Sparse Signal Processing
作者/authors
摘要/abstract
Conventional sampling techniques are based on Shannon-Nyquist theory which states that the required sampling rate for perfect recovery of a band-limited signal is at least twice its bandwidth. The band-limitedness property of the signal plays a significant role in the design of conventional sampling and reconstruction systems. As the natural signals are not necessarily band-limited, a low-pass filter is applied to the signal prior to its sampling for the purpose of antialiasing. Most of the signals we are faced with are sparse rather than band-limited (or low pass). It means that they have a small number of non-zero coefficients in some domain such as time, discrete cosine transform (DCT), discrete wavelet transform (DWT), or discrete fourier transform (DFT). This characteristic of the signal is the foundation for the emerging of a new signal sampling theory called Compressed Sampling, an extension of random sampling. In this chapter, an overview of compressed sensing, together with a summary of its popular recovery techniques, is presented. Moreover, as a well-known example of structured sparsity, the block sparse recovery problem is investigated and the related recovery approaches are illustrated.
目录/contents
1 Abstract Exact and Approximate Sampling Theorems ................ 1M.M. Dodson2 Sampling in Reproducing Kernel Hilbert Space ........................ 23J.R. Higgins3 Boas-Type Formulas and Sampling in Banach Spaceswith Applications to Analysis on Manifolds ............................. 39Isaac Z. Pesenson4 On Window Methods in Generalized Shannon SamplingOperators .................................................................... 63Andi Kivinukk and Gert Tamberg5 Generalized Sampling Approximation for MultivariateDiscontinuous Signals and Applications to Image Processing ......... 87Carlo Bardaro, Ilaria Mantellini, Rudolf Stens, Jörg Vautz,and Gianluca Vinti6 Signal and System Approximation from General Measurements ..... 115Holger Boche and Ullrich J. Mönich7 Sampling in Image Representation and Compression.................. 149John J. Benedetto and Alfredo Nava-Tudela8 Sparse Signal Processing................................................... 189Masoumeh Azghani and Farokh Marvasti9 Signal Sampling and Testing Under Noise ............................... 215Mirosław Pawlak10 Superoscillations ............................................................ 247Paulo J.S.G. Ferreira11 General Moduli of Smoothness and Approximationby Families of Linear Polynomial Operators ............................ 269K. Runovski and H.-J. Schmeisser12 Variation and Approximation in Multidimensional Settingfor Mellin Integral Operators ............................................. 299Laura Angeloni and Gianluca Vinti13 The Lebesgue Constant for Sinc Approximations ...................... 319Frank Stenger, Hany A.M. El-Sharkawy, and Gerd Baumann14 Six (Seven) Problems in Frame Theory .................................. 337Ole Christensen15 Five Good Reasons for Complex-Valued Transformsin Image Processing ........................................................ 359Brigitte Forster16 Frequency Determination Using the Discrete Hermite Transform ... 383Dale H. Mugler and Stuart Clary17 Fractional Operators, Dirichlet Averages, and Splines................. 399Peter Massopust18 A Distributional Approach to Generalized StochasticProcesses on Locally Compact Abelian Groups ......................... 423H.G. Feichtinger and W. Hörmann19 On a Discrete Turán Problem for `-1 Radial Functions ............... 447Elena E. Berdysheva and Hubert Berens