Architecture for compact and fast associative-memories with All-parallel nearest-match hamming-distance search

摘要

Finding the nearest-match between an input-data word of W bit length and a number R of reference-data words is a basic operation for pattern recognition as well as data compression. Associative-memory architecture for Hamming-distance search, compact implementation and short nearest-matches times up to large distances is proposed. Main ideas are fast analog word comparison and self-adaptive winner-line-up amplification. A 0.6 um, 2-poly, 3-metal CMOS design with 32 rows and 128 columns verifies the key concepts.