Channel Selection in Multi-Channel Opportunistic Spectrum Access Networks with Perfect Sensing

We study optimal transmission strategies in multi-channel cognitive radio networks where one secondary user (SU) opportunistically accesses multiple orthogonal channels that are owned and utilized by primary users (PU). In dynamic spectrum access networks, the protection of PU's is vital, since no PU would accommodate SU access to its own detriment. Therefore the objective of the problem we study is to maximize the SU throughput while protecting PUs on all channels. At a given time, the SU decides if it transmits and if so on which PU channel it should transmit on in order to protect PU performance. We use a constraint on the expected PU packet collision probability as the protection metric. We consider a general setting where the PUs are unslotted and may have different idle/busy time distributions and protection requirements. Under general idle time distributions, we determine the form of the SU optimal access policy. We also study the special case where PUs have independent, exponentially distributed idle time. For this case, we formulate a linear program that yields an optimal randomized strategy for the secondary user, and also present a tunable heuristic which allows for a tradeoff between complexity and throughput performance.