Accession Number : ADA321811
Title : Quasi-Optimal Processing in Spread Spectrum Environments.
Descriptive Note : Final rept. Apr 93-Apr 94,
Corporate Author : ILLINOIS INST OF TECH CHICAGO DEPT OF ELECTRICAL AND COMPUTER ENGINEERING
Personal Author(s) : Ucci, Donald R. ; Jacklin, William R. ; Tanas, John
PDF Url : ADA321811
Report Date : DEC 1996
Pagination or Media Count : 89
Abstract : The derivations of robust Locally Optimum (LO) detection algorithms (high interference/signal ratios) with and without memory, and their performance in a spread spectrum communications system, are the subject of this report. There are three forms of LO detection considered: (1) detection with memory; (2) detection in independent and identically distributed (iid) noise; and (3) memoryless detection. Several algorithms are investigated, all using the usual assumption that the small signal corrupting the noise estimate is acceptable. Simulation results show, however, that this assumption is not always valid, even in high interferencel signal ratio environments. Detailed investigations focused on regions of poor performance, using ideal signal-free estimates of the noise as a baseline. The ideal estimator yielded a truly robust detector over a wide range of scenarios. Recommended research should thus be focused on means for obtaining true noise estimates in practical environments.
Descriptors : *SIGNAL PROCESSING, *SPREAD SPECTRUM, ALGORITHMS, COMPUTERIZED SIMULATION, OPTIMIZATION, COMPUTER COMMUNICATIONS, COMMUNICATIONS TRAFFIC, MULTIVARIATE ANALYSIS, SIGNAL TO NOISE RATIO, PROBABILITY DENSITY FUNCTIONS, NONLINEAR SYSTEMS, COMMUNICATIONS NETWORKS, NOISE REDUCTION, ANTIJAMMING, PHASE SHIFT KEYERS.
Subject Categories : Radiofrequency Wave Propagation
Distribution Statement : APPROVED FOR PUBLIC RELEASE