Opportunistic spectrum access based on channel quality under general collision constraints

This paper studies the problem of cognitive access in a hierarchical network with practical time varying channels. We propose an access policy which is not only based on the sensing outcomes about the occupation states of the channel, but also based on the channel quality situations. The contribution in this paper is as follows. First we propose an access policy which takes advantages of the characteristics of the channel and thereby fully utilizes the access opportunities that channel could offer. The optimization objective of cognitive access is designed to maximize the effective throughput subject to collision constraints imposed by channels. It is shown that our proposed algorithm can achieve high throughput performance by grasping the access opportunities which have good channel qualities. Second we extend the throughput region achieved by secondary users under general collision constraints. Specifically, our algorithm can achieve better throughput performance not only when collision constraints are tight, but also under loose collision constraints, which has not been studied in previous related work. Besides, We also give the practical algorithms to implement our policy, which makes it easily to be deployed in practice. In order to validate the theoretical analysis, numerical results are presented to show that by introducing the proposed policy, which can fully utilize the characteristics of application scenarios and take the advantages of access opportunities of the channel when it is of high quality, the effective throughput can be improved significantly under general collision constraints.