1、 本文由惩戒之疾风贡献ppt文档可能在WAP端浏览体验不佳。建议您优先选择TXT，或下载源文件到本机查看。CHAPTER 5. SPECTRUM SENSINGIFA2007COGNITIVE CLASS1What is Spectrum Sensing ?How to detect spectrum holes by the COGNITIVE RADIO so that it can adapt itself to its environment !IFA2007COGNITIVE CLASS2Spectrum SensingTransmitted SignalRadio Environm
2、entRF StimuliSpectrum MobilityPrimary User Detection Decision RequestSpectrum SensingSpectrum HoleSpectrum SharingChannel CapacitySpectrum DecisionIFA2007COGNITIVE CLASS3EFFICIENT WAY TO DETECT SPECTRUM HOLESA general CR based communication scenarioCR User 2No interaction between CR user and Primary
3、 Tx/Rx CR user must rely on locally sensed signals to infer primary user activityLicensed band 2 CR User 1 Licensed band 1Primary Tx Primary RxChannels found occupied by CR user (Licensed bands 1 and 2) are now avoided during communication between CRsIFA2007COGNITIVE CLASS4EFFICIENT WAY TO DETECT SP
4、ECTRUM HOLES !Detect primary users that are receiving data within the communication range of a CR user.In reality Difficult for a CR to detect primary user activity in the absenceof interaction between primary users and itself. RECENT RESEARCH How to detect primary users based on “local observation“
5、 of CR users (from its environment)IFA2007COGNITIVE CLASS5Classification of Spectrum Sensing TechniquesSpectrum SensingTransmitter Detection Receiver Detection Interference Temperature ManagementMatched Filter DetectionEnergy DetectionCyclostationary Feature DetectionIFA2007COGNITIVE CLASS6Transmitt
6、er DetectionCR should distinguish between Used and Unused spectrum bands.CR should have the capability to determine if a signal from primary user (transmitter) is locally present in a certain spectrum. Transmitter Detection Approach Detection of the signal (weak signal as the worst case) from a prim
7、ary user through local observations of CR users.COGNITIVE CLASS7IFA2007Basic Hypothesis Model for Transmitter DetectionThe signal r(t) received (detected) by the CR (secondary) user isH0 n(t) r(t) = hs(t) + n(t) H1where n(t) s(t) AWGN (Additive White Gaussian Noise) Transmitted signal of the primary
8、 user Amplitude gain of the channel Null hypothesis No licensed user signal in a certain spectrum band. Alternative hypothesis There exists some licensed user signal. COGNITIVE CLASSh H0 H1IFA20078Transmitter DetectionSensing for Cognitive Radios,“ in Proc. 38th Asilomar Conference on Signals,System
9、s and Computers, pp. 772-776, Nov. 2004.D. Cabric, S. M. Mishra, and R. W. Brodersen, “Implementation Issues in SpectrumThree schemes are generally used for the transmitter detection according to the hypothesis model. Matched Filter Detection Energy Detection and Cyclostationary Feature Detection Te
10、chniquesIFA2007COGNITIVE CLASS9Matched Filter DetectionSpectrum SensingTransmitter Detection Receiver DetectionInterference Temperature ManagementMatched Filter Energy Detection DetectionCyclostationary Feature DetectionIFA2007COGNITIVE CLASS10Matched Filter DetectionMatched Filter Received Signal r
11、(t) = s(t) + n(t) Sample at t = T Threshold Devicet0r( )s(T t + )dY Y Y | H1 = Qm( 2 , ) d (m, / 2) Pf = PY | H0 = (m)where is the SNR m = TW is the (observation/sensing) time bandwidth product () and (, ) are complete and incomplete gamma functions Qm( ) is the generalized Marcum Q-function.COGNITI
12、VE CLASS19IFA2007Energy DetectionFading Environment: The amplitude gain of the channel varies due to the shadowing/fading - variation of SNR Pf is the same as that of non-fading case (independent of SNR,) non(independent SNR, Pd gives the probability of the detection conditioned on instantaneous SNR
13、 as:P = Qm ( 2 , ) f (x)dx dxwhere f(x) is the probability distribution function of SNR under fading.COGNITIVE CLASS20IFA2007Energy DetectionA low Pd missing the presence of the primary user with high probability increases the interference to the primary user A high Pf low spectrum utilization (sinc
14、e false alarms increase the number of missed opportunities). Implementation is easy!COGNITIVE CLASS21IFA2007Problems of Energy DetectionPerformance is susceptible to uncertainty in noise power. SNR problem! (Pilot tone from primary user helps to improve the accuracy of the energy detector)Energy det
15、ector cannot differentiate signal types but can only determine thepresence of the signal. Energy detector is prone to the false detection triggered by the unintended signals. Energy detector needs longer sensing time Matched filter: T1/SNR when detecting weak signals: Energy Detector: T1/SNR2SNR 1(-
16、10dB to -40 dB) 1(dB)COGNITIVE CLASS22IFA2007Cyclostationary Feature DetectionSpectrum SensingTransmitter Detection Receiver DetectionInterference Temperature ManagementMatched Filter DetectionEnergy DetectionCyclostationary Feature Detection23IFA2007COGNITIVE CLASSCyclostationary Feature DetectionS
