Support Detection for SAR Tomographic Reconstructions from Compressive Measurements

Area of Science:

• Geophysics
• Remote Sensing
• Signal Processing

Background:

• Multibaseline Synthetic Aperture Radar (SAR) tomography enables 3D imaging by analyzing data from multiple viewpoints.
• Detecting and locating multiple scatterers is crucial for accurate SAR tomography reconstruction.
• Compressive sensing and support detection techniques offer potential for reducing measurement requirements.

Purpose of the Study:

• To address the challenge of detecting and locating multiple scatterers in multibaseline SAR tomography using compressive measurements.
• To analyze different support detection techniques for sparse vector recovery.
• To propose and evaluate a novel support detection method, the Generalized Likelihood Ratio Test (Sup-GLRT).

Main Methods:

• Analysis of support detection techniques for identifying the positions of non-zero elements in sparse vectors.
• Implementation and comparison of the proposed Sup-GLRT method with the SequOMP method.
• Evaluation based on probability of detection and probability of false alarm across varying numbers of measurements.

Main Results:

• Support detection techniques significantly reduce the number of required measurements for reliable SAR tomography solutions.
• The proposed Sup-GLRT method demonstrates competitive performance in detecting multiple scatterers.
• Performance comparison highlights the trade-offs between Sup-GLRT and SequOMP regarding detection probability and measurement count.

Conclusions:

• The Sup-GLRT method is a viable approach for multiple scatterer detection in SAR tomography.
• Support detection is a key strategy for efficient and accurate SAR tomography.
• Further research can explore optimizations and applications of the Sup-GLRT method in various SAR imaging scenarios.