一种低功耗的WSN数据多节点联合预测冗余控制模型  

A Low Power WSN Data Control Model with Multi-Node Joint Predictive Redundancy

作　　者：汪宏海[1,2] 

机构地区：[1]浙江旅游职业学院,浙江杭州311231 [2]苏州大学计算机科学与技术学院,江苏苏州215000

出　　处：《安阳师范学院学报》2017年第5期30-37,共8页Journal of Anyang Normal University

摘　　要：随着物联网的快速发展,无线传感器及其网络的应用进入了一个全新的高度,而传感器节点的寿命和能耗问题一直制约着其进一步发展。基于此,本文提出了一种预测回归控制模型,能够对节点进行回归预测分析,通过将融合多个节点构建多元回归模型,能够对历史数据进行融合,从而能够提高单个节点的准确性并有效降低能耗,还可以提供一定的节点冗余支持,即在部分节点失效时仍可采用控制模型进行较为准确的预测。通过在样本区域部署测试系统,结合"小气候"特点对控制模型进行了测试,验证了预测回归模型的有效性。With the development of internet, WSNs （Wireless Sensor Networks） and its applications are going to the new height. However, both lifetime and energy consumption of sensor nodes have restricted its further development. In this paper, a joint predictive regression control model is proposed, which implement regres-sion prediction Regression analysis for single - node and multi - node in the region at the same time, and then merge the result. Therefore, our method can improve the measurements accuracy of individual nodes and re-duce their energy consumption, and provide redundancy support for nodes as well, which can still implement prediction utilizing control model even if some nodes are not working. By deploying the actual test system in the sample area and incorporating ＂ microclimate ＂ feature into test, the validation of the regression model is examined.

关 键 词：无线传感器网络 回归预测 能耗控制 系统冗余 

分 类 号：TP393.03[自动化与计算机技术—计算机应用技术]