A Digital Filter Bank Based Method for Data Reduction in Spaceborne SAR Electronics

Abstract

On-board data reduction is a major concern in designing of a spaceborne synthetic aperture radar (SAR) system since the volume of acquired SAR data is generally much higher than the downlink capability to the ground-station. The dechirp-on-receive (DoR) method is considered the state-of-the-art SAR data reduction technique. If there is deramping hardware in the SAR instrument, DoR is an effective method to reduce the volume of images for high resolution modes. In this paper, we use a digital signal processing method which applies a modified discrete Fourier transform (MDFT) filter bank for SAR data reduction. We call this method MDFT band selection filter (MDFT-BSF). This paper presents a quantitative analysis of the MDFT-BSF on data reduction performance. In addition, we provide a relationship between the data reduction performance and the number of sub-bands (M), and show that there is an optimal value of M that minimizes the volume of on-board SAR data. An evaluation of MDFT-BSF including a comparison with the DoR method presents that MDFT-BSF has an advantage for SAR data reduction. We show that MDFT-BSF can be implemented and operated on commercial space-grade FPGAs by the development of a spaceborne SAR digital receiver.INDEX TERMS dechirp-on-receive, field programmable gate array, modified discrete Fourier transform filter bank, satellite electronics, synthetic aperture radar