THROUGHPUT ANALYSIS OF DIGITAL PARTITIONING WITH ERROR-CORRECTING CODES FOR OPTICAL MATRIX-VECTOR PROCESSORS

Digital partitioning and error-correcting codes provide a technique fo r achieving high-accuracy computations with analog optical matrix-vect or processors. We present the results of a detailed throughput analysi s of this technique. The results indicate that using one processor per submatrix provides the best compromise between system throughput and hardware requirements over a range of matrix sizes. Error-correcting c odes are shown to not significantly degrade system throughput for larg e matrix sizes. Finally, a comparison to digital electronic computers is made.