回归的能源有效网络大数据流汇聚算法研究

为了降低传感器网络数据流汇聚时的能源消耗,提出了一种基于回归的能源有效数据流汇聚算法。首先,将传感器节点分为活跃节点和能源有效节点。然后,以活跃节点为中心点将所有节点进行聚类,并应用回归方法通过活跃节点的数据流对能源有效节点的数据进行预测。接下来,通过节点预测值的累积误差不断修正活跃节点集。最后,应用活跃节点的数据流信息对能源有效节点的数据进行预测。实验表明,本文提出的算法与其它相关算法相比具有更好的预测准确性。

Regression Based Power-efficient Data Streams Aggregation Algorithm

In sensor networks, power consumption is a critical factor in aggregating data streams. In order to decrease the power consumption while aggregating data streams in sensor networks, this paper proposed a regression based power-efficient data streams aggregation algorithm. Firstly, we classified the sensor nodes into active nodes and power-efficient node. Secondly, we clustered all nodes using the active nodes as the center nodes, and predicted the data of power-efficient nodes using the active nodes with regression method. Thirdly, we modified the active node set with the cumulative error of the predicted data. Finally, we predicted the data of power-efficient nodes with the active nodes. The experiments show that, the proposed method is more accurate than related works while predicting.