Title :
Design and analysis of low-complexity interference mitigation on vector channels
Author :
Müller, Ralf R. ; Verdu, Sergio
Author_Institution :
Vienna Res. Center for Telecommun., Austria
fDate :
8/1/2001 12:00:00 AM
Abstract :
Linear multiuser detectors for vector channels with crosstalk are approximated by weighted matrix polynomials. The weight optimization problem is overcome using convergence results from random matrix theory. The results are also extended to receivers with subsequent successive decoding. In the case of subsequent successive decoding, a novel low-complexity implementation is found for the first-order approximation that is based on matched filter banks only and does not require matrix algebra. Spectral efficiency is obtained analytically and found to be fairly close to the optimum. The paper is focussed on multiuser detection for CDMA, but the results can be easily extended to communication via antenna arrays
Keywords :
antenna arrays; channel bank filters; code division multiple access; convergence of numerical methods; crosstalk; decoding; filtering theory; interference suppression; least mean squares methods; matched filters; matrix algebra; multiuser channels; polynomials; radiofrequency interference; signal detection; spectral analysis; CDMA; MMSE multiuser detection; antenna arrays; code division multiple access; convergence results; crosstalk; first-order approximation; linear multiuser detectors; low-complexity implementation; low-complexity interference mitigation; matched filter banks; random matrix theory; receivers; spectral efficiency; successive decoding; vector channels; weight optimization; weighted matrix polynomials; Convergence; Crosstalk; Decoding; Detectors; Interference; Matched filters; Matrices; Multiuser detection; Polynomials; Vectors;
Journal_Title :
Selected Areas in Communications, IEEE Journal on
