Research on High Reliable Wireless Channel Data Cleaning Method

Abstract

Recent developments in the 5th generation wireless communication system have heightened the need for the propagation characteristics and modeling of wireless channels. As the propagation characteristics and variation rules of radio waves in different scenes, frequency points and bandwidth are all hidden in the channel test massive data that have the big data features, it is necessary to carry out effective data cleaning methods to make better use of test data. This paper analyzes and compares a variety of data cleaning methods first, then designs a data cleaning strategy according to the characteristics of wireless channel test data. Finally, the effectiveness of the data cleaning strategy is verified through simulation. This paper provided significant theoretical and technical support for the wireless environment reconstruction and model construction in the big data era.