Rui Zhu

Radar systems in practical environments are often affected by radio-frequency interference, which is commonly mitigated by spatial cancellation using auxiliary antennas. However, the use of auxiliary antennas inevitably introduces frequency-dependent amplitude and phase distortions between antennas, degrading signal synthesis and suppression performance. Existing calibration methods usually rely on prior measurements or explicit parameter estimation, making them sensitive to model mismatch, especially in multi-interference scenarios with spectral overlap. This letter proposes a spectral flatness measure (SFM)-based method for distortion compensation and interference suppression. The received signal is decomposed into subbands, where distortions are approximated as locally constant and compensated independently, and all parameters are jointly optimized via a genetic algorithm using the SFM criterion. The method requires no prior information or external calibration signals and achieves robust interference suppression. Simulation and experimental results validate its effectiveness in real interference environments.

The Polar Format Algorithm (PFA) is widely regarded as an efficient imaging algorithm, typically performing well when either the out-of-plane motion of the radar is negligible or the imaging scene is approximately planar. However, in Earth-based radar imaging of the Moon, these conditions are often violated. The unavoidable out-of-plane motion of the radar and the non-planar terrain of the lunar surface introduce space-variant phase errors that cause geometric distortion and defocus. To address this issue, an improved PFA with out-of-plane motion compensation is proposed in this Letter. First, an optimal focus plane is selected to eliminate the linear component of the out-of-plane motion and the resulting geometric distortion errors. Then, based on the mapping relationship of phase errors between the phase history domain and the wavenumber domain, space-variant filters are designed in the azimuth wavenumber domain to efficiently compensate for the space-variant defocus errors in PFA images. Finally, the effectiveness of the proposed algorithm is validated by the results obtained from experimental data.

In the calibration of distributed radar for radio astronomy, deep space radio sources are commonly used as calibration sources to correct inter-array delay errors, and accurate delay estimation is critical. Traditional correlation methods are limited by sampling frequency, achieving accuracy only at the sampling interval level. To achieve higher accuracy, sub-sample estimation is necessary. This letter proposes a precise delay calibration method using maximum likelihood iteration for sub-sample delay estimation. The proposed algorithm starts with the frequency domain features, first transforming the delay estimation problem into a phase estimation problem, and then calculating the likelihood function of the phase difference. A cost function is established based on the maximum likelihood criterion, and the optimal solution is obtained using the Newton iteration method. Compared to other algorithms, the proposed algorithm achieves superior accuracy in sub-sample delay estimation, meeting stringent calibration requirements in radio astronomy. Simulation and experimental results verify the validation of the algorithm.

The distributed coherent aperture radar (DCAR) utilizes full coherent processing (FCP). Compared to single radar unit observations, N radar units can achieve an N ³ times increase in signal-to-noise ratio (SNR), providing an advantage in observing distant targets. However, its stringent requirements for time and phase of multiple radar units make coherent parameters (CPs) estimation the crucial aspect of FCP. The paper introduces a satellite data-driven CPs estimation method via hierarchical and hybrid generalized Radon-Fourier transform (HHGRFT). Firstly, the FCP procedure is outlined, and the signal model with CPs is established. Secondly, the principles for selecting satellites and observation time in the experimental setup are induced, along with an analysis of the SNR variations during processing. Furthermore, through a hierarchical processing approach using the generalized Radon-Fourier transform (GRFT), inter-pulse coherence processing, inter-unit-radar coherence processing, and inter-sub-aperture coherent or non-coherent processing are sequentially conducted. The utilization of generalized sharpness and gradient descent method facilitated CPs estimation, subsequently enhancing the SNR post-coherence processing. Finally, the proposed method has been validated through simulation and successfully applied to a real DCAR system.

