面向移动群智感知的多任务分发算法

发布时间：2018-06-05 08:53

本文选题：移动群智感知 + 机会通信　；参考：《计算机应用》2017年01期

【摘要】：针对在移动群智感知中基于机会通信完成数据传输会消耗大量时间成本的问题,提出了一种基于中枢节点的多任务分发(HTA)算法。该算法利用节点在移动网络中社交关系属性不同的特点,通过中枢节点选择算法将部分节点作为中枢节点,并将其用于协助任务请求节点分发任务。在任务请求节点与中枢节点相遇时,同时给中枢节点本身和它的从属节点分配任务,并由中枢节点负责向从属节点分发任务与回收任务结果。基于The ONE模拟器进行实验,与在线任务分配(NTA)算法相比,HTA算法时间成本平均降低了24.9%,同时任务完成率平均提高150%。实验结果表明,HTA算法能够提高任务的完成速度,降低时间成本消耗。
[Abstract]:In order to solve the problem that data transmission based on opportunity communication consumes a lot of time cost in mobile swarm intelligence perception, a multi-task distribution (HTA) algorithm based on central nodes is proposed. This algorithm takes advantage of the different attributes of social relations between nodes in mobile networks and uses some nodes as central nodes through the central node selection algorithm and uses them to assist task request nodes to distribute tasks. When the task request node meets the central node, the central node itself and its subordinate node are assigned tasks, and the central node is responsible for distributing the task to the subordinate node and retrieving the results of the task. Experiments based on the one simulator show that the time cost of the HTA algorithm is 24.910% lower than that of the on-line task assignment (NTA) algorithm, and the task completion rate is increased by 150% on average. The experimental results show that the HTA algorithm can improve the task completion speed and reduce the time cost.
【作者单位】：北京信息科技大学计算机学院;网络文化与数字传播北京市重点实验室(北京信息科技大学);
【基金】：国家自然科学基金资助项目(61370065,61502040) 北京市优秀人才培养资助青年骨干个人项目(2014000020124G099) 网络文化与数字传播北京市重点实验室资助项目(ICDD201406) 现代测控技术教育部重点实验室/机电系统测控北京市重点实验室资助项目(KF20151123205)~~
【分类号】：TN929.5

【相似文献】