<title>Equalization for wireless optical communications over multipath channels</title>

Abstract

We review the advantages of digital pulse-position modulation (PPM) for wireless infrared communications, and then examine several equalization strategies for detecting PPM signals in the presence of intersymbol interference and additive white Gaussian noise. Our approach is based on an equivalent vector channel, obtained by viewing PPM as a binary block code, and we use this vector channel to derive the ML sequence detector. We also propose three sub- optimal block-by-block detectors, all of which use block decision-feedback equalization (DFE) to mitigate inter-block interference but differ in how they mitigate intra-block interference: the ML-BDFE accounts for it in an optimal way, the comparator-BDFE uses linear equalization, and the correcting-scalar-DFE uses DFE. We illustrate low-complexity implementations of these equalizers, and compare their performance to the ML detector using Monte-Carlo simulations.