Citation

Abstract

Techniques to reduce the effect of co-channel interference (CCI), intersymbol interference (ISI) and noise in a time-division multiple access (TDMA) radio cellular radio system will allow an increase in network capacity without a loss in quality of service.

Two new analytical results have been derived and presented. The first result is the probability of error of a method for blind maximum likelihood (ML) data and channel estimation in the presence of noise. The second result is the mean square error of a displaced linear equalizer with infinite length on a time-varying channel in the presence of additive white Gaussian noise. These results provide measures of effectiveness of the blind maximum likelihood data and channel estimation method and the displaced linear equalizer used in the receiver.

In order to achieve good system performance in a fast time-varying environment, a novel block adaptation with blind channel estimation strategy has been developed. It involves combinations of channel estimation and interpolation. With this strategy, in addition to the channel estimates obtained from known sequences in a TDMA time slot, channel estimates are obtained using small sequences of unknown data and the ML blind data and channel estimation method. The time-varying channel during a time slot can then be reconstructed by interpolating these channel estimates. Subsequently, the equalizer coefficients can be determined by using the Wiener-Hopf equations. A similar strategy can be applied to deal with time variations in the CCI. The effectiveness of this novel block adaptation strategy has been proven in the IS-136 application.