Performance Analysis And Optimization For Spectrum Sensing In Cognitive Radio

With the rapid development of wireless communication technologies, the demand for the radio spectrum increases significantly. Most of the available radio spectrum has been assigned for various services, and it is hard to find available spectrum for new communication systems. On the other hand, some measurements show that the licensed spectrum is rarely utilized continuously across time and space. Therefore, spectrum scarcity is largely due to inefficient fixed frequency allocations rather than any physical shortage of spectrum. Cognitive radio is an intelligent wireless system with the function of dynamic spectrum access. It is able to opportunistically access the primary channels and improve the utilization of the allocated spectrum, and alleviate the issue of spectrum scarcity.The key technologies of cognitive radio include three categories in general: spectrum sensing, spectrum coexistence and spectrum management. Among them, spectrum sensing is an important task for the realization of CR networks. The main objective of spectrum sensing is to provide more spectrum access opportunities for secondary users without interfering with the operations of the licensed network. In the dissertation, performance analysis and optimization for spectrum sensing is addressed, and the main achievements and results are listed as follows:1. The performance of cooperative spectrum sensing with equal gain combination over Nakagami-m channels is investigated. By using the equivalent representation of generalized Marcum Q-function and integrals of some special functions, the series form of detection probability is derived, where the fading factor can be any number not less than 0.5. In the simulation results, the correctness of the theoretical analysis results are validated, and a group of practical parameter settings are given.2. Taking periodic spectrum sensing for search background, throughput optimization problem is formulated in cognitive radio networks based on spectrum sensing for single primary channel under the constraint that the primary users are sufficiently protected. Three detection methods are considered: traditional data fusion detection, sequential detection for single secondary user and sequential cooperative detection for multiple secondary users. The existence and uniqueness of the solutions are proved by optimization theory, and the corresponding optimization algorithms are given. In the simulation results, the maximum throughput and the corresponding parameters are given under different simulation parameters. It can be seen from the simulation results that the achievable throughputs of the cognitive radio networks based on sequential detection are larger, but it requires making more decisions. In practice, the detection algorithm should be chosen according to the actual situation.3. The issue of parameter setting in cognitive radio is addressed when the primary communication system is a multi-carrier modulation system. In this case, the received signals at secondary users are converted to samples in frequency domain, and then energy detector is used at each subchannel. In the dissertation, single user detection and multi-user decision-fusion detection are considered, respectively. The models of the detectors are given firstly, then the throughput optimization problem is formulated and the optimization algorithm is given. In the simulation results, the correctness of theoretical analysis is verified.4. When there exists multiple primary channels in cognitive radio networks, the secondary user senses the primary channels sequentially. A particle swarm optimization based method is proposed to optimize the sensing duration for each primary channel to maximize the throughput. The problem is formulated as a constrained optimization problem, and the sensing durations are optimization parameters. The detailed specifications on each step of the procedure and the process of dealing with the constraints are given. The simulation results show that the proposed method is stable and efficient and the optimal achievable throughput of the method is much larger than that of the existing ones.