Joint Transmitter / Receiver Optimization for Cognitive Radio Applications

Abstract

The increasing demand for high data rate services is a primary motivation for the development of next generation mobile radio networks. As such, efficient utilization of the available radio spectrum is an important objective to consider. Conventional static frequency allocation schemes are unable to accommodate the increasing number of users. The scarce bandwidth availability thus necessitates some form of spectrum sharing between existing and new users. One such spectrum sharing paradigm is cognitive radio, which offers a highly flexible alternative to the conventional single frequency band wireless devices. Cognitive radio is based on the concept of dynamic spectrum sharing in which the new users use the radio channel only when it is not in use by the primary license-holder user. As such, multiple users may be accommodated within the common frequency band. In this thesis, we examine an alternative paradigm to cognitive radio in which new (overlay) users can operate simultaneously with existing (legacy) users. This may potentially unveil an even more efficient utilization of the shared spectrum. However, this mode of operation must contend with mutual interference caused by potentially simultaneous transmissions by the two users. The design of the optimal transmitter and receiver of the overlay system should thus attempt to mitigate the interference received from the legacy user and also the interference caused by the overlay system to the legacy user. A weighted sum of the mean-squared error (MSE) of the overlay system plus the excess MSE in the legacy system due to the introduction of the overlay system is therefore used as a figure of merit. The effects of varying key parameters such as the overlay transmitter power and the amount of channel overlap between the legacy and the overlay systems are investigated. The sensitivity of the system to accuracy of the channel estimate and signal-to-noise ratio (SNR) estimate is also examined. In addition, to reduce the complexity, a system having a fixed transmitter or receiver pulse shape has been studied and the optimum receiver or transmitter for that has been designed, and its performance has been compared with the joint optimization case. Finally, a dual-transmit-antenna overlay system has been proposed and its incurred gain compared to the single-transmit-antenna case has been studied. Simulation results show that using single-transmit-antenna system, 10 dB improvement in the system MSE can generally be achieved by the proposed optimization technique. Additionally, using dual-transmit-antenna system can reduce the system MSE by further 6 dB. Search Terms-Cognitive radio (CR), mean square error (MSE), crosstalk, overlay system, transmitter/receiver optimization