Transfer learning for context recognition based on Wi-Fi channel state information

2023 Fiscal Year Final Research Report

• Project/Area Number: 21H03428
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 60060:Information network-related
• Research Institution: Osaka University
• Principal Investigator: Maekawa Takuya 大阪大学, 大学院情報科学研究科, 准教授 (50447025)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: Wi-Fiセンシング

Outline of Final Research Achievements

testIn this research, we studied environment-independent context recognition using Wi-Fi radio wave information. Specifically, we studied recognition and estimation methods for the following contexts that are important in mobile and ubiquitous computing. (1) Angle of Arrival (AoA) information of Wi-Fi radio waves: Estimating the angle of arrival of received radio waves by using Wi-Fi propagation information. (2) Physical distance between Wi-Fi receivers: Estimating the physical distance between Wi-Fi devices, such as smartphones using information on received radio waves from nearby Wi-Fi access points.

Free Research Field

ユビキタスコンピューティング

Academic Significance and Societal Importance of the Research Achievements

本研究では、Wi-Fiセンシング研究において環境依存性の低い技術に注力を置いて研究開発を行った。環境に依存しないWi-Fi受信情報の成分を極力利用して、上述するような様々な屋内コンテキストを推定する手法を開発し、今後の様々なコンテキスト依存応用に繋がる基盤的な技術を築いた。本研究を通じて、IEEE Sensors JournalやElsevier Pervasive and Mobile Computingに論文が採録されるなどの顕著な成果が得られた。