Joint effects of on-board Doppler filtering and quantization in spaceborne SAR systems

Abstract

A novel strategy for data volume reduction in spaceborne synthetic aperture radar (SAR) system has been recently proposed, which consists of low-pass filtering and decimating the data prior to downlink and compensating for the filter’s transfer function in the processing. However, if data are further quantized prior to downlink, an amplification of the quantization noise may occur, when compensating for the transfer function of the filter. In a context of data volume reduction, data are likely to be quantized. This paper analyzes the joint effects of on-board Doppler filtering and block-adaptive quantization (BAQ) in spaceborne synthetic aperture radar (SAR) systems. For the present analysis, SAR raw data acquired by the German satellite TerraSAR-X are used. As is apparent, the removal of the high Doppler frequency components by a low-pass filter leads to better performance in terms of noise equivalent sigma zero (NESZ) for the same quantization rate. However, if the transfer function of the filter is not flat in the processed bandwidth, an amplification of the quantization noise occurs, when compensating for the filter’s transfer function in the processing.