Bayesian LRT Wideband Detector Assisted by Scattering Characteristics for Space Target

With the increasing frequency of space activities and the rapid growth in the number of on-orbit space targets, the limited orbital resources have become increasingly scarce. Uncontrolled space targets significantly elevate the risk of collisions, posing a serious threat to the safety of the orbital environment. Therefore, it is of great significance to use radar detection methods to efficiently and accurately detect such targets. However, conventional wideband target detection methods have difficulties in parameters selection and extended energy accumulation, which results in detection performance loss in low signal-to-noise ratio (SNR) condition. To address this issue, this paper proposed a wideband Likelihood Ratio Test (LRT) detector assisted by target scattering characteristics. First, the scattering characteristics of the target across the entire spatial domain were modeled using Gaussian Mixture Model (GMM) to obtain a prior probabilistic distribution. Subsequently, these priors were incorporated into the energy accumulation-based LRT detector to construct a Bayesian LRT posterior detector. Next, the posterior distribution of the detector was derived analytically, and the detection threshold was calculated through Monte Carlo (MC) simulations and Kernel Density Estimation (KDE). In order to solve the problem of observation angle sensitivity, a fusion decision-making framework was proposed. Based on this, a multi-angle target characteristic fusion LRT posterior detector was designed to enable robust detection of space targets under low-SNR or long-distance conditions. Finally, simulation experiments were conducted to verify the effectiveness of the proposed method.