Abstract

Novel digital information applications such as multimedia and video-on-demand require the storage of a large amount of data at very low bit-error rates (BER), fast access to this data, and the efficient interface of the storage system to high speed (gigabit per second) networks [1]. Optical page-oriented memory (OPOM) technology [2], [3] is one candidate that simultaneously provides large capacity (10¹² bits/cm³ theoretically) and high data access rate (10⁹ bits/second or more). Unfortunately, the high raw BER (10^-4 - 10^-7) of the retrieved data is a limitation. The use of error detecting/correcting coding is one way to reduce the BER to a desirable rate (10^-15 or better) while improving the overall memory capacity and maintaining high data access rates. In holographic memories, for example, recording a large data page, consisting of perhaps 10³ x 10³ bits (a megabit), yields a low signal-to-noise ratio (SNR) and thus high BER because of the low distributed power per bit and inter-pixel crosstalk. By adding a small number of redundant (parity) bits, the BER of the decoded data can be reduced to desirable levels while maintaining the same data access rate [4]. This paper is concerned with the design of data encoding and decoding procedures for OPOMs which provide high effective storage capacity and output data rates, along with parallel-to-serial conversion and reformatting. These functions are implemented with optoelectronic smart-pixel interfaces.

© 1996 Optical Society of America