSignal y(t)=m(t)c(t)transmitsignaltransmitterdemodulatorreceivedsignalradiocarrierXChipping sequence, c(t)receiverintegratorproductsdecisiondatasampledsumscorrelator

DS/SS Comments IIIPseudonoise(PN) sequence chosen so that its autocorrelation is very narrow => PSD is very wideConcentrated around t < TcCross-correlation between two users codes is very small

DS/SS Comments IVSecure and Jamming ResistantBoth receiver and transmitter must know c(t)Since PSD is low, hard to tell if signal presentSince wide response, tough to jam everythingMultiple accessIf ci(t) is orthogonal to cj(t), then users do not interfereNear/Far problemUsers must be received with the same power

FH/SS (Frequency Hopping Spread Spectrum) IDiscrete changes of carrier frequencysequence of frequency changes determined via PN sequenceTwo versionsFast Hopping: several frequencies per user bit (FFH)Slow Hopping: several user bits per frequency (SFH)Advantagesfrequency selective fading and interference limited to short perioduses only small portion of spectrum at any timeDisadvantagesnot as robust as DS/SSsimpler to detect

FHSS (Frequency Hopping Spread Spectrum) IIuser dataslowhopping(3 bits/hop)fasthopping(3 hops/bit)01Tb011tff1f2f3tTdff1f2f3tTdTb: bit periodTd: dwell time

FHSS (Frequency Hopping Spread Spectrum) IIImodulatoruser datahoppingsequencemodulatornarrowbandsignalSpread transmitsignaltransmitterreceivedsignalreceiverdemodulatordatafrequencysynthesizerhoppingsequencedemodulatorfrequencysynthesizer

Applications of Spread Spectrum Cell phonesIS-95 (DS/SS)GSMGlobal Positioning System (GPS)Wireless LANs802.11b

Performance of DS/SS SystemsPseudonoise (PN) codes Spread signal at the transmitterDespread signal at the receiverIdeal PN sequences should beOrthogonal (no interference)Random (security)Autocorrelation similar to white noise (high at t=0 and low for t not equal 0)

PN Sequence GenerationCodes are periodic and generated by a shift register and XORMaximum-length (ML) shift register sequences, m-stage shift register, length: n = 2m 1 bits+Output

Generating PN SequencesTake m=2 =>L=3cn=[1,1,0,1,1,0, . . .], usually written as bipolar cn=[1,1,-1,1,1,-1, . . .]+Output

mStages connected to modulo-2 adder21,231,341,451,461,681,5,6,7

Problems with m-sequencesCross-correlations with other m-sequences generated by different input sequences can be quite highEasy to guess connection setup in 2m samples so not too secureIn practice, Gold codes or Kasami sequences which combine the output of m-sequences are used.

Detecting DS/SS PSK SignalsXBipolar, NRZm(t)PNsequence, c(t)Xsqrt(2)cos (wct + q)Spread spectrumSignal y(t)=m(t)c(t)transmitsignaltransmitterXreceivedsignalXc(t)receiverintegratorz(t)decisiondatasqrt(2)cos (wct + q)LPFw(t)x(t)

Optimum Detection of DS/SS PSKRecall, bipolar signaling (PSK) and white noise give the optimum error probability

Not effected by spreadingWideband noise not affected by spreadingNarrowband noise reduced by spreading

Signal SpectraEffective noise power is channel noise power plus jamming (NB) signal power divided by N

Multiple Access PerformanceAssume K users in the same frequency band, Interested in user 1, other users interfere413526

Signal ModelInterested in signal 1, but we also get signals from other K-1 users:

At receiver,

Interfering Signal After mixing and despreading (assume t1=0)

After LPF

After the integrator-sampler

At Receiverm(t) =+/-1 (PSK), bit duration TbInterfering signal may change amplitude at tk

At User 1:Ideally, spreading codes are Orthogonal:

Multiple Access Interference (MAI)If the users are assumed to be equal power interferers, can be analyzed using the central limit theorem (sum of IID RVs)

Example of Performance DegradationN=8N=32

Near/Far Problem (I)Performance estimates derived using assumption that all users have same power levelReverse link (mobile to base) makes this unrealistic since mobiles are movingAdjust power levels constantly to keep equal1k

Near/Far Problem (II)K interferers, one strong interfering signal dominates performanceCan result in capacity losses of 10-30%

Multipath Propagation

RAKE ReceiverReceived signal sampled at the rate 1/Ts> 2/Tc for detection and synchronizationFed to all M RAKE fingers. Interpolation/decimation unit provides a data stream on chiprate 1/Tc Correlation with the complex conjugate of the spreading sequence and weighted (maximum-ratio criterion)summation over one symbol

RAKE ReceiverRAKE Receiver has to estimate:Multipath delaysPhase of multipath componentsAmplitude of multipath componentsNumber of multipath componentsMain challenge is receiver synchronization in fading channels