Development of Maintenance-free Context Recognition by Passive Sensing

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K11941
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60060:Information network-related
• Research Institution: Osaka University
• Principal Investigator: Uchiyama Akira 大阪大学, 大学院情報科学研究科, 准教授 (70555234)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: Wi-Fi CSI / バッテリレスセンシング / 状況認識

Outline of Final Research Achievements

In this research project, we focused on developing maintenance-free context recognition techniques using passive sensing with batteryless tags that require no power supply and ultra-low power tags capable of continuous operation through environmental power generation. We designed batteryless tags using conductive materials such as copper foil tape and developed a tag identification method using deep learning based on the reflected RF signal patterns. Additionally, we developed an indoor people counting method using Wi-Fi CSI (Channel State Information) and context recognition techniques using Wi-Fi CSI with ultra-low power tags. These achievements were presented at various venues including international conferences and published in international journals.

Free Research Field

モバイルコンピューティング

Academic Significance and Societal Importance of the Research Achievements

Wi-Fi CSIを用いた状況認識技術は導入コストの低い手法として注目を集め，活発に研究が行われてきた.一方，これまでにWi-Fi CSIを用いた見守りや在室検知などいくつかのサービスが登場しているが，普及には至っていない.この原因として，対象が複数存在する場合にそれらの区別がつかない，明確な電波の変化が現れにくいため認識可能な行動が限られる，という課題が存在する.本研究成果は，これらの課題解決に資するものであり，Wi-Fi CSIを用いた状況認識技術の普及につながるいくつかのアプローチを確立できたと考えている.