Effect of system imperfections on the performance of CDMA receivers with CCI cancellation
Panicker, P. C. John
Code Division Multiple Access (CDMA) is an emerging technology for future cellular and personal communication systems. Indoor radio communication is an integral part of personal communication systems. In an indoor CDMA system, the Co-Channel Interference (CCI) from other users of CDMA limits the system capacity and Bit Early Rate (BER) performance. Regeneration and cancellation of CCI from the received signal is an approach that improves capacity and BER performance. CCI cancellation can be performed in single path correlator receiver and multipath combining RAKE receiver. In practice, there are system imperfections and their effect on the receiver performance has to be accurately evaluated. The power control error is modeled as a log-normal random variable. The estimate errors in channel parameters and spreading code phase are modeled by zero mean Gaussian random variables. In this thesis, analysis is done initially on the effect of system imperfections on the BER performance of a CDMA system having correlator receiver and multipath combining RAKE receiver. Numerical results indicate that the BER performance degrades when the variance of normalized received power is above 0 dB, or the mean square of normalized estimate errors is above 0.01. The effect of system imperfections on the performance of CCI canceling correlator and RAKE receiver is then analyzed. Numerical results indicate that significant improvement in system capacity is possible with two or three stages of CCI cancelers if the variance of normalized received power is below 2 dB, or the mean square of normalized estimate errors is below 0.01. However, for CCI cancellation to be effective when all imperfections are present, the power control error and the estimate error have to be minimized. Computer simulations of indoor CDMA system are also performed to verify theoretical results. It is concluded that the simulated results are in agreement with the theoretical results for all practical values of signal to noise ratio. Simulation techniques based on importance sampling can be used to reduce considerably the computational burden involved in simulation of CDMA systems.