《现代数字信号处理AdvancedDigitalSignalProcessing-introdu》由会员分享，可在线阅读，更多相关《现代数字信号处理AdvancedDigitalSignalProcessing-introdu（12页珍藏版）》请在金锄头文库上搜索。
1、Advanced Digital Signal ProcessingSchool of Information Engineering, Hebei University of TechnologyZeng Cheng20011.4IntroductionThe Classical Digital Signal Processing (CDSP) and the Advanced Digital Signal Processing (ADSP)nCDSPnSignal: one-dimensional deterministic signalnMethods: nonparametric (w
2、ave-based) methods, such as DFTnLinear, causal and minimum-phase systems (mostly)nInvariant sampling ratenCDSP and its generalizationMethod: wave-basedSystem: LTI, minimal-phaseSample rate: invariantSignal: 1-D, deterministicCDSPStationary, GaussianRandom signalNon-GaussianMulti-dimensionVariant rat
3、e: multirateParametric-basedNon- linearNon-minimal phaseTime-variantNon-stationarynMain topics in ADSPnClassical statistical signal processingnSignals: stationary random (stochastic) signal (often with Gaussian or normal distributions)nMethods: model-based methods (e.g. AR linear prediction model),
4、Bayesian Statistical Signal Processing (based on stochastic process analysis)nMain subjectslStationary stochastic signal analysislBayesian estimationlOptimal filtering (Wiener filter & Kalman filter)lAdaptive signal processinglPower spectrum estimation (especially the modern methods)nNon-stationary
5、stochastic signal processingnHidden Markov Models (HMM) Widely applied in the processing of “naturally generated” signals, such as audio signals, image signals, biomedical signals and seismic signalsnTime-frequency analysis (short-time Fourier transform (STFT), Gabor decomposition, Wigner distributi
6、on)nWavelet analysisnMultirate signal processingnApplications: discrete-time systems with signals sampled in unequal sampling ratesnDown-sampling, up-sampling, filter banksnNon-Gaussian stochastic signal processingnSignals: stochastic signals without Gaussian distributionsnMethods: high-order statis
7、tic analysisnMulti-dimension digital signal processingnSignals: multi-dimension (often two-dimension) stochastic or deterministic signals, such as digital image signals nSome theories based on one-dimension signals can not be generalized into multi-dimension signals directlynThe processing of multi-
8、dimension signals confronts some new problems which never occur in the case of one-dimensionnNonlinear signal processingnProcessing signals with nonlinear system theory and method (notice that what is nonlinear here is the system, not the signal)nMethodslNonlinear filteringlCepstrum and homogenous d
9、econvolutionlNeural network signal processingAbout Our CoursesnContent organization Basic theories and algorithms for the processing of discrete (digital) random signals, including their implementations and applicationsnFundamentals of discrete random signal AnalysisnWiener filter and discrete Kalma
10、n filternAdaptive signal processingnPower spectrum estimationnTime-frequency analysis (STFT)nWavelet analysisnThe principal objectivesGive an introduction to basic concepts and methodologies that can provide the foundation for further study, research and applicationnBasic requirementsFor all discuss
11、ed subjects (theory or algorithm),nUnderstanding their fundamental mathematical principlesnKnowing their characteristics and appreciating their limitationsnImplementation & applicationsnUsing them to analyze simple practical problemsnPrerequisitesnSignal and systemnDigital signal processing (classic
12、al)nProbability theory and stochastic processnLinear algebranFinal examinationnA written examination for the students majoring in “communication and information system”，“circuit and system”nA term paper for others?nTextbook and referencesn丁玉美，阔永红，高新波.数字信号处理时域离散随机信号处理（研究生教材）. 西安电子科技大学出版社，2002 (textbo
13、ok)n姚天任.现代数字信号处理. 华中理工大学出版社，1999 n陆传贲.现代信号处理导论. 北京邮电大学出版社，2003n张明友, 吕明. 近代信号处理理论与方法. 国防工业出版社，2005nS. V. Vaseghi. Advanced Digital Signal Processing and Noise Reduction, Second Edition. John Wileys Sons. LTD., 2000nV. K. Ingle and S. M. Kogon. Statistical and Adaptive Signal Processing. McGraw-Hill, 2000 nOthersAbout the Teacher曾曾 成成E-mail: zeng_Tel: 26647962(H), 13920326097Office: 信息学院楼信息学院楼406室室
