Details

OPTIMAL CANCELLATION OF FREQUENCY-SELECTIVE COSITE INTERFERENCE

by MAXSON, BEN DAVID

Abstract (Summary)
As the demand for higher throughput has increased the operating bandwidths of communications systems, the effectiveness of traditional interference cancellation techniques has been diminished. Next-generation wideband military radio communications hardware will necessitate development of novel algorithms and architectures to provide cancellation of frequency-selective cosite interference under strict complexity constraints. This thesis represents the culmination of preliminary research into the feasibility of such a canceller. For a given signal bandwidth, channel transfer function, and number of filter taps, one important metric is the best possible cancellation. Calculation of the optimal tap weights and corresponding cancellation for a given set of tap delays is straightforward. However, finding the optimal delay for each tap is a more difficult non-linear optimization problem which sometimes involves a problematic error surface. Brute force searches of the tap delay space were performed to lend insight into the nature of the solution. Two sub-optimal algorithms were investigated that trade performance with computational complexity: the Multiple Model (MM) method and an echo-cancellation technique called Scrub Taps Waiting in a Queue (STWQ). A comprehensive MATLAB program was developed to simulate these algorithms and to investigate other aspects of a digital canceller, including quantization effects and performance in time-varying channels. A number of these simulation results are presented and interpreted. Finally, the results of the simulation are analyzed and directions for future research presented.
Bibliographical Information:

Advisor:

School:University of Cincinnati

School Location:USA - Ohio

Source Type:Master's Thesis

Keywords:adaptive filter noise cancellation cosite interference

ISBN:

Date of Publication:01/01/2002

© 2009 OpenThesis.org. All Rights Reserved.