Performance analysis of a subspace algorithm for cochannel TDMA signals

Abstract

We analyze the performance of a subspace algorithm that separates cochannel TDMA signals using an adaptive antenna array. The algorithm processes a block of data in two passes. In the first pass, the weight vector is constrained to lie in a subspace that contains the direction vector of the signal of interest and is orthogonal to the subspace spanned by the direction vectors of the interferers. In the second pass, this weight vector is modified by projecting it onto an appropriate subspace determined by the cochannel interferers. The analysis includes expressions for the beamformer weights and the mean-square error in terms of the signal direction vectors. These results allow one to numerically evaluate the performance of the subspace algorithm for different cochannel scenarios.