On Opportunistic Spectrum Access Theory And Application

Opportunistic spectrum access (OSA) has emerged as one of the exciting technologies in the future cellular networks, and has drawn much attentions from academic and industrial circles. In the OSA-based wireless network, the cognitive user (CU) is allowed to operate in the frequency bands originally allocated to the licensed user only when the licensed user is absent. However, when the licensed user comes back into operation, the CU should vacate the spectrum instantly to avoid interference with the licensed user.This thesis studies the sensing scheduling problem in the OSA-based CRN, and inves-tigates the application of the OSA in the cellular networks and femtocell networks, respec-tively. The contents of this thesis are listed as follows.The sensing scheduling problem in the OSA-based wireless network is studied. We consider a CRN which coexists with multiple licensed users, and CUs transmit over time-varying channels. We propose a novel prediction-based MAC-layer sensing algorithm. In the proposed algorithm, the CUs’channel quality information is predicted by the Markov chain, and the probability of the licensed channel being idle is predicted by employing the traffic parameters of the licensed users. Through the above predicted information, we schedule the CUs to sense and transmit on different licensed channels. Specifically, multiple significant factors, including network throughput and fairness, are jointly considered in the proposed algorithm. Then, we formulate the prediction-based sensing scheduling problem as an optimization problem, and solve it with the Hungarian algorithm in polynomial time. The proposed prediction-based sensing scheduling algorithm could achieve a good tradeoff between network throughput and fairness among CUs.The analysis of grade of service (GoS) in the OSA-based cellular networks is stud-ied. Firstly, we present the network framework of OSA-based cellular networks. According to the non-hopping and hopping modes, two OSA strategies are proposed. Then, a three dimension (3D) continuous-time Markov chain (CTMC) is constructed to model the OSA strategies, and we derive the GoS performance metrics in the OSA-based cellular network-s. By employing the GoS parameters, the spectrum utilization of the OSA-based cellular networks is studied. Moreover, for a given licensed user traffic, the range of achievable CU traffic is derived with guarantying the GoS of the CU.The incentive mechanism for hybrid access OSA-based femtocell is studied. We propose a novel dynamic spectrum allocation method for the hybrid access OSA-based femtocell. In the proposed method, the macro base station allocates a portion of subchan-nels that originally allocated to the macro users (MUs) to the femto base station (FBS) to spur the femtocell to serve the MUs. A protocol is designed to implement the proposed method. Then, we formulate the corresponding resource allocation problem as a sum utility maximization problem, and propose an optimization method to solve it via the dual decomposition method. Both the wireless operator and the femtocell could benefit from the proposed mechanism. On one hand, the wireless operator saves a portion of subchannels while guaranteeing the performance of the MUs. On the other hand, the FBS obtains more spectrum resource to improve the performance of the femto users.