The advantages of multivalued logic in optical parallel computation need no introduction. There are lots of proposals, already reported, where tristate, quarternary state logic operations can be performed with optics. Here we report a new approach to implement tristate logic based all optical flip-flop using optical nonlinear material. The concept and the principle of operation of this type of flip-flop are different from that of the conventional binary one.

All-optical multibit address recognition at 20 Gb/s is demonstrated based on a special AND logic of terahertz optical asymmetric demultiplexer (TOAD). The semiconductor optical amplifier (SOA) used in the TOAD is biased at transparency status to accelerate the gain recovery. This is the highest bit rate that multibit address recognition is demonstrated with SOA-based interferometer. The experimental results show low pattern dependency. With this method, address recognition can be performed without separating address and payload beforehand.

Optics is a potential candidate in information, data, and image processing. In all-optical data and information processing, optics has been used as information carrying signal because of its inherent advantages of parallelism. Several optical methods are proposed in support of the above processing. In many algebraic, arithmetic, and image processing schemes fundamental logic and memory operations are conducted exploring all-optical devices. In this communication we report an all-optical matrix multiplication operation with non-linear material based switching circuit.

An optical cross connection network which adopts coarse wavelength division multiplexing (CWDM) and data packet is introduced. It can be used to realize communication between multi-CPU and multi-MEM in parallel computing system. It provides an effective way to upgrade the capability of parallel computer by combining optical wavelength division multiplexing (WDM) and data packet switching technology. CWDM used in network construction, optical cross connection (OXC) based on optical switch arrays, and data packet format used in network construction were analyzed. We have also done the optimizing analysis of the number of optical switches needed in different scales of network in this paper. The architecture of the optical interconnection for 8 wavelength channels and 128 bits parallel transmission has been researched. Finally, a parallel transmission system with 4 nodes, 8 channels per node, has been designed.

The limitations in electronics in arithmetic, algebraic & logic processing are well known. Very high speed performance (above GHz) are not expected at all in conventional electronic mechanism. To achieve high speed performance we may think on the introduction of optics instead of electronics for information processing and computing. Non-linear optical material is a successful candidate in this regard to play a major role in the optically controlled switching systems and therefore in all-optical parallel computation these materials can show a very good potential aspect. In this paper, we have proposed a new method of an optical half adder as well as full adder circuit for binary addition using non-linear and linear optical materials.