Layered space-time MUD with adaptive group order successive interference cancellation

This paper proposes a simple, yet effective layered space-time (LAST) multiuser detection (MUD) with adaptive group order successive interference cancellation (AGOSIC). The new approach sequentially peels off the detected signals group-by-group in a descending order of power levels, and, most noticeable, the number of users in each group is adaptively determined based on the average of the post signal to noise ratio (SNR) to achieve a better ordering of signals. More specifically, according to the threshold, which is stage dependent, the remaining signals in each stage are partitioned into two groups: the desired group and the residual group. After conducting the layered space-time processing scheme, the signals in the desired group, whose SNRs are all greater than or equal to the threshold, are then removed. The whole process is repeated until all users are detected. Furnished simulations show that the performance of the proposed approach is very close to, and sometime even superior than, the LAST-MUD but with lower computational complexity and latency.