Guest Editorial Special Issue On Multirate Systems, Filter Banks, Wavelets, And Applications - Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on

THE LAST decade has seen a tremendous amount of activity and emergence of applications in the areas of filter banks and wavelets. These topics are of such wide interest that there have been papers in many different journals, conferences, and workshops in diverse disciplines. However, many aspects of the theory, design, and application of filter banks and wavelets are of great interest to the circuits and systems community as well. Our editorial team felt that this was a perfect time to put together a special issue with stateof-the-art papers on these popular topics. All of the papers have been peer-reviewed according to the usual practice of this TRANSACTIONS. Almost in parallel, there is also a similar special issue (April 1998) by the IEEE TRANSACTIONS ON SIGNAL PROCESSING, with a slightly greater emphasis on applications. Many well-known authors have contributed articles to these special issues and we expect these to serve as valuable references for a long time to come. In this Special Issue, we have 11 regular papers and 19 transactions briefs, covering a wide range of theory and applications. The paper by Chen, Liu, and Bai considers multirate systems in very general building blocks, with natural application in nonuniform filter bank design. The work by Unser and Zerubia deals with sampling theorems for signals that are not band-limited, clearly showing the connection to filter bank theory. Cotronei, Montefusco, and Puccio present the recent theory of multiwavelets using recursive matrices and a simpler notation emphasizing the importance of choice of the prefilter, and show application in compression. The next paper, by Wei and Bovik, considers the effect of vanishing moments on the phase linearity near zero-frequency, and also the effect of distributing the moments, showing how to construct nearly symmetric coiflets with performance comparable to some biorthogonal splines. The paper by Hsu and Rao, which is an application in High Definition Television (HDTV) coding, shows how to use wavelets and lapped transforms for motion compensation. The application to image coding is also considered in the paper by Kim, Hu, and Nguyen, where the authors consider entropy constrained vector quantizers for the wavelet coefficients. The work by Kim, Lee, Isshiki, and Kunieda focuses on hardware considerations, emphasizing VLSI implementation of one–dimensional wavelet transforms using scalable, possibly pipelined, architectures.