In a wireless sensor network, sensor nodes transmit sensor data with information about detected events including local clock values at times when occurrences of the events are detected to a fusion center. The sensor data is stored in a database and analyzed. Since local clocks in wireless sensor nodes are usually not synchronized precisely due to unexpectable transmission delay of control messages for clock synchronization which requires high communication overhead, this paper proposes sensor data synchronization which adjust the local clock values associated to stored sensor data. For local clock value adjustment, this paper proposes a novel method based on estimation of relative skew and offset between local clocks in neighbor wireless sensor node whose observable areas are overlapped. Here, estimation of commonly detected events by the wireless sensor nodes is a key technique. By our proposed method, with no additional communication overhead, local clock values are consistently adjusted to analyze their temporal relations such as precedence ones.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Sensor Data Synchronization Based on Estimation of Commonly Detected Events in Wireless Sensor Networks

  • Katsuhiro Nagano,
  • Hiroaki Higaki

摘要

In a wireless sensor network, sensor nodes transmit sensor data with information about detected events including local clock values at times when occurrences of the events are detected to a fusion center. The sensor data is stored in a database and analyzed. Since local clocks in wireless sensor nodes are usually not synchronized precisely due to unexpectable transmission delay of control messages for clock synchronization which requires high communication overhead, this paper proposes sensor data synchronization which adjust the local clock values associated to stored sensor data. For local clock value adjustment, this paper proposes a novel method based on estimation of relative skew and offset between local clocks in neighbor wireless sensor node whose observable areas are overlapped. Here, estimation of commonly detected events by the wireless sensor nodes is a key technique. By our proposed method, with no additional communication overhead, local clock values are consistently adjusted to analyze their temporal relations such as precedence ones.