Product Description
A Unified Treatment of Non-Gaussian Processes and Nonlinear Signal Processing Nonlinear signal processing methods are finding numerous applications in such fields as imaging, teletraffic, communications, hydrology, geology, and economics–fields where nonlinear systems and non-Gaussian processes emerge. Within a broad class of nonlinear signal processing methods, this book provides a unified treatment of optimal and adaptive signal processing tools that mirror tho… More >>
Nonlinear Signal Processing: A Statistical Approach
Tags: adaptive signal processing, gaussian processes, geology, hydrology, nonlinear systems, optimal and adaptive signal processing, processing tools, statistical approach, teletraffic, tho
#1 by G. Liu on July 3, 2010 - 2:36 pm
I’m a Ph.D student in Operations Research. I loved this book since I first read it. The book provides comprehensive and up-to-date analysis and application description of two types of very important non-normal (nonGaussian) models: generalized Gaussian distribution model and stable distribution model. These two types of models are key models in successfully modeling real data in an impulsive enviornment. The book focuses on Laplace distribution models for generalized Gaussian distribution models and on stable distribution models.
The book provides many tools (algorithms, smoothers, and filters) to assist in applying nonGaussian models in a real life. Order statistics is presented in one early chapter and then used to produce these useful tools, which include weighted median smoother, stack smoother, LAD (least absolute deviation) algorithm, weighted median in LAD regression, LMA (least mean absolute) algorithm, weighted median filter, optimal weighted median filter, recursive weighted median filter, complex weighted median filter, weighted myriad smoother, and weighted myriad filter. To me, the concept of myriad smoother and myriad filter is very new. They are, as I understand, especially deveoped for stable distribution models.
In recent years, stable distribution has attracted more and more attentions. However, the parameter estimation of stable distribution is always difficult. This book uses FLOM (fractional low-order moments) procedure to estimate stable distribution parameters, based on Kuruoglu’s early work.
The book’s CD contains MATLAB files of algorithms, smoothers and filters. Thus all tools can be tested for quick results and modified for furthering improvement. Certainly, these MATLAB files are a plus for (expert) people looking into and examining details.
Rating: 5 / 5
#2 by Robert Birgam on July 3, 2010 - 4:41 pm
I have been searching for a book on this topic that covers the fundamentals all the way to more advanced topics, particularly on design tools. This book is perfect for someone who is interested in learning the field, broadly. Other books I looked at seemed too specialized without the basics or simply a gathering of test cases. The software provided is a BIG plus!
Rating: 5 / 5