Abstract
Researchers have proposed a great number of optical matrix processing architectures and devices, each having significant advantages and disadvantages. However, no general metrics have been established to compare processors, and there has been little work to implement optical matrix processors for specific applications. In this article, we examine the range of proposed optical matrix processors both to determine the most promising classes of applications and, by establishing performance criteria, to select the best architectures and algorithms for these applications. This examination concludes that the greatest current potential is for analog optical matrix-vector processors to analyze slowly varying or invariant systems with high data throughput rates.
Get full access to this article
View all access options for this article.