Optical Computing
INTRODUCTION
Computing today is done almost entirely through the electrical medium. Even with the advent of optical fiber communication links, the routing of such fibers currently goes through an electrical conversion before being resent. Similarly, input devices such as scanners and detectors blindly transfer all information into electrical signals for analysis by computer software. To scan a document, for example, much code goes into filtering unwanted light, distinguishing between the text and the paper, and recognizing characters. Much time and effort in programming can be saved if optical filtering methods were done by the device itself rather than by a software program.
An optical computer is a system of optical elements that manipulate input signals in such a way that the output signal is the desired calculation, performed without electronics. At the heart of optical computing is the lens, whose inherent Fourier transformation capabilities allow for instant calculations that before would take much time by computer. The linear nature of the Fourier transform, along with its convolution and correlation properties, allow an input signal to be compared with a reference through spatial filtering, yielding a result that is either the sum or the difference of the two. This input signal can be either coherent or incoherent, leading to two different processes for its analysis. Such techniques have led to "real-time" processes for pattern recognition, logic operations, and even optical memory storage.