Robust multiuser detection based on probability constrained optimization of the MMSE receiver

The performance of multiuser detection (MUD) algorithms for code- division multiple-access (CDMA) systems depends on the accuracy of channel estimates. Such estimates are typically affected by errors, which can lead to significant degradation of the performance. In this paper, we develop a MUD technique which is based on probability constrained optimization of the minimum mean-square error (MMSE) multiuser receiver, and is robust against channel estimation errors. Its relationship to the recently proposed robust worst case optimization based MUD technique is established. The uncertainty parameter of the worst case based design is quantified in terms of the outage probability used in the probability constrained design and second-order statistics of the channel estimation errors. Simulation results demonstrate the potential of the proposed technique to outperform the existing robust techniques.