车联网中基于BP神经网络的智能协同频谱感知算法  

An Intelligent Cooperative Spectrum Sensing Algorithm Based on Back Propagation Neural Network in IoV

作　　者：张月霞[1,2] 赵义飞 ZHANG Yuexia;ZHAO Yifei(School of Information Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China;Key Laboratory of Modern Observation and Control Technology of Ministry of Education,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学信息与通信工程学院,北京100101 [2]北京信息科技大学现代测控技术教育部重点实验室,北京100101

出　　处：《电讯技术》2021年第8期919-924,共6页Telecommunication Engineering

基　　金：国家重点研发计划子课题(2020YFC1511704);国家自然科学基金资助项目(61971048);北京市科技计划课题(Z191100001419012);北京信息科技大学2020年促进高校内涵发展科研水平提高项目(2020KYNH212)。

摘　　要：针对传统信噪比加权频谱感知方法在车联网(Internet of Vehicles,IoV)环境中受噪声影响较大、感知准确率较低的问题,提出了一种基于反向传播(Back Propagation,BP)神经网络的IoV协同频谱感知(Cooperative Spectrum Sensing based on BP Neural Network,BP-CSS)算法。该算法首先将本地次用户能量检测结果进行协方差处理,然后通过BP神经网络对次用户信噪比进行权值优化,使用训练好的模型进行协同频谱感知。仿真结果表明,在信噪比0~25 dB范围内、10个次用户协同感知时,该算法在噪声干扰较大的环境中的平均检测准确率为90%,比基于信噪比加权频谱感知方法提升20%,比基于门限值频谱感知方法提升30%。In the Internet of vehicles(IoV)environment,the traditional spectrum sensing method based on the weighted signal-to-noise ratio(SNR)is greatly affected by noise,resulting in low sensing accuracy.To solve this problem,this paper proposes a cooperative spectrum sensing algorithm based on Back Propagation neural network(BP-CSS).The algorithm first processes the covariance of the local secondary user(SU)energy detection results,then optimizes the SNR of the SU through a BP neural network,and uses the trained model for cooperative spectrum sensing.The simulation results show that when the SNR is in 0~25 dB and 10 times of users cooperative perception,the average detection accuracy of the algorithm is 90%in the noisy environment,which is 20%higher than that of the algorithm based on SNR and 30%higher than that of the algorithm based on threshold.

关 键 词：车联网 BP神经网络 协同频谱感知 

分 类 号：TN919.1[电子电信—通信与信息系统]