Adaptive compensation of amplitude and phase distortions introduced by SAR equipment into received signal

Abstract

SAR (synthetic aperture radar) is examined where FSK (frequency-shift keyed) signals are used to obtain high resolution in range. Using FSK signals in SAR goes along with distortions emerging in recieved signal which are due to dependence on amplitude frequency and phase of through transmission ratio of signal path, including transmitter, antenna and receiver. Amplitude-phase distortions are hard to account for due to randomness and temporal variability of their manifestations. The paper is aimed at adaptive algorithm development for evaluating and compensation for signal amplitude-phase distortions caused by SAR equipment, as well as at verification of its performance in synthesizing  radar images in SAR. Synthesis of compensation algorithm is done in two steps and it is for a complex signal, which is compressed in azimuth coordinate and is represented by module and argument (phase). First, signal module is evaluated by averaging amplitudes throughout RF hologram according to maximum likelihood criterion. Second, the phase is evaluated using non-parametric iteration autofocusing algorithm which is optimal as per minimum variance criterion. Complex vector was obtained as a result, which is used to compensate for amplitude and phase distortion. The developed autocalibration algorithm performs real-time distortions’ evaluation and compensation. Block diagram synthesis is done, which implements the proposed adaptive algorithm. Algorithm research results in laboratory and flight environment are prsented. It is made clear that powerful sidelobes at –15 dB level are observed if amplitude-phase distortions are in the signal. Such sidelobes in RI (radar images) make false marks, deform radar targets, disguise marks of small-size low-reflection targets. Flight tests with airborne Ku-band SAR testified that using developed algorithm enabled to reduce the level of maximum sidelobe down to –39 dB and boost quality of synthesized RI. Proposed algorithm capabilities are noted for processing operations’ automation of RF holograms and radar images’ decryption.