Joint routing and link scheduling for cognitive radio networks under uncertain spectrum supply

The essential impediment to apply cognitive radio (CR) technology for spectrum utilization improvement lies in the uncertainty of licensed spectrum supply. In this paper, we investigate the joint routing and link scheduling problem of multi-hop CR networks under uncertain spectrum supply. We model the vacancy of licensed bands with a series of random variables, and introduce corresponding scheduling constraints and flow routing constraints for such a network. From a CR network planner/operator's point of view, we characterize the network with a pair of (α; β) parameters, and present a mathematical formulation with the goal of minimizing the required network-wide spectrum resource at the (α; β) level. Given that (α; β) is specified, we derive a lower bound for the optimization problem and develop a threshold based coarse-grained fixing algorithm for a feasible solution. Simulation results show that i) for any (α; β) level, the proposed algorithm provides a near-optimal solution to the formulated NP-hard problem; ii) the (α; β) based solution is better than expected bandwidth based one in terms of blocking ratio as well as spectrum utilization in CR networks.