Electromagnetic Environment Data Mining And Visualization Research

With the rapid development of wireless communication technology,the number of various advanced communication equipment is increasing,and the communication system is becoming more and more diversified,which makes the contradiction between low spectrum utilization rate and spectrum resource shortage become more and more serious,and the electromagnetic environment is becoming more and more complicated.In the past,the application and industry development of communication equipment were limited.Therefore,it is of great significance to study the mining,analysis,prediction and visualization technology of electromagnetic environment.This subject uses data mining on the electromagnetic environment to obtain the similarity analysis results of the electromagnetic environment in the frequency domain and space domain,and uses it as a priori information to improve the accuracy of electromagnetic spectrum prediction,and finally studies the visualization of complex electromagnetic environments method to provide support for the comprehensive management of the electromagnetic environment.The main work content is as follows:(1)Considering the complexity of the electromagnetic environment in the frequency domain and airspace due to the allocation of different service frequency bands and user differences,this topic is based on the method of correlation analysis and the Spearman correlation coefficient to study the electromagnetic environment in the frequency domain.And the implicit relationship of the airspace,and build a correlation network diagram.The experimental results show that the similarity of the electromagnetic environment spectrum data is ubiquitous,and most frequency points are related to the situation evolution of their adjacent frequency points.However,there is no unified rule for airspace correlation.(2)Considering that the existing spectrum prediction algorithm based on deep learning lacks the use of existing information and knowledge,this topic uses the electromagnetic environment similarity correlation network graph obtained by mining as prior information,and designs a graph convolutional network-based The spectrum prediction model realizes multi-domain joint spectrum prediction,and at the same time makes full use of the results of mining analysis to improve the accuracy and reliability of spectrum prediction.(3)Considering that the complexity of the electromagnetic environment is difficult to describe intuitively,this topic not only represents the similarity relationship network graph in the frequency domain,but also characterizes the spectrum occupancy at the coarse-grained and fine-grained levels by combining line graphs and box graphs degree,and express the activity mode of electromagnetic signals in a visual abstract way,and finally design and develop the electromagnetic environment visualization interface.