17、ensing for Cognitive Radios,“ in Proc. 38th Asilomar Conference on Signals,Systems and Computers, pp. 772-776, Nov. 2004. A. Fehske, J. D. Gaeddert, and J.H. Reed, “A New Approach to Signal Classification Using Spectral Correlation andNeural Networks,“ in Proc. IEEE DySPAN, pp. 144-150, Nov. 2005. ,
18、 H. Tang, “SomePhysical Layer Issues of Wideband Cognitive Radio System,“ in Proc. IEEE DySPAN,pp. 151-159, Nov. 2005. ,D. Cabric, S. M. Mishra, and R. W. Brodersen, “Implementation Issues in SpectrumModulated signals are in general coupled with sine wave carriers, pulse trains, repeating spreading,
19、 hopping sequences, or cyclic prefixes, which resultin built-in builtperiodicity.COGNITIVE CLASS24IFA2007Cyclostationary Feature DetectionThese modulated signals are characterized as cyclostationary since their mean and autocorrelation exhibit periodicity. These features are detected by analyzing a
20、spectral correlation function. Advantage of the spectral correlationfunction: differentiates the noise energy from modulated signal energyIFA2007COGNITIVE CLASS25Cyclostationary Feature Detection Sine based Cyclostationary DetectionPrimary Tx frequency repeats over symbol durations at regular interv
21、als TProblem: Can these cyclical regularities be detected at the CR user?IFA2007COGNITIVE CLASS26Cyclostationary Feature DetectionSensing for Cognitive Radios,“ in Proc. 38th Asilomar Conference on Signals,Systems and Computers, pp. 772-776, Nov. 2004. r(t) Correlate R(f+)R*(f- ) Average over T Feat
22、ure detectD. Cabric, S. M. Mishra, and R. W. Brodersen, “Implementation Issues in Spectrumr(t): Received signal R(f): Fourier transform of r(t) : Cyclic frequency R*(f): Complex conjugate of R(f) If cyclostationary with period T then cycle autocorrelation has component at =1/TCOGNITIVE CLASS27IFA200
23、7Cyclostationary Feature DetectionIf the correlation factor is high (greater than the threshold), there is a primary userIFA2007COGNITIVE CLASS28Cyclostationary Feature DetectionThis scheme performs better than the energy detector in discriminating against noise due to its robustness to the uncertai
24、nty in noise power. Computationally complex and requires significantly long observation time.IFA2007COGNITIVE CLASS29Limitations of the Transmitter DetectionReceiver Uncertainty Problem Shadowing ProblemIFA2007COGNITIVE CLASS30Receiver Uncertainty ProblemWith the transmitter detection, the CR user c
25、annot avoid the interference due to the lack of the primary receivers information (Fig.a). Moreover, the transmitter detection model cannot prevent the hidden terminal problem.IFA2007COGNITIVE CLASS31Shadowing ProblemA CR transmitter can have a good line-of-sight to a line-ofreceiver, but maynot be
26、able to detect the transmitter due to the shadowing (Fig. b). Consequently, the sensing information from other users is required for more accurate detection Cooperative DetectionCOGNITIVE CLASS32IFA2007Limitations of the Transmitter DetectionInterference Primary Transmitter RangePrimary Base-station
27、CR Transmitter RangeCR UserHidden Terminal Problem due to ShadowingPrimary UserCannot detect transmitterInterference CR UserPrimary Transmitter RangeCR Transmitter RangeInterference due to uncertainty of receiver locationPrimary Base-stationPrimary UserCannot detect transmitterIFA2007COGNITIVE CLASS
28、33Transmitter DetectionNonNon-Cooperative vs Cooperative DetectionDetection Method Transmitter Detection Detection Behavior Transmitter DetectionMatched Filter DetectionEnergy DetectionCyclostationary Feature DetectionNon-cooperative DetectionCooperative DetectionIFA2007COGNITIVE CLASS34NonNon-Coope
29、rative vs Cooperative DetectionNonNon-Cooperative Detection CR users detect the primary transmitter signal independently through theirlocal observations.Cooperative Detection Refers to spectrum sensing methods where information from multiple CR users are incorporated for primary user detection. Allo
30、w to mitigate the multi-path fading and shadowing effects, which multiimproves the detection probability in a heavily shadowed environmentIFA2007COGNITIVE CLASS35Cooperative DetectionG. Ganesan and Y.G. Li, “Cooperative Spectrum Sensing in Cognitive Radio Networks,“ in Proc. IEEE DySPAN 2005 S. M. M
31、ishra, A. Sahai and R. W. Brodersen,“Cooperative sensing among cognitive radios ,“ in Proc. IEEE ICC 2005.Cooperative Detection can be implemented either in a centralized or in a distributed manner. Centralized MethodCR base-station plays a role to gather all sensing information from basethe CR user
32、s and detect the spectrum holes.Distributed Method require exchange of observations among CR users.COGNITIVE CLASS36IFA2007Cooperative DetectionCooperative Methodsprovide more accurate sensing performance, they cause adverse effects on resource-constrained resourcenetworks due to the additional oper
