Compact high-frequency (HF) radar is a promising tool for maritime surveillance owing to its over-the-horizon and all-weather ability. However, it typically suffers from low tracking accuracy due to limited azimuth resolution. Cross localization using multiple radars provides a solution, but track-to-track association among radars remains a critical challenge. Previous association methods are mainly subject to that each radar’s range and azimuth measurement errors are comparable and sufficiently small, which is not satisfied in HF radar. In this article, we propose a novel track association algorithm based on the target region match (TRM) strategy, which employs the joint range and azimuth gates intersection criterion to combine the effect of range and azimuth measurement errors. Considering the azimuth error is much larger than that of the range, their contribution to the track association is balanced through the area intersection operation. Both simulation and experiment results show that the proposed TRM method produces better association performance than the other investigated methods.

Compact high-frequency (HF) surface wave radar can be used for remote sensing of sea surface dynamic parameters such as winds, waves and currents, as well as hard targets such as ships and low-altitude aircrafts. Due to its small antenna aperture, compact HF radar usually has a large error in direction of arrival (DOA) estimation, which significantly reduces its accuracy of target tracking. For a high-speed target, the DOA estimation error is greater due to range migration (RM) and Doppler frequency migration (DFM) as well as rapid change in the DOA of target echo. A high-speed target tracking method for compact HF radar is proposed in this paper, which utilizes the zero-mean property of the DOA error. First, the generalized Radon Fourier transform (GRFT) is used to estimate the target's motion parameters, and the multiple signal classification (MUSIC) algorithm is used to estimate the target's DOA. The original target track is obtained by combining the range and DOA estimates. Next, multiple positions on the original track are selected as the base points, and for each base point, a predicted track can be calculated based on the target motion parameter estimates, thus multiple predicted tracks are obtained. Then, these predicted tracks are screened according to their headings. The remaining predicted tracks after screening are fused to obtain the final track. Numerical simulations and field experiments demonstrate that this method reduces the impact of DOA error on tracking accuracy and expands the application of compact HF radar for high-speed target tracking.

Early-Warning Radar (EWR) is an advanced detection system capable of monitoring aerial targets over long distances with high precision, providing critical information support for defense security. However, EWR faces challenges such as a limited number of pulses, low coherent integration gain, small target Radar Cross Section (RCS), and complex clutter and electromagnetic interference environments. Conventional Constant False Alarm Rate (CFAR) detection algorithms struggle to effectively detect weak targets while maintaining an acceptable false alarm rate. To address these issues, this paper introduces a deep learning approach. A high target-clutter/interference/noise discriminative feature spectrum is obtained through phase difference transformation, upon which a dual-branch collaborative architecture network is constructed. In this architecture, the main network focuses on extracting spatiotemporal amplitude–phase characteristics, while the auxiliary branch implicitly mines the target’s physical boundary features from frequency-domain echoes. Through a self-attention mechanism, the features from both branches are semantically aligned and fused. This method significantly enhances the weak target detection capability of EWR under the constraint of a controlled false alarm rate. Test results show that under the false alarm rate ranging from ${10}^{-3}$ to ${10}^{-4}$, the SNR gain of the proposed algorithm is about 2∼5 dB, which is equivalent to increasing the radar detection range by 10∼30%.