Multi-input optical parallel logic processing with the shadow-casting technique

A generalized method of encoding multiple input patterns employing only two distinguishable states of output light levels was proposed. The technique was patterned after the spatial coding method developed previously for a two-input system without use of polarization states. Though the proposed method can theoretically handle any number of input patterns, the number of inputs was confined to 16 when diffraction limitations of the system was considered. It was also demonstrated that the correlation and the decoding of a coded pattern can be done for all possible combinational logic operations in a single step. As a result, the method can be used for addition of binary numbers with the carry-look-ahead method of addition. Experimental confirmation of the proposed approach was presented for simultaneous processing of three input patterns.