Reviews and New Releases: Multirate Digital Signal Processing

Abstract

Multirate digital signal processing is a mature topic of research: applications are numerous in very different fields, from FSK modems to Digital Audio Broadcasting and from Asynchronous Conversion of sampling rates to SpeechiAudio and Image/Video Coding.The book by N.J. Fliege presents in a clear way algorithms of Multirate DSP, emphasizing their computational efficiency and possible applications. The contents of the book are better understood by readers with a background on basic "Signals and Systems" elements."Multirate Digital Signal Processing" by N.J. Fliege is an useful book both for beginners and experts. Teachers could adopt it as a reference text at graduated level, and some part of it at undergraduate level. Moreover, engineers can found here a compact and clear reference book fir developing applications.In the first chapter the Polyphase time/frequency representation of discrete signals is given and the important relationship between polyphase representation and modulation is derived and properly emphasized. This chapter also includes the theoretical basis of decimation and interpolation.Chapters 2 and 3 form a well organized part of the book, dealing with the design of Finite Impulse Response (FIR) filters. First (chapter 2), the most important topological structures and characteristics of FIR filters are derived; then (chapter 3), the standard design techniques of FIR filters are analyzed in details. Thought these topics can also be found in general DSP books, their inclusion in the book is advantageous to the reader for two main reasons: symbols and definitions that are used in more advanced chapters are clearly introduced, topics that att usually scattered in a number of books are usehlly summarized here.,' In chapter 4 decimation and interpolation combined with anti-alias filters are dealt with. The main scope of this chapter is the identification of cornputationally efficient structures. Strictly connected to the material presented in chapter 4 is the use of multi-rate techniques to improve the computational efficiency of redundant digital filters which is dealt with in chapter 5 .Chapters from 6 to 9 are dedicated to the analysis and design of filter banks. These four chapters form the core of the book since filter banks is probably the main application of Multirate DSP. In chapters 6 and 7 important topics such as QMF banks, perfect reconstruction, and linear phase are dealt with for the two and M-channel cases. Chapter 8 is mainly devoted to Polyphase filter banks and to computational efficiency of DFT polyphase filter banks. In chapter 9 wavelet transform is introduced and the strict relationship between dyadic wavelet and octave filter banks is derived in a clear way.Finally chapter 10 gives an overview of the most important applications of multirate DSP.Usehl numerical examples are also provided in each chapter helping the reader to become familiar with the different topics.In conclusion, the book "Multirate Digital Signal Processing" by N.J. Fliege is an useful tool for researchers and en...

Compressão Seletiva de Imagens Coloridas com Base em Redes Neurais

“…Neste método foram utilizadas as wavelets diádicas , que possuem relação com banco de filtros de reconstrução perfeita [7]. O sinal é decomposto utilizando dois filtros: um passa alta e outro passa baixas freqüências.…”

A Study on Performance Analysis for Error Probability in SWSK Systems

“…To compute the next level of resolution, the decimated sequence of detailed coefficients are inputted to the next set of quadrature filters. In this fashion the wavelet coefficients at the n th level of decomposion can be obtained [4] [5].…”

“…The simplest wavelet is Haar [4] [5] and the communication signals considered are amplitude shift keying (ASK), frequency shift keying(FSK) and phases shift keying(PSK), which are transmitted over an AWGN channel.…”

Selection of Wavelet Subbands Using Genetic Algorithm for Face Recognition

“…The wavelet transform [23,24] is expressed as an inner product of a signal f (x) with a family of functions which are translations and dilations of a mother wavelet function ψ(x) and the scaling function φ(x). This is used to split the signal into approximation and detail.…”