The Fourier series implementation of a locally optimum detector

Reception of signals in the presence of jammers is of paramount importance in military spread spectrum(SS) communications systems. Locally optimum (LO) detection is a method which is particularly suited to noisy environments characterised by large jammer-to-signal ratios (J/S). Interest has focused on robust techniques, in which the required nonlinearity is formed from an estimate of the noise probability density function (PDF). This paper presents a novel method of implementing a robust LO detector which does not require an intermediate estimate of the interference PDF. Instead, an algorithm based on the Fourier series is used to approximate the nonlinearity directly. Furthermore, an analytical expression for the signal-to-noise ratio (SNR) gain at the output of the detector is derived. Finally, the detection of a direct sequence (DS) SS signal in Cauchy noise is presented as an example of the performance gains provided by this method.<
>