Research On Cooperative Spectrum Sensing Algorithm Based On SNR Threshold

With the great development of wireless communication technology, the demand for radio spectrum resources also increases correspondingly, leading to the intension of wireless spectrum resources. Therefore, how to enhance the efficiency in the use of spectrum has become a hot research topic. Cognitive Radio technology (CR) is expected to solve the problem. Spectrum detection is the key technology that prevents the authorized user from interference and which can improve the spectrum efficiency by discovering available frequency resources in the cognitive radio technology. Spectrum detection performance in practical scenario, however, is often influenced through the multipath-fading and shadow effect and the uncertainty problem of receiver. In order to reduce the effects of these problems, the cooperative spectrum detection is put forward to improve the detection performance.This thesis first expounds the basic principle of spectrum sensing in cognitive radio technology. And with energy detection as a local detection method, the simulation shows that the impact signal to noise ratio of cognitive users received signal plays on the performance of different collaborative detection method. Considering that different nodes locate in different areas and transmission channel is independent and different, spectrum sensing optimization algorithm based on SNR threshold of collaborative is proposed. The thesis’s main work is as follows:Firstly, On the basis of the traditional hard decision collaborative perception technology, the simulation shows that the impact on the performance of collaborative spectrum sensing under the condition of the poor SNR environment. Assuming that the average signal-to-noise ratio the cognitive user receives is under the condition of random distribution, optimization is focused on cooperative sensing algorithm based on the trust value of nodes. Considering the user’s primary and secondary channel environmental, evaluation parameters of the trust value of nodes are given, and a trust standardization function is set up to evaluation parameters. The simulation results of optimization algorithm illustrates the improved algorithm will change the number of collaborative nodes according to the channel environment under different SNR environment. So when the number of nodes are the same to participate in the collaborate detection, false alarm probability of the improved algorithm system is relatively small, and utilization rate of control bandwidth is reduced as well.Secondly, In order to distinguish the wireless environment which is responsible for the diversity, the testing time the system takes is decreased. Based on the weighted algorithm of the trust value of perceptual node, parameter named the signal-to-noise ratio to determine the threshold is introduced according to different SNR environment using different fusion rules. Analysis for the new algorithm is processed, and its false-alarm probability and detection probability is simulated and analyzed. The selection of the optimal signal-to-noise ratio to determine the threshold needs to take a different threshold for the comparison of the system detection probability, the threshold having the best detection effect is selected as the optimal value.