Research Of Wideband Spectrum Sensing For Cognitive Radio Based On Flexible Detection Criterion

With the rapid development of wireless communication, a variety of new communication technologies and new communication services are emerging, and hence the shortage of spectrum resource is becoming increasingly prominent. To ease the contradiction between the growing demand for wireless communications services and the limited spectrum resources, increasing the utilization of the limited wireless spectrum resources has become a hot research topic.Cognitive radio, which can effectively increase the utilization of the spectrum resources because of its dynamic spectrum access capability, provides a new solution to the problem of spectrum shortage. As the key technology of the cognitive radio, spectrum sensing is the important prerequisite for spectrum allocation and spectrum sharing. In order to increase the spectrum utilization, the secondary users can find and use the spectrum holes by sensing the spectrum surroundings.This thesis first describes the basic concept and the current research status of cognitive radio, and analyzes several traditional wideband spectrum sensing algorithms. Aiming to enhance the current wideband spectrum sensing schemes, the improved forward consecutive mean excision based wideband spectrum sensing algorithm and flexible detection criterion based wideband spectrum sensing algorithm were proposed:1. Compared with the traditional energy detection based wideband spectrum sensing, the forward consecutive mean excision based wideband sensing algorithm can effectively solve the problems of demanding a prior knowledge of the noise power and the poor detection performance under the low SNR circumstance. Aided by the knowledge of signal contiguity, frequency bin combination, cluster detection and binary integration were proposed to improve the forward consecutive mean excision algorithm. Theoretical analysis and simulation results show, the improved algorithm inherits the advantage of low algorithm complexity. Additionally, the improved algorithm can effectively control the false-alarm rate and improve the detection performance.2. In the wideband spectrum sensing scheme based on signal set cardinality estimation, the information theoretic criterion always underestimate the number of the subbands occupied by primary users in the licensed frequency band, which usually deteriorate the detection performance of spectrum sensing. In order to solve this drawback, flexible detection criterion based wideband spectrum sensing algorithm was proposed for cognitive radio to improve the traditional information theoretic criterion. Theoretical analysis and simulation results also illustrate, the proposed algorithm can not only effectively enhance the spectrum sensing performance, but can also adjust the flexible parameter to meet the requirements in various sensing scenarios.