Adaptive Wiener Filters for Control and Signal Processing: A Project Summary

Anders Ahlén and Mikael Sternad,

NUTEK Workshop on Digital Communications, Kista, Sweden, June 1-2, 1993.

In Postscript (without figures) : compressed(gz) 40K , uncompressed 178K.
In Pdf 185K.

Outline:

Our research has partly been supported by the Nutek program on Digital Communications and later on Telecommunications since 1988. The paper summarizes our activities in 1992-1993.

Abstract:

In the project, work has been carried out on problems spanning the fields of Signal processing, Communications, and Control. A guiding principle has been to formulate general problems, originating from relevant applications. Here equalization of fading mobile radio channels has served as a main source of inspiration, and also a useful application for testing new ideas. The aim is, however, to develop general tools and methods useful for a large range of problems.

Very central in our work is the desire to obtain explicit solutions and to gain engineering insight. As a means to accomplish that, the Polynomial Systems framework has been used. It describes linear dynamic systems in input-output form, and has been useful in our study of general IIR-filter structures, implicit adaptive schemes, probabilistic descriptions of model errors, and utilization of a priori information. Our long term goal is to provide a general model-based design concept, for solving a variety of problems in Signal Processing, Communications and Control. The publication of the book chapters [B2] , [B3] and [B4] are steps towards this goal.

A short summary of the main activities, and a list of publications, are given in the paper.

Related publications:

Licentiate Thesis by L Lindbom 1992, on tracking of time-varying mobile radio channels.
Paper in Automatica 1993, on the design of (scalar) robust filters.
Paper in Signal Processing 1994, on filter design via inner-outer factorization.