Bargaining-based Spectrum Sharing for Cognitive Radio Networks With Incomplete Information

To improve the utilization of spectrum resource, a primary network can share its unused bandwidth with a closely located secondary network. As a compensation, the latter should pay a certain price for the obtained bandwidth. To determine this price under the constraint that the primary network only knows an incomplete information for the utility of the secondary network, the present paper consider the repeated offering of a price from the primary network to the secondary network, which then can accept or reject it, until they reach a price agreement. Since this bargaining process helps the primary network estimate the real utility of the secondary network as accurate as possible, the agreed price is mutually beneficial for both networks. Moreover, to increase the satisfaction of all secondary terminals in transmitting heterogeneous applications, i.e., voice, video, image, and data, we also propose a spectrum allocation solution within the secondary network based on Nash bargaining model. Numerical simulation shows that the benefits of the two networks from spectrum sharing varies with their maximal bargaining times and the proposed spectrum allocation scheme can improve the total satisfaction of all secondary terminals while maintain a reasonable fairness among them.