Transient Interference Excision and Spectrum Reconstruction with Partial Samples Using Modified Alternating Direction...

Area of Science:

• Radar Systems Engineering
• Signal Processing
• Electromagnetics

Background:

• Over-the-horizon radar (OTHR) faces challenges with transient interference and sparse echo data, degrading target detection.
• Conventional methods for sparse OTHR signals produce grating lobes, further diminishing performance.
• Multi-target detection from sea to air exacerbates data sparsity in modern OTHR systems.

Purpose of the Study:

• To develop a robust method for reconstructing target and clutter spectra from sparse OTHR signals.
• To enhance the performance of OTHR target detection by addressing signal sparsity and interference.
• To improve the accuracy and efficiency of OTHR signal spectrum reconstruction.

Main Methods:

• A modified Alternating Direction Method of Multipliers Network (ADMM-Net) is proposed for spectrum reconstruction.
• Transient interferences are identified using sparse basis representation and subsequently excised.
• Doppler sparsity-constrained optimization is solved using the trained network to reconstruct the complete Doppler spectrum.

Main Results:

• The proposed ADMM-Net effectively reconstructs the Doppler spectrum for OTHR target detection.
• The method successfully identifies and removes transient interferences, yielding a processed sparse OTHR signal.
• The approach demonstrates a balance between efficiency and accuracy in OTHR signal spectrum reconstruction compared to traditional methods.

Conclusions:

• The modified ADMM-Net provides a significant advancement in reconstructing sparse OTHR signals for improved target detection.
• The method's effectiveness is validated through both simulated and real-measured OTHR data.
• This approach offers a promising solution for enhancing OTHR capabilities in complex environments.