Text
E-book Photonic Neural Networks with Spatiotemporal Dynamics
Optical computingis a general term for high-performance computing technologiesthat make effective use of the physical properties of light; it is also used as the nameof a research area that attracted attention from the 1980s to the 1990s. This wasexpected to be a solution for image processing and large-capacity information pro-cessing problems that could not be solved by electronics at that time, and a widerange of research was conducted, from computing principles to device developmentand architecture design. Unfortunately, with the rapid development of electronics,its superiority as a high-performance computing technology was lost, and the boomsubsided. However, as seen in the recent boom in artificial intelligence (AI), tech-nological development has repeated itself. In the case of AI, the success of deeplearning has triggered another boom. It should be noted that technologies with com-parable potential for further development have been developed in the field of opticalcomputing. Consequently, the momentum for research on new optical computing isincreasing. Optical computing is not a technology that emerged suddenly. Figure1 showsthe relationship among the research areas related to optical computing. Its rootsareoptical information processing, represented by a Fourier transform using lenses,and technologies such as spatial filtering and linear transform processing. They arebased on the physical properties of light waves propagating in free space, where thesuperposition of light waves is a fundamental principle. Light waves emitted fromdifferent sources in space propagate independently and reach different locations. Inaddition, light waves of different wavelengths propagate independently and can beseparated. Thus, the light-wave propagation phenomenon has inherent parallelismalong both the spatial and wavelength axes. Utilizing this property, large-capacityand high-speed signal processing can be realized. In particular, optical informationprocessing has been researched as a suitable method for image processing. Fromthis perspective, optical information processing can be considered an informationtechnology specialized for image-related processing.
Tidak tersedia versi lain