利用報告書 / User's Reports


【公開日:2024.07.25】【最終更新日:2024.04.15】

課題データ / Project Data

課題番号 / Project Issue Number

23KT1145

利用課題名 / Title

Reservoir Computing with Surface Acoustic Wave Resonators

利用した実施機関 / Support Institute

京都大学 / Kyoto Univ.

機関外・機関内の利用 / External or Internal Use

内部利用(ARIM事業参画者以外)/Internal Use (by non ARIM members)

技術領域 / Technology Area

【横断技術領域 / Cross-Technology Area】(主 / Main)加工・デバイスプロセス/Nanofabrication(副 / Sub)-

【重要技術領域 / Important Technology Area】(主 / Main)高度なデバイス機能の発現を可能とするマテリアル/Materials allowing high-level device functions to be performed(副 / Sub)-

キーワード / Keywords

Reservoir Computing, Surface Acoustic Waves,CVD,リソグラフィ/ Lithography,光リソグラフィ/ Photolithgraphy


利用者と利用形態 / User and Support Type

利用者名(課題申請者)/ User Name (Project Applicant)

Amit Banerjee

所属名 / Affiliation

京都大学 大学院工学研究科

共同利用者氏名 / Names of Collaborators in Other Institutes Than Hub and Spoke Institutes

Claude Meffan

ARIM実施機関支援担当者 / Names of Collaborators in The Hub and Spoke Institutes
利用形態 / Support Type

(主 / Main)機器利用/Equipment Utilization(副 / Sub)-


利用した主な設備 / Equipment Used in This Project

KT-205:プラズマCVD装置
KT-154:両面マスクアライナー露光装置
KT-111:ウエハスピン洗浄装置
KT-110:レジスト現像装置
KT-103:レーザー直接描画装置


報告書データ / Report

概要(目的・用途・実施内容)/ Abstract (Aim, Use Applications and Contents)

Reservoir computing is a neural network algorithm that intends to improve the power efficiency of neural networks by reducing the required training. Recently, non-linear MEMS sensors have been used as non-linear neurons in reservoir computers. In this work, we propose a new SAW device which couples a SAW resonator to a circular delay path. This delay path allows the device to function as a time-multiplexed reservoir computer.

実験 / Experimental

The delay line coupled SAW resonators were fabricated on chemically reduced YZ-LiNbO3 (Yamaju Ceramics Co., Ltd.) using a standard image reversal lift-off procedure (KT-154: Double-Sided Mask Aligner. AZ5214E, pre-bake at 110℃, 50s,  8.66 mJ.cm-2 exposure, image reversal bake at 117℃ for 2 min, flood exposure). The wafers were then diced into devices using a dicing SAW (Disco, DAD3231) and glued onto a custom PCB for measurement. The non-linear processing capacity of the resonators was evaluated using the time delayed binary parity task, using a custom measurement platform, developed on the Moku pro reconfigurable measurement system.

結果と考察 / Results and Discussion

The fabricated device, and results of the binary parity task are shown in Figure 1, below. The integrated device has an excellent linear memory capacity of up to 8 bits, however the non-linear memory capacity is lower than our previous results. This is likely due to the lower feedback gain of the physical memory, compared to the previous electronically implemented memory system result.

図・表・数式 / Figures, Tables and Equations


Figure 1 The delay coupled SAW reservoir computer. a) The control system for the reservoir computer b) The SAW structure used to generate the mask function c) A detailed view of the SAW resonator’s IDTs, reflector, and the coupler d) an overview of the overall delay-coupled structure. e) The short term memory of the reservoir computer f) The non-linear memory capacity of the reservoir computer.


その他・特記事項(参考文献・謝辞等) / Remarks(References and Acknowledgements)


成果発表・成果利用 / Publication and Patents

論文・プロシーディング(DOIのあるもの) / DOI (Publication and Proceedings)
口頭発表、ポスター発表および、その他の論文 / Oral Presentations etc.
  1. C. Meffan, T. Ijima, A. Banerjee, J. Hirotani, T. Tsuchiya, “Optimisation of a physical reservoir computer’s parameters using a genetic algorithm non-linear search”, 49th international conference on Micro and Nano Engineering, Berlin, Germany
  2. C. Meffan, T. Ijima, A. Banerjee, J. Hirotani, T. Tsuchiya, “Optimisation of a physical reservoir computer’s parameters using a genetic algorithm non-linear search”, ECML PKDD 2023, Turin, Italy, September 18–22
  3. C. Meffan, T. Ijima, A. Banerjee, J. Hirotani and T. Tsuchiya, "Exploring Surface Acoustic Wave Devices for Sound-Based Reservoir Computing," 2024 IEEE 37th International Conference on Micro Electro Mechanical Systems (MEMS), Austin, TX, USA, 2024, pp. 529-532, doi: 10.1109/MEMS58180.2024.10439559.
特許 / Patents

特許出願件数 / Number of Patent Applications:0件
特許登録件数 / Number of Registered Patents:0件

印刷する
PAGE TOP
スマートフォン用ページで見る