Bayesian High Resolution Range Profile Reconstruction of High-speed Moving Target from Under-sampled Data

Area of Science:

• Radar Systems Engineering
• Signal Processing
• Target Recognition

Background:

• High-resolution range profiles (HRRP) are crucial for target recognition, especially when Inverse Synthetic Aperture Radar (ISAR) imaging is not feasible.
• High-speed targets introduce high-order phase errors that stretch HRRP, complicating analysis and recognition.
• Traditional HRRP reconstruction and velocity estimation algorithms fail with under-sampled data.

Purpose of the Study:

• To propose a novel HRRP reconstruction algorithm for high-speed targets using under-sampled data.
• To accurately compensate for high-order phase errors induced by target velocity.
• To enhance HRRP resolution and concentration for improved target recognition.

Main Methods:

• Utilizing Laplacian scale mixture (LSM) as a sparse prior for HRRP.
• Employing variational Bayesian inference to derive the posterior distribution of HRRP.
• Jointly estimating target velocity via entropy minimization and HRRP sparseness during reconstruction.

Main Results:

• Successfully reconstructed high-resolution HRRP from under-sampled data.
• Effectively compensated for high-order phase errors caused by high-speed target motion.
• Validated the proposed Bayesian algorithm's effectiveness using both simulated and measured data.

Conclusions:

• The proposed Bayesian HRRP reconstruction algorithm is effective for high-speed targets with under-sampled data.
• The method accurately estimates target velocity and compensates for phase errors, leading to concentrated HRRP.
• This approach offers a viable solution for target recognition in challenging radar scenarios.