Exploiting multi-spatial correlations of motion data in a body sensor network

Chun Hao Wu*, Yu-Chee Tseng

*Corresponding author for this work

研究成果: Article同行評審

4 引文 斯高帕斯(Scopus)


Human body motions usually exhibit a high degree of coherence and correlation in patterns. This allows exploiting spatial correlations of motion data being captured by a body sensor network. Since human bodies are relatively small, earlier work has shown how to compress motion data by allowing a node to overhear at most κ = 1 node's transmission and exploit the correlation with its own data for data compression. In this work, we consider multi-spatial correlations by extending κ = 1 to κ > 1 and constructing a partial-ordering directed acyclic graph (DAG) to represent the compression dependencies among sensor nodes. While a minimum-cost tree for κ = 1 can be found in polynomial time, we show that finding a minimum-cost DAG is NP-hard even for κ = 2. We then propose an efficient heuristic and verify its performance by real sensing data.

頁(從 - 到)662-665
期刊IEEE Communications Letters
出版狀態Published - 1 五月 2012

指紋 深入研究「Exploiting multi-spatial correlations of motion data in a body sensor network」主題。共同形成了獨特的指紋。