Dynamic Spectrum Access in Underlay Cognitive Radio System with SINR Constraints

How to improve the spectrum utility of the unlicensed uses (secondary users) without interfering the licensed spectrum holders' usage becomes a key issue in the cognitive radio system. In this paper, we consider an underlay cognitive radio system in which the secondary users can make use of the allowance of the primary users' SINR constraints and operate on the same spectrum band as the primary signals. The dynamic spectrum occupation process of this underlay cognitive radio system is modeled by continuous-time Markov chains. Moreover, appropriate access probabilities for the secondary users are adopted to decrease the mutual interference and coordinate the spectrum access among the secondary users. Under three kinds of optimization criteria, the performance of the underlay cognitive radio system is analyzed. The simulation results show that the access opportunities of the secondary users in the underlay cognitive radio system, compared with the overlay system, can be enhanced considerably, and thus the overall throughput of the secondary users can be improved greatly. Also, the proportional fairness criterion used in this system can not only provide more fair spectrum sharing among secondary users than the maximal-throughput criterion, but also achieve much higher throughput than the max-min criterion.