Research On Spectrum Detection Based On Sequential Pattern

The main technology of cognitive radio is the dynamic utilization of the spectrum in idle state. Spectrum detection is to detect the available spectrum in the communication environment. Then the spectrum is allocated to the cognitive user communication. Cognitive radio (CR)terminal analysis the specific regions of the frequency band. Without affecting the primary user (PU), the secondary user (SU) can be accessed through the detection algorithm to determine whether the secondary user can access the frequency band to carry out the data transmission of the cognitive network. In addition, the spectrum access priority of the secondary users is lower than that of the primary user. Secondary users in the process of using the frequency band must continue to detect the external environment. Once the primary user appears to occupy the frequency band, the secondary users have to quickly detect and exit the frequency band to prevent the interference caused by the signal.Therefore, the fast spectrum sensing plays an important role in cognitive radio, and the current industry is more and more research on spectrum sensing. Spectrum sensing performance has become a key issue in the study.Based on sequential pattern, a new phase detector is used in this paper. The setting of the threshold value is no longer affected by the noise power and the implement is easier than the relative matched filter detection and others. The phase difference between adjacent sampling signals is very different from that of the primary signal, so the probability likelihood ratio of the phase difference is used to detect spectrum. In addition, there is a certain correlation between the detection status of the current frequency band and the state of the last time slot, which can be used to improve the detection performance. When the Doppler frequency of the channel is high, the change of the channel signal is fast, the correlation of the channel is low relatively. On the contrary, the Doppler frequency is low, the channel correlation is high. The traditional detection methods have not been able to take advantage of past observations. In this paper, a sequential detection algorithm of q value is proposed to improve the detection performance and reduce the complexity of the algorithm as much as possible, and to reduce the number of samples used in the past.Through the above analysis, it is indicated that the weight phase detection based on sequential pattern is feasible and will bring a certain gain. Also,this new spectrum detection method is robust to the uncertainty of channel noise and multipath fading.