SUM: Spectrum Utilization Maximization in Energy-Constrained Cooperative Cognitive Radio Networks

Cooperative cognitive radio networks (CCRNs) enable secondary users (SUs) to access primary resource by cooperation with active primary users (PUs). For the cooperation-generated resource, existing schemes in CCRNs allocate the resource only to the relay SUs. However, this may lead to inefficient spectrum utilization, when the relay SUs have poor channel condition or little traffic load for their own secondary transmissions. In this paper, considering user diversity in secondary networks, we focus on network-level throughput optimization for secondary networks, by allowing all SUs to optimally share the cooperation-generated period. Besides, considering the energy constraint on SUs, we formulate the resource allocation problem from long-term perspective, to reflect the time-varying change of user diversity in channel condition, traffic load and energy amount. We present an online SUM scheme to solve the long-term optimization problem. Although a mixed-integer and non-convex problem is involved in the SUM scheme, we transform the problem into multiple convex subproblems, and then optimally solve it with low computational complexity. Extensive simulations show that the proposed SUM scheme significantly outperforms the existing schemes.