Research on LSI design methods to identify Trojan circuits in IP cores

2022 Fiscal Year Final Research Report

• Project/Area Number: 18K11228
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60040:Computer system-related
• Research Institution: Kyoto Sangyo University
• Principal Investigator: Yoshimura Masayoshi 京都産業大学, 情報理工学部, 准教授 (90452820)
• Co-Investigator(Kenkyū-buntansha): 細川 利典 日本大学, 生産工学部, 教授 (40373005)
• Project Period (FY): 2018-04-01 – 2023-03-31
• Keywords: トロイ回路 / ブラックボックス / ホワイトボックス / SATソルバー / モンテカルロツリーサーチ / 入力系列生成

Outline of Final Research Achievements

We have developed a technique to detect trojan circuits that may be included in IP cores, which are LSI design data. Trojan circuits that are activated in states that are rarely used during operation are difficult to detect in the LSI desgin phases. To detect these states, we developed a method using a SAT solver and Monte Carlo tree search. For small and medium benchmark circuits, we were able to identify these states with high accuracy in a relatively short time. On the other hand, search times for large benchmark circuits were long.

Free Research Field

コンピュータサイエンス

Academic Significance and Societal Importance of the Research Achievements

情報化社会において，LSIは基盤となる部品であり，LSIの信頼性や安全性が損なわれると，情報化社会の信頼性安全性にも大きな影響を与える．本研究は，LSIの信頼性と安全性をLSI設計レベルで高める技術である．本研究は，LSIの設計段階において，LSIに悪意のある回路が含まれていないかの判定に用いられる．LSIへの悪意のある回路の混入を防ぐことで，情報化社会の信頼性と安全性の向上に貢献する．