Detection and tracking of a moving target using SAR images with the particle filter-based track-before-detect...

Area of Science:

• Radar Systems Engineering
• Signal Processing
• Target Tracking

Background:

• Traditional track-after-detect (TAD) methods struggle with detecting low signal-to-noise ratio (SNR) moving targets in synthetic aperture radar (SAR) imagery.
• Moving target detection in SAR is crucial for various applications, but low SNR significantly hinders performance.

Purpose of the Study:

• To propose a novel particle filter (PF) based track-before-detect (TBD) algorithm for enhanced moving target detection and tracking in SAR.
• To address the limitations of TAD approaches in low SNR scenarios.

Main Methods:

• Implementation of a PF-based TBD algorithm incorporating a SAR moving target signal model.
• Utilizing sub-area calculations for likelihood ratio computation to improve efficiency.
• Optimizing the number of particles for a balance between performance and computational cost.

Main Results:

• The proposed TBD approach successfully detects and tracks moving targets with SNR as low as 7 dB.
• Demonstrated superior performance compared to TAD methods when SNR is below 14 dB.
• Achieved improved computational efficiency with minimal loss in detection and tracking accuracy.

Conclusions:

• The novel PF-TBD algorithm offers a robust solution for detecting and tracking low SNR moving targets in SAR systems.
• The method effectively resolves target ambiguities and provides true estimations.
• This approach significantly advances SAR-based moving target detection capabilities.