Michelson Interferometric Optical Complex Convolution for Optical Encryption
Open AccessOptical real-time dynamic data processing finds a wide range of applications, notably in tensor algebra acceleration, often a bottleneck in machine learning, cryptography, digital holography, and other optically computed algorithms due to their high computational load. Moreover, cryptographic algorithms lean heavily on computationally demanding data compression algorithms to ensure the required security level and fulfill standard network specifications. In parallel, image processing algorithms in machine learning heavily rely on the speed of convolution and area dot product multiplication as they consume a major share of the computation budget. Quantum mechanics laws restrict the speed of intricate calculations based on classical electronic semiconductor hardware. Conversely, in optics, Optical Fast Fourier Transform (FFT) and dot-product multiplication executed passively by a single lens or metalens demonstrate reduced computational complexity compared to logarithmic scaling in electronic processing units. Optical computing systems based on Spatial Light Modulators (SLMs) are generally limited by the device's refresh rate but benefit from their high resolution and reconfigurability. Here we are to announce progress on a novel, 'full' (i.e., complex) convolution-based Reconfigurable Complex Convolution Module (RCCM) capable of independently modulating both phase and amplitude over two million pixels and its preliminary applications. Such a system is of extreme utility in rapidly progressing fields of optical computing, hardware acceleration, encryption, and machine learning, where neglecting phase modulation can lead to impractical bit-error rates.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.