Enhancing reliability of real-time traffic via cooperative scheduling in cognitive radio networks

In this paper, we study the cooperation model to design and develop a system that improves the reception rate of delay sensitive packets in realtime applications. While making a cooperation decision, a user with real-time traffic must consider the queue size, packet deadline, the traffic rate, channel condition. Existing cooperation models do not take these factor into consideration and therefore, are not applicable to the improvement of reliability of real-time traffic. We formulate the cooperation model using a Markov decision process (MDP) which is NP-Hard. The MDP-based optimization problem of a single pair of primary and secondary user is analyzed to find the optimum transmission decision in different states of the network. Based on the findings of a single pair case, we develop a distributed cooperation algorithm where a user (primary or secondary) quantifies the immediate and future impact of cooperation and takes its decision accordingly. Extensive simulation is performed to evaluate the performance of the proposed cooperation algorithm which reveals its efficacy compared to the non-cooperative scheme.