An adaptive space-time DFE combined with successive interference cancellation

We investigate the use of a multi-element antenna array (MEA) at the base station (BS) in a time division multiple access (TDMA) broadband fixed wireless access (BFWA) system. At high data rates the intersymbol interference (ISI) induced by dispersive channels becomes a technological bottleneck. Space-time processing at the BS offers a solution to improve the wireless link quality in the uplink. First, we investigate the performance of a space-time decision feedback equalizer (DFE) trained with the LMS algorithm for a single user BFWA system (i.e. one user per frequency channel and per time slot). Then we propose an algorithm for space division multiple access (SDMA) in the uplink based on a combination of adaptive space-time equalization and successive interference cancellation (SIC). It is shown via simulations that this combined approach offers an improvement in the probability of symbol error (P_s) for a two-user system (i.e. two users per frequency channel and per time slot) compared to a space-time DFE without SIC. The RLS algorithm is used for the two-user system to ensure fast convergence. The proposed iterative approach offers a low complexity solution compared to minimum mean square error solutions based on direct matrix inversion, which would be costly for the expected delay spreads associated with BFWA systems.