论文部分内容阅读
无线传感器节点持续感知的数据具有高度的时间关联性,所产生的冗余数据会给传感器节点带来不必要的数据传输和能量消耗.数据预测算法通过预测节点的感知数据序列可以有效避免上述问题.提出一种基于马尔科夫链的数据预测算法(MC-DP),该算法首先对传感器节点感知的数据进行弱化处理来提高数据序列的光滑性,然后采用离散灰色预测模型对节点需要向sink传输的数据序列进行预测,当节点预测的数据精度不够时,进一步使用基于马尔科夫链的数据修正过程来对其进行改进.实验结果表明,MC-DP算法预测准确率更高,预测序列的数据误差率更低,使用该预测算法使得传感器节点可以节约更多能量.
The continuously sensed data of wireless sensor nodes have a high degree of temporal correlation, and the redundant data generated will bring unnecessary data transmission and energy consumption to the sensor nodes. The data prediction algorithm can effectively avoid the above problem by predicting the perceived data sequence of the nodes A data prediction algorithm based on Markov chain (MC-DP) is proposed in this paper. Firstly, the data perceived by sensor nodes are weakened to improve the smoothness of the data sequence. Then, Which is used to predict the accuracy of the data when the node is not accurate enough, and further improve the data using Markov chain based data modification.The experimental results show that the MC-DP algorithm has higher prediction accuracy and predictive sequence The data error rate is lower, and using this prediction algorithm allows sensor nodes to save more energy.