Adaptive Spread-Spectrum Systems Using Least-Squares Lattice Algorithms

Abstract
Practical communication systems must cope with many uncertainties in addition to determining the transmitted data, e.g., the direction, timing, and distortion of the desired signal, and the spectral and spatial distribution of the interference, all of which may change with time. This paper describes exact least-squares (LS) recursive lattice algorithms which resolve these uncertainties in a direct-sequence spread-spectrum digital communication system. The adaptive LS algorithm is recursive beth in order and time, and converges rapidly to the uncertain parameters. Time-discrete algorithms may be mechanized by a receiver containing integrate-and-dump circuits operating at the chip rate of the pseudonoise (PN) sequence, one in each in-phase and each quadrature channel of each sensor array element's output. Different configurations of optimal time-discrete receivers are presented and transformed into adaptive receivers by taking advantage of the spectral properties of the different kinds of LS filters. Simulation results are presented and some guide lines are given for the architecture of an adaptive direct-sequence spread-spectrum system.