The PCA feature extraction of target based on carrier-free UWB radar

Abstract

Target recognition of obstacles in the highway is a challenge problem. Most of the abnormal events on the expressway are pigs, traffic cones, stones, etc. This question of radar target recognition includes: echo signals detection and echo signals processing. The echo signals of this paper were produced by impulse UWB radar. And the signals were processed with feature extraction by The PCA (Principal Component Analysis) machine learning algorithm. The PCA is an unsupervised method by calculating the covariance matrix of the data and performing eigenvalue decomposition on the covariance matrix. The engineering research in this paper shows that extracting “information principal components” of obstacles can greatly help to identify the target on the expressway. The research results can be widely used in time sensitive target recognition.