摘要： 为减少无线传感器网络的数据流量，提出了一种基于预测的数据分发机制。在源节点和sink节点分别配置相同参数的预测器，用以预测未来采样值。当在源节点处预测值和实际采样值的差异大于某门限值时，当前采样数据才会被传送到sink节点。预测器采用双指数平滑预测模型，根据历史数据去预测未来数据。实验表明，该节能的数据分发机制可以滤除83%的冗余数据而仍然保持较高的精度。

关键词: 无线传感器网络, 预测模型, 节能策略

Abstract: In order to decrease the traffic in the wireless sensor networks, a data dissemination scheme based on forecast is presented. Two matching predictors with the same parameters are deployed at the source and the sink respectively, which predict the next immediate sampling value simultaneously. The sampling value is sent to the sink only when the difference between the actual and the predicted value at the source exceeds a certain threshold. The predictors adopt the double exponential smoothing forecast model to estimate the future values based on the past samples. The experiment results show that this energy-saving data dissemination scheme is very effective, which can filter about 83% data and remain high precision yet.

Key words: Wireless sensor networks, Forecast model, Energy-saving policy

中图分类号: 

• TP393