Tracking and transient performance of code-constrained blind adaptive MUD's

Abstract

This paper derives the closed form expressions for the excess mean square error (EMSE) of code constrained blind adaptive multiuser detectors (MUD) in a non-stationary channel, using the feedback approach. The tracking performance of those algorithms is thus quantified. Closed form expressions are derived to quantify the transient performance of the adaptive algorithms. It is shown that the exact time evolution of the signal to interference plus noise ratio (SINR) may be predicted from initialization to steady-state. The accuracy of the analysis is confirmed via simulation.