Spatial and temporal independent component analysis of micro-Doppler features

Abstract

Micro-Doppler features can be regarded as a unique signature of an object with movements and provide additional information for classification, recognition and identification of the object. Independent component analysis (ICA) can decompose micro-Doppler features into independent basis functions that represent salient physical movement attributes of the object. To study ICA of micro-Doppler features, we used a dataset generated by simulation of radar returned signals from rotating objects and tumbling objects. Fast ICA algorithm was used in our study to decompose micro-Doppler features into a set of spatial and temporal independent components. Spatial characteristics of the independent components combined with the corresponding temporal characteristics can be used to improve performance of classification, recognition and identification.