Research On The Spectrum Sensing And Resource Allocation In Cognitive Radio Networks

As the enable technologies, the spectrum sensing and resource allocation promote the cognitive radio (CR) wide application, which has attracted great research attention. This dissertation is supported by the State 863 projects and other fundings. The dissertation contains valuable work on both theories and applications.Based on the deeply researching on the principles and progress of cognitive radio, the following contributions and innovative ideas are achieved:Considering the spectrum sensing problems over fading channels in cognitive radio networks, the average sensing performance is evaluated under the assumption that the number of secondary users (SUs) is distributed according to a homogeneous Spatial Poisson process. Based on the derived average probabilities of detection and false alarm, the cumulate interference of primary user (PU), the outage probability at the PU's receiver, two operative methods are presented based on minimizing the total spectrum sensing error and maximizing the average opportunistic throughput respectively. By optimizing the sensing time, the proposed methods enhance the spectrum sensing performance.A novel limited wideband sensing capability model is proposed based on the constant bit rate (CBR) traffic of the SU. The joint optimization of subchannel selection and spectrum sensing time for multiband under the sensing capability constrains is developed. Moreover, a semi--analytical optimization scheme is constructed to achieve the optimal solution.For the power allocation problem in CR networks, a joint spectrum overlay and underlay access model is proposed. Two efficient power allocation schemes are developed aiming at maximizing the system utility and the achievable throughput respectively while satisfying the QoS requirement of SUs and interference constrains of the PU. A modified optimal power allocation scheme is formulated specifying the case where the sensing error is not zero. A QoS sensitivity based admission control algorithm is developed to deal with the case that the optimal power allocation scheme is infeasible.Considering the spectrum sharing problem, a game-based distributed spectrum offering scheme with dynamic topology is proposed. Moreover, an iterative distributed spectrum offering algorithm for the case when the topology is rapidly changing and the PUs can not observe the current action and utility adopted by others is proposed. The iterative algorithm holds the characteristic of fast convergence.Finally, the content of the whole dissertation is summarized, and several valuable research directions of cognitive radio are discussed.