Joint Routing and Spectrum Selection for Multihop Cognitive Radio Networks

We present a new routing paradigm for cognitive radio networks that computes paths from a source to a destination by considering t he activity and the placement of primary users. Our algorithm makes use of the possibility of exploiting multiple frequencies in the same time between two cognitive nodes. For a given capacity demand ”D” from the application, our protocol tends to find the most probable path that satisfies this demand. This is done by accounting the primary nodes activity pattern and their positions distributions to choose the path that has the highest probability to satisfying this demand. Then the algorithm will then add frequencies along the path until the capacity on every link is greater than the demand. We show through simulation and numerical results that the found most probable path is the best path from the source to the destination. Furthrmore, the average capacity calculated over that path is highly accurate which can be considered as a positive indicator on satisfying the application demand D.