33、ations and overhead traffic.PROBLEM The primary receiver uncertainty problem caused by the lack of the primary receiver location knowledge is still unsolved!COGNITIVE CLASS37IFA2007Primary Receiver DetectionSpectrum SensingTransmitter DetectionReceiver DetectionCyclostationary Feature DetectionInter
34、ference Temperature ManagementMatched Filter DetectionEnergy DetectionIFA2007COGNITIVE CLASS38Primary Receiver DetectionB. Wild and K. Ramchandran, “Detecting Primary Receivers for Cognitive RadioApplications“ in Proc. IEEE DySPAN, pp. 124-130, Nov. 2005.A direct receiver detection method is develop
35、ed for detection of primary receivers where the local oscillator (LO) leakage power emitted by the RF front-end of the primary receiver front-IFA2007COGNITIVE CLASS39Primary Receiver DetectionHowever, since LO leakage signal is typically weak, implementation of a reliable detector is not trivial. Cu
36、rrently this method is only feasible in the detection of the TV receivers.IFA2007COGNITIVE CLASS40Interference Temperature ManagementSpectrum SensingTransmitter Detection Receiver DetectionInterference Temperature ManagementMatched Filter DetectionEnergy DetectionCyclostationary Feature DetectionIFA
37、2007COGNITIVE CLASS41Interference Temperature ManagementInterference is typically regulated in a transmittertransmittercentric way Interference can be controlled at the transmitter through * radiated power, * out-of-band emissions and out-of* location of individual transmitters.COGNITIVE CLASS42IFA2
38、007Interference Temperature ManagementFCC, ET Docket No 03-289 Notice of Inquiry and Notice of Proposed RulemakingNov. 2003.However, interference actually takes place at the receivers Therefore a new model for measuring interference, referred to as interference temperature introduced by the FCC.IFA2
39、007COGNITIVE CLASS43Interference Temperature ModelLicensed SignalNew Opportunities for Spectrum Access Minimum Service Range with Interference CapPower at ReceiverInterference Temperature LimitService Range at Original Noise FloorOriginal Noise Floor Distance from Licensed Transmitting AntennaIFA200
40、7COGNITIVE CLASS44Interference Temperature ModelThe model shows the signal of a radio designed to operate in a range at which the received power approaches the level of the noise floor. As additional interfering signals appear, the noise floor increases at various points within the service area, as
41、indicated by the peaks above the original noise floor.COGNITIVE CLASS45IFA2007Interference Temperature ModelModel manages interference at the receiver through the interference temperature limit, which is represented by the amount of new interference that the receiver could tolerate.IFA2007COGNITIVE
42、CLASS46Interference Temperature ModelI.o.w., the interference temperature model accounts for the cumulative RF energy from multiple transmissions and sets a maximum cap on their aggregate level. As long as CR users do not exceed this limit by their transmissions, theycan use this spectrum band.COGNI
43、TIVE CLASS47IFA2007Spectrum Sensing Challenges:Interference Temperature MeasurementNo practical way for a CR to measure or estimate the interference temperature. Only the noise and signals due to other CR users are considere as interference, and CR users must be separated from the received primary s
44、ignals to calculate the interference temperature. If CR users cannot measure the effect of their transmission o all possible receivers, a useful interference temperature measurement may not be feasible.The increase in the background interference will affect existing systems: reduced capacity and sma
45、ller cell sizesCOGNITIVE CLASS48IFA2007Spectrum Sensing Challenges:Spectrum Sensing in Multi-User Networks Multi- CR networks reside in a multi-user environment multi(multiple CR users and primary users). CR networks can also be co-located with other CR conetworks competing for the same spectrum ban
46、d.- Difficult to sense the primary users and to estimate the actual interference.Spectrum sensing functions need to be developed considerin the possibility of multi-user/network environment multiCOGNITIVE CLASS49IFA2007Spectrum Sensing Challenges:SpectrumSpectrum-Efficient SensingW.Y. Lee and I.F. A
47、kyildiz, “An Optimal Framework for Spectrum Sensing“ submitted for publication, April 2007. Sensing cannot be performed while transmitting packets. Hence, CR users should stop transmitting while sensing, which decreases the spectrum efficiency.Balance between the spectrum efficiency and sensing accuracy Sensing time directly affects the transmission performance ! Novel spectrum sensing algorithms need to be developed sensing time is minimized within a given sensingaccuracy.COGNITIVE CLASS50IFA20071
