Abstract: The uncertainty of the moving speed of the target nodes in the sensor network, brings a serious challenge for dynamic ranging. An improved dynamic RSSI-based distance estimation method is proposed, which utilizes pattern matching in term of the sliding window. By measuring the received signal strength indicator in the communication process of nodes, the mapping relation between the received signal strength indicator and time is analyzed and determined. The linear treatment for the real-time RSSI data streams obtained is produced in the moving process. The sliding window pattern matching is used to realize high precision dynamic distance estimation for multi-mode speed mobile nodes, which contains uniform, uniform variable and variable acceleration nodes. The experimental test shows that the method can overcome the uncertainty of RSSI data, and can realize the dynamic distance estimate error for multi-mode speed mobile nodes less than 2.6% at the same time.

Key words: Uncertainty, dynamic ranging, data stream, sliding window, pattern matching

摘要： 针对传感器网络中目标节点移动速度的不确定性，给距离估计带来了现实挑战，论文提出了改进的基于滑动窗口匹配的动态距离估计方法。通信过程中，节点通过对信号强度（Received Signal Strength Indicator，RSSI）的测量，分析确定信号强度与时间（Received Signal Strength Indicator-Time，RSSI-T）的映射关系。基于此，在移动过程中，对实时获得的RSSI数据流进行在线线性处理，通过滑动窗口模式进行匹配，实现在匀速，匀变速和变加速多模式速度下的高精度动态距离估计。经实验测试，该方法在克服RSSI数据不确定性的同时，能对多模式速度的移动目标节点进行误差小于2.6 %的动态距离估计。

关键词: 不确定性, 动态距离估计, 数据流, 滑动窗口, 模式匹配