Maxiumum likelihood receivers for nonlinear digital communication systems

Nonlinear distortion in digital communication systems was studied. Maximum likelihood receivers (ML) were derived to combat any single valued nonlinearity and the normally present additive Gaussian noise. The ML receivers were derived for (1) binary transmission and adjacent intersymbol interference (i.s.i.); (2) M-ary transmission and the case in which there is i.s.i. form L sub p previous symbols and L sub f future symbols; and (3) based on the Viterbi algorithm, minimizing the probability of error over the entire message sequence. Performance of the receivers was obtained through computer simulation. It is observed that for all the receivers considered, the quadratic distortion degrades the performance more than either the error function (soft limiter) or cubic distortion.