Adaptive antenna array system for wireless communication systems using spatial-temporal high order statistics filtering

Radio frequency propagation characteristics in wireless digital radio communications directly impact the performance of the system in terms of capacity. When channel distortions are severe, countermeasures should be taken in order to limit link degradations at tolerable levels. One of the well known methods of increasing the reliability of a wireless system is the use of spatial diversity. The article analyzes a K-th order space diversity system for digital transmission over a frequency selective fading channel. If the K channels can be modeled as arbitrary impulse responses it is possible to state an alternative criterion for restoring signals at the receiver. The usual approach (the minimum mean square error criterion) to this problem is in fact based on SOS (second order statistics) of the received signal: but there is evidence that this method suffers from large degradations in particular channel conditions despite its relative simplicity. We propose the application of the multichannel generalization of the deconvolution method presented by Shalvi and Weinstein (see IEEE Trans., Information Theory, vol.39, p.504-519, March 1993).