Optical computing using optical flip-flops in Fourier processors: use in matrix multiplication and discrete linear...

Area of Science:

• Optoelectronics
• Computer Architecture
• Digital Signal Processing

Background:

• Matrix multiplication is a fundamental operation in many scientific and engineering fields.
• Existing optical processing methods face challenges in speed and architectural complexity.
• Residue arithmetic offers potential for parallel computation but requires efficient hardware implementation.

Purpose of the Study:

• To propose a novel architecture for matrix multiplication using optical flip-flops (OFFs).
• To develop algorithms for residue arithmetic-based matrix multiplication in optical processors.
• To leverage optical Fourier processors for parallel element processing.

Main Methods:

• Design of an optical processor architecture incorporating optical flip-flops (OFFs).
• Implementation of algorithms based on residue arithmetic for matrix multiplication.
• Utilizing the parallel information retrieving ability of optical Fourier processors.
• Enabling bidirectional data flow with OFFs for architectural simplification.

Main Results:

• The proposed system enables parallel processing of all matrix elements.
• Bidirectional data flow facilitated by OFFs simplifies the architecture.
• Reduced residue-to-decimal conversion time due to parallel processing.
• Calculated operation times indicate suitability for real-time 2-D linear transforms.

Conclusions:

• The proposed architecture offers a promising approach for high-speed matrix multiplication.
• Optical flip-flops and residue arithmetic are key to achieving efficient parallel processing.
• The system demonstrates potential for real-time applications in optical signal processing.