ABSTRACT
In this chapter, incoherent fiber-optic systolic array processors (FOSAPs), which employ a digital-multiplication-by-analog-convolution (DMAC) algorithm and the extension of the DMAC algorithm, are proposed for real-valued digital matrix computations. The important role of optics in optical computing and a variety of existing optical architectures using the DMAC algorithm are described in Section 3.1.2. Section 3.1.2.4 presents mathematical formulations of the DMAC algorithm and the twos-complement binary (TCB) arithmetic, while Section 3.1.3 outlines the operational principles of the elemental processors of the FOSAP architectures. The performances of the FOSAP multipliers (Section 3.1.5) are compared with the performances of digital electronic multipliers and other optical DMAC multipliers. Means of overcoming the limitation of the FOSAP architectures are discussed in Section 3.1.6. The theoretical aspects of incoherent fiber-optic signal processing described in Chapter 2 are applied in this chapter where intensity-based signals are considered.
