GMSK is an efficient modulation technique for digital mobile communications. In this paper a simple solution to the modeling and simulation of GMSK systems impaired by fading channels and additive white Gaussian noise is given, and its effects on the channel characteristics are also considered. The modeling is based on the well-known linearized representation of phase-modulated signals. The analysis is not only useful for modeling and simulation purposes, but also gives a tool for the design of simple receivers.

A new tight analytical upper bound is presented for the performance of finite-delay symbol-by-symbol (SBS) Abend-Fritchman-like decoding of trellis-encoded data crossing arbitrary (eventually time-varying) discrete memoryless channels (DMCs). On the basis of this bound, a related criterion for the optimal design of trellis codes with SBS decoding is then proposed. It gives rise to an effective procedure for the construction of good trellis codes (generally time varying and nonlinear) with assigned decoding complexity.

The least square estimation problem for pinned-to-zero discrete-index reciprocal Gaussian processes in additive white noise is solved, thus completing and extending some previous results available in the literature. In particular, following the innovations approach a (finite) set of recursive equations is obtained for the filter and for the three standard classes of smoothers (fixed-point, fixed-interval, fixed-lag). Recursive expressions for the mean square performance of the proposed estimators are also given.>

The Marcum's Q-function is extended to complex arguments. In particular, a set of modified generalized Q-functions for imaginary values of one or both arguments are defined and their properties are investigated. An extended probabilistic interpretation for the modified Q-functions is given appealing to a proposed generalization of the concept of random variable. Complex pseudo-random variables, pseudo-events, and pseudo-probability measures are defined, which formally obey the usual rules of composition of probability measures of independent events. The modified Q-functions are then used to derive the Rayleigh multivariate multidimensional distribution function and the error probability for partially coherent detection of equicorrelated signals.>