(The) BER performance improvement for IDMA systems in overloaded environments

Abstract

This thesis proposes a modified multiuser iterative detection for synchronous uncoded IDMA system using the scaling factor at the receiver. The existing IDMA receiver feeds back the extrinsic log-likelihood ratios (LLRs) to multiuser detector to improve its estimates in the next iteration. In overloaded scenarios especially where the number of users is greater than the maximum number of users supported in fixed spreading length, the degradation of the performance is caused by increased interferences. However, the proposed iterative multiuser detection provides better estimates based on the extrinsic LLRs manipulated by the scaling factor. Simulation results show that the proposed scheme compensates the degradation of BER performance of IDMA system without additional iterations, compared to the existing iterative multiuser detection scheme.