Detecting Significant Locations from Raw GPS Data Using Random Space Partitioning
スポンサーリンク
概要
- 論文の詳細を見る
We present a fast algorithm for probabilistically extracting significant locations from raw GPS data based on data point density. Extracting significant locations from raw GPS data is the first essential step of algorithms designed for location-aware applications. Most current algorithms compare spatial/temporal variables with given fixed thresholds to extract significant locations. However, the appropriate threshold values are not clearly known in priori, and algorithms with fixed thresholds are inherently error-prone, especially under high noise levels. Moreover, they do not often scale in response to increase in system size since direct distance computation is required. We developed a fast algorithm for selective data point sampling around significant locations based on density information by constructing random histograms using locality-sensitive hashing. Theoretical analysis and evaluations show that significant locations are accurately detected with a loose parameter setting even under high noise levels.
著者
-
Morikawa Hiroyuki
Rcast The University Of Tokyo
-
Kami Nobuharu
System Platforms Research Laboratories Nec Corporation
-
Yoshikawa Takashi
System Platforms Research Laboratories Nec Corporation
-
BABA Teruyuki
System Platforms Research Laboratories, NEC Corporation
-
Ikeda Satoshi
System Platforms Research Laboratories, NEC Corporation
関連論文
- B-20-58 Appliance Recognition Using Power State Segregation at Low Sampling Rate
- Study on Dissemination Patterns in Location-Aware Gossiping Networks
- Detecting Significant Locations from Raw GPS Data Using Random Space Partitioning
- Detecting Significant Locations from Raw GPS Data Using Random Space Partitioning
- Study on Dissemination Patterns in Location-Aware Gossiping Networks
- PAVENET OS : A Compact Hard Real-Time Operating System for Precise Sampling in Wireless Sensor Networks