A study on pattern recognition of infrasonic multivariate time-series data

2016 Fiscal Year Final Research Report

• Project/Area Number: 26730021
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science
• Research Institution: Kyushu University
• Principal Investigator: OKUSA KOSUKE 九州大学, 芸術工学研究院, 助教 (30636907)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: パターン認識 / ストリームデータ / 多次元時系列データ / 低周波信号

Outline of Final Research Achievements

In this research, we focused on statistical analysis of multivariate low frequency stream data, especially the pattern recognition method. To confirm the performance of proposed method, we mainly focused on the sensing data analysis which is a typical example of multivariate low frequency stream data.
In this study, as a specific example, we focused on two cases of sensing data analysis. First case is developing the life-watching system using microwave Doppler rader. Second case is statistical indoor location estimation system using ToA/RSS measurement system. Finally, our proposed method showed good performance for both examples than previous methods.
These results were reported to peer-reviewed papers and international conferences.

Free Research Field

計算機統計学