Joint maximum likelihood approach in overloaded array processing

Abstract

We demonstrate that, signal extraction in overloaded environments is achieved by employing a nonlinear maximum-likelihood detector. We define the overloaded interference scenario as one where the number of near equal-power co-channel interferers exceeds the number of antenna array elements. The iterative least squares with enumeration (ILSE) technique is a maximum likelihood joint detector that provides an optimum solution for the overloaded antenna array scenarios in the presence of synchronous cochannel users in an AWGN channel. A linearized version of ILSE, called iterative least-squares with projection (ILSP) is shown to fail in overloaded environments due to an implicit beamformer assumption. ILSE is shown to extract signals with very small angular separations in highly overloaded scenarios at reasonable SNRs. We also show that when the channel is a priori unknown, ILSE can still demodulate more users than elements by incorporating short training sequences for channel estimation as well as employing a completely blind channel estimation technique.