Cooperative Spectrum Sensing Algorithm Based On Sequential Detection In Cognitive Radios

The scarcity of radio spectrum resources seriously hampers the development of wireless communication technology in the future, introduced cognitive radio to improve spectrum utilization is an effective way to solve this problem, it has become the next generation of wireless communications network core technology. As the foundation for realization of cognitive radio, cooperative spectrum sensing can overcome the limitations of single-user local spectrum sensing and draws more and more attention, but introduces additional overhead while improving sensing performance. Sequential detection algorithm can reduce data exchange latency in local sensing and optimize spectrum detection algorithm timeliness to ensure detection performance under the premise under the premise to ensure detection performance.This paper focuses on the application of sequential detection algorithm in cooperative spectrum sensing.The paper firstly introduces the concepts of cognitive radio and the latest developments of spectrum sensing, discusses the traditional cooperative spectrum sensing technology, integration guidelines and problems exist in it, and based on the previous introduction, combines with sequential detection algorithms and cooperative spectrum sensing.A cooperative spectrum sensing algorithm based on energy probability distributions and sequential sensing algorithm is presented for the problem that the overhead of the traditional sequential detection in the case of low SNR may be too large. In the algorithm,cognitive users divide the signal sequence to fragment, count the number of energy value in each period greater than a preset threshold, and upload them to the fusion center, the fusion center uses sequential test to make a decision. Performance analysis and simulation results show that average number of samples of the proposed algorithm is significantly less than energy detection algorithm at low SNR, and found the best segment length to make the system throughput maximum.Afterwards, the paper studies sequential change point detection cooperative spectrum sensing methods and an improved algorithm is presented for the problem that uncertainty primary user signal impacts detection performance. In the algorithm, the test statistic of local sensing is the sum of a plurality of sample points to reduce system overhead by reducing the number of decision times, while improve the cumulative amount of changes in fusion center to reduce the computational complexity by cancel the likelihood ratio calculation. Simulation results show that the proposed improve algorithm can accurately determine the signal change while the primary user signal status is unknown.