Abstract
This paper presents the simulation of a communications
system which contains an adaptive
filter. The filter is introduced in order to compensate
for the channel effects, thus emulating
a channel adaptive equalizer. At the discretion
of the user, various system parameters can be
changed. The simulation results can be postprocessed,
as a hardcopy is created and passed
to the MATLAB workspace.
1 Introduction
Adaptive filters can be integrated in systems with different
functionalities, being prediction, system identification,
equalization and interference canceling examples
of such. In the present case, one such filter has been incorporated
in a typical adaptive equalizer configuration.
By dynamically adjusting the finite impulse response
(FIR) filter coefficients, as dictated by an adaptation
algorithm, one can compensate the channel effects. The
algorithm choosen for this simulation is the Recursive
Least Squares (RLS), which is one of the most efficient
algorithms for implementation of adaptive filters. That
is the main reason why the Least Mean Squares (LMS)
algorithm still dominates most of the applications in
this area. Despite taking more iterations to converge,
it is possible to do all the processing between each iteration
with much less computational effort in the LMS algorithm.
On the other hand, with the decreasing prices
of the Digital Signal Processors (DSPs) and also with
the increasing capabilities of general purpose microprocessors,
the RLS is an algorithm whose time has come.

DOWNLOAD FULL REPORT
http://googleurl?sa=t&source=web&cd=3&ve...1996_1.pdf&ei=o7EBTuu_EJHKrAe0rbmGAw&usg=AFQjCNHflgwEC2VmD5BAlv02DvUlKRbTjA&sig2=00upuJz8sbuJZVEdgWGdIw