Synthetic aperture radar (SAR) systems mounted on unmanned aerial vehicles (UAVs) are usually sensitive to trajectory deviations that cause serious motion error in the recorded data. In this letter, a novel range-dependent map-drift algorithm (RDMDA) is developed to accommodate the range-variant characteristics of severe motion errors. Utilizing the algorithm as a core estimate, we come up with a robust motion compensation strategy for the UAV SAR imagery. RDMDA outperforms the conventional MDA in both accuracy and robustness while it keeps similar efficiency. Real data experiment shows that the proposed approach is appropriate for precise imaging of UAV SAR systems equipped with only a low-accuracy inertial navigation system.

Conventional motion compensation (MOCO) under beam-center approximation is usually insufficient to correct severe track deviations for high-resolution synthetic aperture radar imaging. In this paper, a novel MOCO approach is developed for correction of the azimuth-variant motion errors by exploiting a precise angle-to-Doppler relationship within subapertures. The corruption from the residual motion errors to the angle-to-Doppler mapping is investigated and overcome by a compensation scheme of the scaled Fourier transform. Inheriting the high efficiency, the proposed azimuth MOCO approach has dramatically improved precision over the conventional subaperture MOCO method by reducing high side-lobe peaks of the point spread function. Extensive comparisons with other MOCO algorithms are given to show the superiority of the proposed algorithm. Moreover, real-data experiments are provided for a clear demonstration of our proposed approach.

Autofocus has attracted wide attention for unmanned aerial vehicle (UAV) synthetic aperture radar (SAR) systems, because autofocus process is crucial and difficult when the phase error is spatially dependent on both range and azimuth directions. In this paper, a novel two-dimensional spatial-variant map-drift algorithm (2D-SVMDA) is developed to provide robust autofocusing performance for UAV SAR imagery. This proposed algorithm combines two enhanced map-drift kernels. On the one hand, based on the azimuth-dependent phase correction, a novel azimuth-variant map-drift algorithm (AVMDA) is established to model the residual phase error as a linear function in the azimuth direction. Then the model coefficients are efficiently estimated by a quadratic Newton optimization with modified maximum cross-correlation. On the other hand, by concatenating the existing range-dependent map-drift algorithm (RDMDA) and the proposed AVMDA in this paper, a phase autofocus procedure of 2D-SVMDA is finally established. The proposed 2D-SVMDA can handle spatial-variance problems induced by strong phase errors. Simulated and real measured data are employed to demonstrate that the proposed algorithm compensates both the range- and azimuth-variant phase errors effectively.

Highly squinted synthetic aperture radar (SAR) data focusing is a challenging problem with difficulty to correct the severe range-azimuth coupling and motion errors. Squint minimization processing with the range-walk correction is widely adapted to simplify the decoupling processing, while it destructs the azimuth-shift invariance of conventional SAR transfer function. In this paper, a two-stage focusing algorithm (TSFA) is proposed to generate a focused imagery for the highly squinted airborne SAR. In the proposed algorithm, conventional range cell migration correction and azimuth matched filtering are performed and a fine focusing stage is established to correct the azimuth variance. In the fine focusing procedure, the coarse-focused image is divided into azimuth blocks to accommodate the correction of azimuth-variant residual range migration and phase terms. Moreover, precise motion compensation is embedded into the TSFA procedure to form an accurate airborne SAR imagery, which may be called the extended TSFA. In order to balance the processing precision and computational load, optimal selection of block size is investigated in detail. Both simulated and real measured airborne SAR data sets are used to validate the proposed approaches.

Aperture‐dependent motion compensation (MOCO) is crucial for high‐resolution airborne synthetic aperture radar (SAR) imaging, which can suppress image degradation especially for SAR systems working at a high‐frequency waveband, such as millimetre‐wave (MMW) SAR. The current precise topography and aperture (PTA) dependent MOCO algorithm aims to compensate the azimuth‐dependent motion error by using a modified azimuth matched filtering. However, the matched filter is deduced with the principle of stationary phase without considering the effects from the azimuth‐dependent motion errors, which limits its precision in focusing MMW SAR data involving severe motion errors. This study proposes a refined PTA algorithm with a precise post matched filter established by the method of series reversion. The range‐variant characteristic of residual motion errors is also investigated for an improved precision. Simulated and real MMW SAR data sets are used to validate the effectiveness of the proposal.

Robust estimation of frequency hopping radar signal parameters is the key to achieving stable jam in frequency hopping communication. In order to deal with the poor noise resistance performance of conventional time-frequency analysis based parameter estimation method under low signal-noise-ratio conditions. A robust method is proposed in this paper for frequency hopping radar waveform parameters estimation. The method is performed in time domain. Firstly, the pulse repetition interval of the received radar signal is extracted by matching filtering, which is used as the window length to divide the signal into multiple frames. The coherent integration process is performed on the multi-frame signal to increase the signal-noise-ratio, which effectively ensures the robustness of the algorithm under low signal-noise-ratio conditions. Secondly, the short-time Fourier transform is used to realize the rough estimation of the frequency hopping number, which reduces the range of subsequent hopping counts and reduces the amount of algorithm calculation. Finally, an accurate estimation of the waveform parameters including frequency hopping sequence, hop duration, frequency hopping interval and carrier frequency is realized. Simulation experiments verify the effectiveness of the algorithm and the robustness under various signal-noise-ratio conditions.