Research On Rayleigh Channel Scene Recognition Method Based On Statistical Characteristics Analysis Of Small Scale

The wireless channel scene environment is complex and changeable.In order to improve the performance of the communication system,it is necessary to configure different physical layer technologies for each channel scene.At this time,it is particularly important to accurately identify the wireless channel scene.In the research of channel scene identification,the statistical characteristics of small scale fading of different channel scenes are obviously different,so it can be taken as the characteristics of channel scenes.The feature data conforming to the requirements of channel scenes can be obtained by using the small scale fading statistical characteristic analysis method,which can further improve the identification accuracy.In this paper,the theoretical and applied research is carried out on Rayleigh channel scene recognition based on the analysis of statistical characteristics of small-scale fading.The main research work is as follows:(1)A method to analyze the statistical characteristics of Rayleigh channel model in small scale is proposed.Aiming at the existing first-order statistical characteristics analysis methods of Rayleigh channel model,such as complicated data processing method,unstable statistical performance and large error,a first-order statistical characteristics analysis method based on Anderson-Darling test is proposed,which can be used to test the first-order statistical characteristics distribution of Rayleigh channel model.The channel model is verified to meet the Rayleigh fading channel scenarios.In order to solve the problem of quantifying the power spectrum of second-order statistical characteristics of channel,an analysis method of second-order statistical characteristics based on mean absolute error was proposed.The method calculated the mean absolute error of second-order statistical characteristics and analyzed whether it met the Rayleigh fading channel scenario requirements according to the engineering threshold.(2)Realize the Rayleigh channel scene data acquisition.In order to solve the problem that some data in the existing Rayleigh channel model do not meet the requirements of channel scenarios,the statistical analysis method of small scale fading of Rayleigh channel model is adopted to analyze whether the statistical characteristics of small scale fading meet the requirements of channel scenarios.Thus,Rayleigh-classical,RayleighGaussian,Rayleigh-Flat,Rayleigh-Butterworth and Rayleigh-arch scene data can be obtained.(3)A Rayleigh channel scene recognition method based on the statistical characteristics analysis of small scale fading is proposed.In order to solve the problems of low accuracy and slow convergence rate of existing channel scene recognition methods,a Rayleigh channel scene recognition method based on small scale fading statistical characteristics analysis was proposed.WOA-BP neural network was improved by introducing nonlinear convergence factors and adaptive crossover mutation.The selected power spectrum data is used as feature recognition for five Rayleigh channel scenarios.Experimental results show that the proposed statistical analysis method can accurately and stably analyze the statistical characteristics of small scale fading of the channel,and can select the channel data that meets the requirements of the channel scenario.The proposed Rayleigh channel scene recognition method can accurately identify five categories of Rayleigh channel scenes by improving the intelligent algorithm.Compared with the WOA-BP neural network,the recognition speed and accuracy are improved.