Research On Opportunistic Spectrum Access In Cognitive Radio Network

With the exponential growth in wireless services, the contradiction between scarcity ofspectrum resource and under-utilization that the traditional approach of fixed radio spectrumallocation leads to is increasingly prominent. Opportunistic spectrum access based on cognitiveradio is to allow secondary users to exploit local and instantaneous spectrum opportunities withlimited interference to primary users, in order to improve the spectrum efficiency. Thisdissertation is under the background of the study on different primary network, and focuses onopportunistic spectrum access strategy in slotted primary network, unslotted primary networkand learning based spectrum access strategy, and realizes the simulation platform ofopportunistic spectrum access system. The main works and creations are as the following:1. A novel opportunistic spectrum access strategy based on energy efficiency is presented tosolve the energy constrain issue for secondary user in cognitive radio ad hoc network. Wedevelop an analytical framework to maximize energy efficiency based on partially observablemarkov decision process (POMDP) through jointly considering the time-variance characteristicof wireless channels and the accuracy of spectrum detection. The framework can lead secondaryuser to select the channel with the best energy efficiency and adaptively control the transmissionpower according to channel belief state, channel gain and detection probability. Simulationresults indicate that the proposed strategy can improve the energy efficiency of the secondaryuser and achieve the tradeoff between the transmission performance and energy overhead byefficiently regulating the transmission power.2. Aiming at the opportunistic spectrum access system in unslotted primary network,sensing activity of secondary user is modeled and analyzed with continuous time Markovprocess. We design an access strategy to control access probability of secondary user, whichanalysises the sensing duration and optimizes maximizes the energy efficiency of secondary userin the precondition of satisfying the collision constraints. The simulation result shows that, theaccess strategy can control the aceess probability to optimize the energy efficiency.3. The problem of channel selection for opportunistic spectrum access is considered. TheQ-Learning of the intelligent control theory is adopted, and a new Q-Learning based algorithmfor channel selection is proposed considering the channel selection model of secondary user. Thealgorithm can lead secondary user to select channels with maximum cumulative reward, andmaximize secondary user’s throughput. A Boltzmann learning rule is adopted to achieve welltradeoff between channel exploration and exploitation. From the simulation results, compared with random selection algorithm, the algorithm does not require prior knowledge or predictionmodels of the channel environment, yet can still select the optimal channel adaptively, improvethe secondary user capability.4. The simulation platform of opportunistic spectrum access system is designed and realizedwith the VS2010platform, and can simulate different primary network environment. Theperformance of opportunistic spectrum access strategy in slotted primary network is tested, andtest results show that the system simulation platform has better reliability and practicability.