Compressive sensing of sparse radio frequency signals using optical mixing

Area of Science:

• Photonics and Optical Engineering
• Signal Processing
• Electromagnetics

Background:

• Sparse radio frequency (RF) signals present challenges for traditional measurement techniques.
• Compressive sensing (CS) offers a promising alternative for efficient signal acquisition.
• Optical systems can be leveraged for high-speed signal processing and measurement.

Purpose of the Study:

• To demonstrate an optical mixing system for characterizing sparse RF signals.
• To investigate the performance of compressive sensing (CS) in an optical domain for RF signal measurement.
• To compare different recovery methods for sparse RF signals using the developed optical system.

Main Methods:

• An optical mixing system was developed, modulating RF signals onto a wavelength-chirped optical field.
• Time-wavelength-space mapping was employed to interface the optical field with a spatial light modulator (SLM).
• The SLM programmed pseudo-random bit sequences (PRBS) to construct the CS measurement matrix, with measurements obtained via photodiode integration.

Main Results:

• The system successfully recovered both pulsed and sinusoidal sparse RF signals.
• Performance was evaluated based on the measurement matrix dimensions and the penalized ℓ(1) norm recovery method.
• For square measurement matrices, the penalized ℓ(1) norm method demonstrated superior recovery performance compared to matrix inversion.

Conclusions:

• The optical mixing system provides an effective platform for measuring sparse RF signals using CS.
• The penalized ℓ(1) norm method is a robust technique for reconstructing RF signals from compressed measurements.
• This approach offers advantages over conventional recovery methods, particularly with optimized measurement matrix configurations.