Compressive sensing approaches for the prediction of scattered electromagnetic fields

Area of Science:

• Electromagnetics
• Wave Propagation
• Computational Physics

Background:

• Predicting electromagnetic (EM) fields in complex environments is crucial for various applications.
• Traditional methods like ray tracing have limitations in accuracy and scope.
• Accurate EM field prediction requires detailed environmental and material property knowledge.

Purpose of the Study:

• To develop a novel method for predicting EM fields in arbitrary scattering environments.
• To reduce the number of required field measurements for accurate EM field reconstruction.
• To overcome limitations of traditional ray tracing approaches.

Main Methods:

• Utilized Huygens' principle for wave propagation modeling.
• Applied compressive sensing techniques for efficient data acquisition.
• Developed a compressive sensing-based subspace optimization method (CS-SOM).
• Employed a surface integral formulation for exact wave-matter interaction modeling.

Main Results:

• Successfully reconstructed EM fields in indoor environments with up to four scatterers.
• Achieved approximately 12% error in EM field prediction.
• Demonstrated accurate reconstruction with only 55% of the required variables measured.
• Validated the effectiveness of the CS-SOM technique.

Conclusions:

• The novel CS-SOM method accurately predicts EM fields with significantly fewer measurements.
• This approach offers a more efficient and potentially more accurate alternative to traditional methods.
• The technique is applicable to homogeneous media without prior knowledge of scatterer geometry or permittivities.