On correlation matching approaches to multipath parameter estimation in multiuser CDMA systems

In this paper, correlation matching techniques are employed to estimate multipath parameters for a multiuser CDMA system. We first parameterize the receiver's output correlation matrix by different unknown channel parameters for all users. Then we match the correlation matrix against its sample average estimated from the received data. The Frobenious norm of the resulting error matrix is minimized to simultaneously obtain a closed form solution for all channel-parameters. Batch as well as adaptive algorithms are developed. The identifiability of the proposed technique is studied. Finally, some simulation results are presented to verify the proposed methods.