In co-located multiple-input multiple-output (MIMO) radar, slow-time code division multiple access (ST-CDMA) is an essential option of orthogonal waveforms, because the transmit power and bandwidth are well exploited with a relatively low level of hardware complexity. However, the code produces high sidelobes in the Doppler spectrum, which poses a negative impact on weak target detection. To address this issue, we propose a novel sidelobe suppression method that leverages the CLEAN framework in conjunction with a quasi-neural network (Quasi-NN). The novelty lies in the application of Quasi-NN for signal modeling in the target parameter estimation step of CLEAN. Specifically, Quasi-NN is employed to represent the signal after range compression. Its inputs, outputs, and internal weights are determined by the characteristics of antennas, waveforms, and targets. In this way, the estimation of target parameters is transformed into the optimization of weights, which thereby can be solved using a back-propagation (BP) algorithm. Simulation results demonstrate the superior performance of the proposed method under various scenarios. Real-data results using a 77GHz radar also show that the proposed method achieves lower sidelobes and thus improves the detection of weak targets.

The compressive sensing (CS) based tomographic SAR (TomoSAR) three-dimensional (3D) imaging method has the shortcoming of low efficiency, mainly represented in two aspects: first, the CS solver requires iterative calculation and hence is computationally expensive; second, the CS solver needs hyperparameters selection, which commonly requires cost-inefficient try-and-error attempts. Recently, the iterative CS solver is suggested to be replaced by a deep learning network for a tremendous processing speed improvement. However, the existing deep learning based TomoSAR imaging algorithms suffer from the problem of model inadaptability, i.e., being inadaptive to the observation model and the signal energy model and hence is low accuracy. This paper proposes a new model-adaptive network (MAda-Net) to implement deep learning based TomoSAR 3D imaging with a much improved processing accuracy. First, a new adaptive model-solving (AMS) module is introduced to solve the problem of the observation model inconsistency between the real spatially varying one and the approximately fixed one employed by the network. Second, a new adaptive threshold-activation (ATC) module is introduced to solve the problem of the signal energy model inconsistency between the real backscattered echo and the simulated echo for network training. The effectiveness of the proposed method has been verified by the computer simulations and the real unmanned aerial vehicle (UAV) SAR experiments.

Large scale phased array now becomes one of the most critical components in wireless communication system. However, the high cost and complexity of the phase controlling in conventional phased array makes it difficult to be widely deployed in commercial use. Alternatively, a frequency-controlled beam steering method with fixed radiation frequency based on sequentially mixing of the propagating IF and LO waves has been proposed as a low-cost solution. The beam pointing direction can be changed by single frequency controller rather than varying any phase shifters or tunable components on the antennas. However, this technique relies on accurately split the input IF and LO waves to a mixers array through an all passive power distribution network. This paper discusses the practical design consideration of a robust two wave mixing array and two sets of cascaded unequal power dividers are employed to efficiently distribute the IF and LO signal. To ensure constant radiation frequency, an external reciprocal mixer is used to keep the frequency difference between LO and IF being same as desired RF. The measured result shows that the proposed signal distribution network successfully achieves desired phase and amplitude of the converted RF signal.

The detection of low, slow and small (LSS) targets, such as small drones, is a developing area of research in radar, wherein the presence of ground clutter can be quite challenging. LSS targets, because of their unusual flying mode, can be easily shadowed by ground clutter, leading to poor radar detection performance. In this study, we investigated the feasibility and performance of a ground clutter mitigation method combining slow-time multiple-input multiple-output (st-MIMO) waveforms and independent component analysis (ICA) in a ground-based MIMO radar focusing on LSS target detection. The modeling of ground clutter under the framework of st-MIMO was first defined. Combining the spatial and temporal steering vector of st-MIMO, a universal signal model including the target, ground clutter, and noise was established. The compliance of the signal model for conducting ICA to separate the target was analyzed. Based on this, a st-MIMO-ICA processing scheme was proposed to mitigate ground clutter. The effectiveness of the proposed method was verified with simulation and experimental data collected from an S-band st-MIMO radar system with a desirable target output signal-to-clutter-plus-noise ratio (SCNR). This work can shed light on the use of ground clutter mitigation techniques for MIMO radar to tackle LSS targets.