The Research On The Electromagnetic Radiation Of Base Station Based On Wavelet Decomposition

With the rapid development of communication technology and the establishment of a large number of base stations,the electromagnetic radiation in the environment is increasing.Therefore,it is of great practical value to predict the intensity of electromagnetic radiation and to study the characteristics of its change.In the research of base station electromagnetic radiation prediction,the electromagnetic radiation of base station is often predicted by the parameters of antenna characteristic,channel and terrain,but it often changes sharply with the change of traffic volume,and it is difficult to predict it accurately by the previous parameters.In addition,the variation and correlation of electromagnetic radiation of base station are analyzed by statistical analysis and user behavior analysis,but these methods can not fully reveal the internal change cycle of the base station electromagnetic radiation,and can not accurately reflect the correlation of electromagnetic radiation of base stations in different time periods.In view of the above problems,the main contents and innovations of this paper are as followed:(1)In terms of electromagnetic radiation prediction of base stations,a method of electromagnetic radiation prediction based on wavelet is proposed in this paper.In this method,the TD-LTE base station electromagnetic sequence is obtained by different wavelet decomposition,and then the high and low frequency sequences are trained and predicted by the extreme learning machine,and the single branch of each wavelet decomposition is obtained,and then the Pearson correlation coefficient of each prediction single branch and the corresponding time traffic volume are divided.The weights of each unit can be used for combination forecasting.As a result of using the Pearson correlation coefficient to determine the combination weight,the actual situation of the combined forecasting weight allocation is fully considered,so the better prediction performance can be obtained.In this paper,the predicted values of this method are compared with the actual measured values.The errors between them are predicted to be 0.37%,0.57%,0.80%and 1.25%,respectively,which is in advance for 1 hour,3 hours,6 hours and 14 hours.These data show that the prediction method proposed in this paper can predict electromagnetic radiation accurately and effectively.(2)In the research of the electromagnetic radiation variation characteristics of the base station,this paper first proposes a method of wavelet for analyzing the electromagnetic radiation variation characteristics of the base station.In this method,the electromagnetic radiation of base station is changed in multiple time scales through Morlet wavelet,and the wavelet transform coefficients at different scales are obtained,and then the characteristics of the change period of the base station electromagnetic radiation are studied by the wavelet transform coefficient graph.Through the experimental analysis,this paper shows that the measured TD-LTE electromagnetic radiation data have fore major change periods of 12,22,30,45 hours,which can provide important reference for its change and prediction research.Then this paper proposes a method of wavelet for the correlation analysis of electromagnetic radiation based on the base station.Firstly,the electromagnetic radiation data of base station is analyzed by DB4 wavelet to obtain the low frequency sequence which can fully reflect the change of the base station electromagnetic radiation,and then it is used as the correlation analysis of the base station electromagnetic radiation.The experimental results show that the correlation coefficient of TD-LTE and FDD-LTE changes from 0.64 to 0.92 in the same area has a strong correlation,and the daily change of electromagnetic radiation is larger,but there is a certain change in it,which can provide a reference for the electromagnetic radiation measurement and environmental protection of the base station.