Signaling constellations and signal detection for satellite mobile communications

The goal of this thesis is to design optimal constellations and efficient receivers for types of hostile nonlinear fading satellite mobile channels.;In the first part of the thesis, the issue of signaling constellations in nonlinear satellite fading channels is investigated. M-ary QAM (M > 4) constellations are potential candidates for modulation schemes of next generation digital communications. To find optimal constellations in mobile satellite communications, we present the exact symbol error rate (SER) and total degradation (TD) performances for different M-ary QAM signals over nonlinear fading diversity channels. This approach uses the same SER expression as in [Yuan, 2003b] and considers the variation of the system parameters.;The second part of the thesis focuses on signal detection. We propose a receiver that combines neural network (NN) technology and traditional fading estimation methods to identify the nonlinear fading channel. Simulation results in terms of mean square error (MSE) and SER are presented. Both bandwidth efficient 16-QAM and 64-QAM signals are considered for the transmission. (Abstract shortened by UMI.)...