Long-time coherent integration can effectively improve the radar detection ability of a maneuvering target. However, with the increasing of integration time, radial velocity and radial acceleration of the moving target result in linear range migration (LRM), range curvature (RC), and Doppler frequency migration (DFM) of the received signal. These may degrade detection performance seriously. To solve these problems, an efficient coherent accumulation method based on improved time reversal transform (ITRT) and modified radon Fourier transform (MRFT) is proposed in this paper. Different from conventional methods, this approach firstly eliminates the acceleration effects by ITRT. It can remove RC and DFM simultaneously via multiplying its slow-time reversed conjugate data in range-frequency and azimuth-time domains. Then by employing a new Doppler filter bank (DFB), MRFT realizes the echoes spatial temporal decoupling as well as the coherent integration in searchingvelocity and Doppler-frequency domains. The major advantages of the proposed method are that RC and DFM can be rectified without parameter estimation, and computational complexity can be optimized by the new FFT-based filter bank. In addition, due to the nonlinearity of ITRT operation, the influence of cross terms of multiple targets on detection performance is also discussed. Several simulations are provided to demonstrate the effectiveness of our approach.

This study focuses on the key challenges in detecting and estimating motion parameters of ground maneuvering targets for airborne radar sensors. The complex unknown motion states of the ground maneuvering target, including velocity, acceleration, and jerk, result in range migrations (RMs) and Doppler frequency migrations (DFMs). These effects severely degrade the long-time coherent accumulation performance of the airborne radar, thereby limiting the reliable detection and precise parameter estimation of maneuvering targets. To address this issue, a new detection and motion parameter estimation method based on the range frequency reversal transform (RFRT) and searching Lv’s distribution (SLVD), i.e., RFRT-SLVD, is proposed. Specifically, the third-order RM (TRM) and quadratic DFM (QDFM) are considered. The proposed method operates as follows: First, RMs are eliminated simultaneously via the RFRT operation, which multiplies the echo by its reversed data in the range frequency and slow-time domains, leveraging the symmetric equal-interval sampling property of the range frequency. Subsequently, a phase compensation function (PCF) related to the jerk is constructed to compensate the QDFM. Finally, the LVD is performed to remove residual DFMs and achieve effective signal energy accumulation. Additionally, the case of a fast-moving target with Doppler ambiguity is analyzed, and a method for estimating three motion parameters is provided. A key advantage of the proposed technique is its ability to directly compensate the RMs without requiring prior knowledge of the maneuvering target, while also avoiding the blind speed sidelobe (BSSL) effect. In comparison with existing algorithms, RFRT-SLVD achieves a balanced trade-off between parameter estimation performance and computational efficiency. Numerical analyses and experiments are conducted to validate the method, assessing its detection capability for ground maneuvering targets, Doppler ambiguity resolution in parameter estimation, computational complexity, and method applicability in multi-target scenarios.

High-precision drogue localization during terminal guidance is critical to close-range autonomous unmanned aerial vehicle (UAV) docking and hinges on infrared–depth (IR–D) multimodal detection. Yet, deploying such detection on airborne edge computing platforms faces severe challenges due to modal heterogeneity, feature redundancy, and real-time constraints. A lightweight IR–D fusion detection network, termed AWIE-CGAF, is proposed for airborne edge deployment, which integrates frequency-domain, physics-prior-driven input enhancement with decoupled gated attention-based adaptive feature fusion to achieve efficient multimodal detection. A training-free Adaptive Wavelet Image Enhancement (AWIE) module is designed to differentially modulate image structures and details in the frequency domain, improving the signal-to-noise ratio and feature discriminability. Concurrently, a Cross-Gated Attention Fusion (CGAF) module employs decoupled cross-modal attention with independent gating, preserving modality-specific features while dynamically selecting complementary information, mitigating redundancy and feature contamination. Experiments on the self-constructed Drogue Infrared–Depth (DIRD) dataset showed that AWIE-CGAF achieved 89.5% mAP@0.5 and 58.2% mAP@0.5:0.95 with 13.5 M parameters, while maintaining real-time inference at 51.7 FPS on a Jetson AGX Orin edge platform. Among the evaluated methods, the proposed framework achieved the highest detection accuracy while retaining real-time edge inference capability. These results support the feasibility of AWIE-CGAF for resource-constrained IR–D drogue perception.