Associative Memory Using Neuron CMOS Inverters

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K06032
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Electron device/Electronic equipment
• Research Institution: Tokai University
• Principal Investigator: FUJIMOTO KUNIAKI 東海大学, 基盤工学部, 教授 (60229044)
• Co-Investigator(Renkei-kenkyūsha): YAHARA Mitsutoshi 東海大学福岡短期大学, 情報処理学科, 教授 (30259692)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 連想メモリ / ニューロンCMOS / ハミング距離 / マンハッタン距離

Outline of Final Research Achievements

An associative memory system finds the reference data that is the most similar to the input data, and outputs the reference data quickly. The fast search capability of the associative memory, which can find similar data from mass data in database is extremely useful for many fields including character recognition, fingerprint recognition, data compression, color image recognition, and so on. In this study, we aim to improve the performance of associative memory LSI by using a neuron CMOS inverter, which behaves similarly to a nerve cell, or neuron. In this project period, we design, fabricate and verify the associative memory LSI chips using neuron CMOS inverters which search similar data based on the Hamming distance and the Manhattan distance at high speed.

Free Research Field

電子回路