Proposal of a Hamming distance search CAM using neuron MOS transistors

An associative or content addressable memory (CAM) with Hamming distance search functions is crucial in data management and pattern recognition applications. We propose a new Hamming distance search CAM using neuron MOS transistors. The proposed CAM can compare stored data and an input data in parallel, and find stored data with Hamming distance within a certain range. This CAM has a large noise margin and reduces search current and power consumption. The operating characteristics of the CAM are also analyzed in detail and the results obtained from these analyses are fully confirmed by HSPICE.