Rate and Power Adaptation for Increasing Spectrum Efficiency in Cognitive Radio Networks

We propose rate and power adaptation strategies to optimize data transmission over fading channels in a spectrum sharing system operating under average received-interference constraint at the licensed user (primary user). Specifically, considering availability of the channel state information (CSI) pertaining to the secondary link and spectrum-sensing information about the activity state of the primary user, we investigate two adaptation policies at the secondary user's transmitter, namely, variable power and variable rate and power, over Rayleigh fading channels. The adaptation policies are obtained by maximizing the achievable capacity under said constraint and bit error rate (BER) requirements in multilevel quadrature amplitude modulation (M-QAM). We asses the benefits of using soft-sensing information about the primary user's activity on rate and power adaptation in spectrum sharing systems, and provide numerical results and comparisons illustrating performance for different operating scenarios.