Software Reduces Radio-Interference Effects in Radar Data

Abstract

A computer program suppresses the effects of narrow-band radio-frequency interference (RFI) on the data collected by a wide-band radar system. The need for this program arises because some advanced wide-band synthetic-aperture radar systems utilize frequency bands that include frequencies used by other radio services. In this program, the RFI environment is represented by an auto-regressive process, the frequency band of which is narrow relative to that of the radar. Most of the RFI signals, both narrow- and wide-band, are estimated in one pass of a least-mean-square (LMS) adaptive filter. The program implements three popular LMS algorithms: the time-domain LMS, the frequency-domain LMS, and the filter-bank LMS adaptive-filter algorithms. The program can be run in a manual or automatic mode. In the manual mode, the user selects the filter parameters prior to execution. In the automatic mode, the program utilizes median-filter and spectral-estimation techniques plus the variable-step-size LMS algorithm for automatic determination of filter parameters, and the parameters are adaptively changed as functions of the inputs, resulting in better overall performance.