Tracking of Time-varying Mobile Radio Channels,
Part I: The Wiener LMS Algorithm.

Lars Lindbom , Mikael Sternad and Anders Ahlén

IEEE Transactions on Communication, December 2001, pp.2207-2217. © IEEE

A brief version of the paper,
"Channel Tracking with WLMS Algorithms:
High Performance at LMS Computational Loads"
by A. Ahlén, L. Lindbom and M. Sternad, is published at
IEEE Vehicular Technology Conference -VTC200-Spring Tokyo, Japan, May 15-18, 2000, pp 16-20. © IEEE

Outline:

We here propose a novel way of extending and optimizing the structure of LMS-like adaptation laws. These results were originally motivated by the difficult problem of accurately tracking rapidly time-varying channel parameters in the IS-136 cellular system. An early version of the proposed algorithm has successfully been used on IS-136 1900MHz channels and a case study on this particular application can be found in Part II of this work.

Motivated also by other applications such as multi-antenna systems multi-carrier systems and multiuser detectors, a framework has been developed for designing low-complexity algorithms for tracking coefficients of linear regression models under assumptions which are realistic in communications applications. Our aim is to improve upon the sometimes inadequate tracking performance offered by standard LMS and RLS algorithms.

Abstract:

Adaptation algorithms with constant gains are designed for tracking smoothly time-varying parameters of linear regression models, in particular channel models occurring in mobile radio communications. In a companion paper, an application to channel tracking in the IS-136 TDMA system is discussed.

The proposed algorithms are based on two key concepts. First, the design is transformed into a Wiener filtering problem. Second, the parameters are modeled as correlated ARIMA processes with known dynamics. This leads to a new framework for systematic and optimal design of simple adaptation laws based on a priori information. They can be realized as Wiener filters, called Learning Filters, or as ``LMS/Newton'' updates complemented by filters that provide predictions or smoothing estimates.

The simplest algorithm, named Wiener LMS, is presented here. All parameters are here assumed governed by the same dynamics and the covariance matrix of the regressors is assumed known. The computational complexity is of the same order of magnitude as that of LMS for white regressors. The tracking performance is however substantially improved.

Related publications:

Part II: A Case Study on IS-136 channels.

Design of the general constant-gain adaptation algorithms.

Analysis of stability and performance, for slow and fast variations.

PhD Thesis by Lars Lindbom.

Licenciate Thesis by Lars Lindbom, on averaged Kalman designs (KLMS)
and on deterministic sinusoid modelling of fading channels.

Conference paper (IFAC Como 2001) on averaged robust design for uncertain fading models.

Sources:

Conference version (VTC2000S): Postscript, 195K ; Pdf, 443K

Paper version (IEEE-COM): Postscript, 306K ; Pdf, 273K

Matlab design:

wlms_design.m With this file, Wiener LMS adaptation law can be designed if ARIMA models of the parameters to be tracked are given.

spefac.m Function called by wlms_design.m

QLpoly.m Function called by wlms_design.m.