A simulation study of space diversity and adaptive equalization in microwave digital radio

In this paper we analyze the performance of M-level quadrature amplitude modulation digital radio systems subjected to microwave multipath fading. We consider two kinds of adaptive receiver techniques, either singly or in combination: dual space diversity and adaptive equalization. The space diversity is assumed to be of either the selection type or the continuous-combining type, and the equalization is assumed to be ideal. We describe a specific form of combining which is optimal when no post-combiner equalization is used. A primary aim of the study is to quantify the performance of this combining approach and to compare it with alternate strategies. The study uses Monte Carlo simulations of the dual-channel fading response functions based on a recently published statistical model. For each response pair generated, a receiver detection measure is derived analytically in terms of the system parameters and receiver approach. Probability distributions of this measure, obtained by simulating several thousand response pairs, are then computed. They can be interpreted as displaying the link outage probability as a function of the number of modulation levels (M). We find that the appropriate combining scheme can serve in some cases to avoid the need for adaptive equalization. Also, where post-combiner equalization is used, the same scheme, while no longer optimal, can sharply reduce the dispersion seen by the equalizer input.