A Task Execution Framework for Cloud-Assisted Sensor Networks  

作　　者：石海龙 李栋 邱杰凡 侯陈达 崔莉 

机构地区：[1]Institute of Computing Technology,Chinese Academy of Sciences [2]University of Chinese Academy of Sciences

出　　处：《Journal of Computer Science & Technology》2014年第2期216-226,共11页计算机科学技术学报（英文版）

基　　金：supported in part by the Strategic Priority Research Program of the Chinese Academy of Sciences under GrantNo.XDA06010403;the International Science and Technology Cooperation Program of China under Grant No.2013DFA10690;the ational Natural Science Foundation of China under Grant No.61003293;the Beijing Natural Science Foundation under GrantNo.4112054

摘　　要：As sensor networks are increasingly being deployed, there will be more sensors available in the same region, making it strategic to select the suitable ones to execute users＇ applications. We propose a task execution framework, named sTaskAlloc, to execute application energy efficiently by two main parts. First, considering that the energy consumption of an application is inversely proportional to the utilization rate of sensors, we present a hot sensor selection algorithm, HotTasking, to minimize the energy consumption of new added applications by selecting the most suitable sensor. Second, when a sensor is shared by multiple applications, proposed MergeOPT （a concurrent tasks optimization algorithm） is used to optimize energy consumption further by eliminating redundant sampling tasks. Experimental results show that sTaskAlloc can save more than 76% of energy for new added applications compared with existing methods and reduce up to 72% of sampling tasks when a sensor is shared by more than 10 applications.As sensor networks are increasingly being deployed, there will be more sensors available in the same region, making it strategic to select the suitable ones to execute users＇ applications. We propose a task execution framework, named sTaskAlloc, to execute application energy efficiently by two main parts. First, considering that the energy consumption of an application is inversely proportional to the utilization rate of sensors, we present a hot sensor selection algorithm, HotTasking, to minimize the energy consumption of new added applications by selecting the most suitable sensor. Second, when a sensor is shared by multiple applications, proposed MergeOPT （a concurrent tasks optimization algorithm） is used to optimize energy consumption further by eliminating redundant sampling tasks. Experimental results show that sTaskAlloc can save more than 76% of energy for new added applications compared with existing methods and reduce up to 72% of sampling tasks when a sensor is shared by more than 10 applications.

关 键 词：task execution framework sensor network concurrent task optimization Internet of Things 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]