| 電子情報通信学会 研究会発表申込システム 講演論文 詳細 |
技報閲覧サービス [ログイン] 技報アーカイブ |
| トップに戻る | 前のページに戻る | [Japanese] / [English] |
| 講演抄録／キーワード | ||
| 講演名 | 2026-03-05 13:50 ［記念講演］PrometheusFree: Concurrent Detection of Laser Fault Injection Attacks in Optical Neural Networks ○Kota Nishida・Yoshihiro Midoh・Noriyuki Miura（UOsaka）・Satoshi Kawakami（Kyushu Univ.）・Alex Orailoglu（UCSD）・Jun Shiomi（UOsaka）  VLD2025-92 HWS2025-88 ICD2025-103 |
|
| 抄録 | （和） | Silicon Photonics-based AI Accelerators (SPAAs) have been considered as promising AI accelerators achieving high energy efficiency and low latency. While many researchers focus on improving SPAAs’ energy efficiency and latency, their physical security has only recently received attention. While it is essential to deliver strong optical neural network inferencing approaches, their success and adoption are predicated on their ability to deliver a secure execution environment. Towards this end, this paper proposes PrometheusFree, an optical neural network framework that is capable of concurrent detection of laser fault injection attacks. This paper first presents an illustrative threat of laser fault injection attacks on SPAAs, capable of subjecting the optical neural network to misclassifications. The threat then is addressed in this paper by developing techniques for concurrent detection of the laser fault injection attacks. Furthermore, this paper introduces a novel application of Wavelength Division Perturbation (WDP) technique where wavelength-dependent Vector Matrix Multiplication (VMM) results are utilized to boost fault attack detection accuracy. Simulation results show that PrometheusFree achieves over 96% attack-caused misprediction recall as the use of the WDP technique squashes the attack success rate by 38.6% on average. Compared with prior art, PrometheusFree limits the average attack success ratio to 0.019, yielding a 95.3% reduction. The experimental results confirm the superiority of the concurrent detection and the boost in attack detection abilities imparted by the WDP approaches. |
| （英） | Silicon Photonics-based AI Accelerators (SPAAs) have been considered as promising AI accelerators achieving high energy efficiency and low latency. While many researchers focus on improving SPAAs’ energy efficiency and latency, their physical security has only recently received attention. While it is essential to deliver strong optical neural network inferencing approaches, their success and adoption are predicated on their ability to deliver a secure execution environment. Towards this end, this paper proposes PrometheusFree, an optical neural network framework that is capable of concurrent detection of laser fault injection attacks. This paper first presents an illustrative threat of laser fault injection attacks on SPAAs, capable of subjecting the optical neural network to misclassifications. The threat then is addressed in this paper by developing techniques for concurrent detection of the laser fault injection attacks. Furthermore, this paper introduces a novel application of Wavelength Division Perturbation (WDP) technique where wavelength-dependent Vector Matrix Multiplication (VMM) results are utilized to boost fault attack detection accuracy. Simulation results show that PrometheusFree achieves over 96% attack-caused misprediction recall as the use of the WDP technique squashes the attack success rate by 38.6% on average. Compared with prior art, PrometheusFree limits the average attack success ratio to 0.019, yielding a 95.3% reduction. The experimental results confirm the superiority of the concurrent detection and the boost in attack detection abilities imparted by the WDP approaches. | |
| キーワード | （和） | Silicon Photonics-based AI Accelerator (SPAA) / Optical Neural Network (ONN) / Laser Fault Injection Attack / / / / / |
| （英） | Silicon Photonics-based AI Accelerator (SPAA) / Optical Neural Network (ONN) / Laser Fault Injection Attack / / / / / | |
| 文献情報 | 信学技報, vol. 125, no. 382, VLD2025-92, pp. 97-97, 2026年3月. | |
| 資料番号 | VLD2025-92 | |
| 発行日 | 2026-02-25 (VLD, HWS, ICD) | |
| ISSN | Online edition: ISSN 2432-6380 | |
| 著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します．(許諾番号：10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) | |
| PDFダウンロード | VLD2025-92 HWS2025-88 ICD2025-103 |
|
| 研究会情報 | |
| 研究会 | ICD HWS VLD |
| 開催期間 | 2026-03-04 - 2026-03-07 |
| 開催地（和） | みんなの貸会議室 那覇旭町店404会議室 |
| 開催地（英） | |
| テーマ（和） | システムオンシリコンを支える設計技術, ハードウェアセキュリティ, 一般 |
| テーマ（英） | |
| 講演論文情報の詳細 | |
| 申込み研究会 | VLD |
| 会議コード | 2026-03-ICD-HWS-VLD |
| 本文の言語 | 英語 |
| タイトル（和） | |
| サブタイトル（和） | |
| タイトル（英） | PrometheusFree: Concurrent Detection of Laser Fault Injection Attacks in Optical Neural Networks |
| サブタイトル（英） | |
| キーワード(1)（和/英） | Silicon Photonics-based AI Accelerator (SPAA) / Silicon Photonics-based AI Accelerator (SPAA) |
| キーワード(2)（和/英） | Optical Neural Network (ONN) / Optical Neural Network (ONN) |
| キーワード(3)（和/英） | Laser Fault Injection Attack / Laser Fault Injection Attack |
| キーワード(4)（和/英） | / |
| キーワード(5)（和/英） | / |
| キーワード(6)（和/英） | / |
| キーワード(7)（和/英） | / |
| キーワード(8)（和/英） | / |
| 第1著者 氏名（和/英/ヨミ） | Kota Nishida / Kota Nishida / |
| 第1著者 所属（和/英） | The University of Osaka (略称： UOsaka) The University of Osaka (略称： UOsaka) |
| 第2著者 氏名（和/英/ヨミ） | Yoshihiro Midoh / Yoshihiro Midoh / |
| 第2著者 所属（和/英） | The University of Osaka (略称： UOsaka) The University of Osaka (略称： UOsaka) |
| 第3著者 氏名（和/英/ヨミ） | Noriyuki Miura / Noriyuki Miura / |
| 第3著者 所属（和/英） | The University of Osaka (略称： UOsaka) The University of Osaka (略称： UOsaka) |
| 第4著者 氏名（和/英/ヨミ） | Satoshi Kawakami / Satoshi Kawakami / |
| 第4著者 所属（和/英） | Kyushu University (略称： Kyushu Univ.) Kyushu University (略称： Kyushu Univ.) |
| 第5著者 氏名（和/英/ヨミ） | Alex Orailoglu / Alex Orailoglu / |
| 第5著者 所属（和/英） | University of California, San Diego (略称： UCSD) University of California, San Diego (略称： UCSD) |
| 第6著者 氏名（和/英/ヨミ） | Jun Shiomi / Jun Shiomi / |
| 第6著者 所属（和/英） | The University of Osaka (略称： UOsaka) The University of Osaka (略称： UOsaka) |
| 講演者 | 第1著者 |
| 発表日時 | 2026-03-05 13:50:00 |
| 発表時間 | 25分 |
| 申込先研究会 | VLD |
| 資料番号 | VLD2025-92, HWS2025-88, ICD2025-103 |
| 巻番号（vol） | vol.125 |
| 号番号（no） | no.382(VLD), no.383(HWS), no.384(ICD) |
| ページ範囲 | p.97 |
| ページ数 | 1 |
| 発行日 | 2026-02-25 (VLD, HWS, ICD) |
[研究会発表申込システムのトップページに戻る]