Research On Parameter Optimization And Framework Design Based Spectrum Sensing

With the rapid development of various radio technologies and services, the existing radio spectrum has been unable to meet the needs of users. According to relevant research datas, the current spectrum utilization is very low. There are a lot of "spectral holes." Therefore, cognitive radio gradually becomes one of the hottest areas of communication research. The cognitive radio include four key technologies: spectrum sensing, spectrum management, spectrum sharing and spectrum mobility. Among them, spectrum sensing is an important prerequisite for the realization of cognitive radio. The main objective of spectrum sensing is to obtain the authorized channel state information. Based on the information, the secondary users can decide whether access to the authorized channel or not。Spectrum sensing is to provide the spectrum access opportunities for secondary users without interfering with the primary users. In order to provide the primary users, the accuracy of spectrum sensing must be high, namely, the spectrum sensing should have good sensing performance. For secondary users, they hope to gain more opportunities to access the authorized channel so as to maximize the throughput of secondary users. With the current problem of global warming is worse, green communication is getting more and more attention, so the energy consumption is gaining more and more concern. So, the energy efficiency, which associated with the energy consumption and throughout, gradually attracts the scholars’ interest. In this thesis, researches on the aspects of perception performance, energy efficiency and throughput have been carried out.In this thesis, a method of joint optimization of sensing performance and energy efficiency is proposed. In the cognitive radio networks using "K-rank" criteria, the parameters which affects system performance may be the threshold, the number of secondary users and the sensing time. So, the process of solving the optimization problem can be divided into two stages. In the first one, given the sensing time, the threshold and the number of secondary users are optimized by minimizing the false probability. In the next, given the threshold and he number of secondary users, the sensing time is optimized by minimizing the energy efficiency. Two optimization problems are proved to have the optimum value by employing the mathematical derivation. The optimal threshold, number of secondary users and sensing time can be obtained through the joint iterative algorithm.In addition, aiming at the optimization problem of throughput, a new frame structure is designed based on spectrum prediction and continuous spectrum sensing. Firstly, the band of channel is divided into two subbands, then we add the spectrum prediction that based on hidden Markov in the subband. The secondary users only choose the sensing channel which is predicted to be free, so as to improve the probability of sensing the idle spectrum. And continuous information transmission is carried out in the other subband to improve the throughput of the system.