一种基于自适应信息素蒸发系数的WSN蚁群路由算法-《郑州大学学报(工学版)》

文章信息/Info

Title:: An Adaptive Pheromone Evaporation Coefficient Based Ant Colony Routing Algorithm for Wireless Sensor Networks

作者:: 王恭¹; 孙铭阳¹; 孙汇阳²; 腾子铭³; 1.东北电力大学自动化工程学院!吉林吉林 132012；2.北京电子科技学院密码科学与技术系，北京 100070；3.吉林大学通信工程学院!吉林长春 130012

Author(s):: WANG Gong¹; SUN Mingyang¹; SUN Huiyang²; TENG Ziming³; 1.School of Automation Engineering, Northeast Electric Power University, Jilin 132012, China;
2.Department of Cryptographic Science and Technology, Beijing Electronic Science and Technology Institute, Beijing 100070, China; 3.College of Communication Engineering, Jilin University, Changchun 130012, China

Keywords:: wireless sensor network; loop effect; adaptive pheromone evaporation coefficient; routing; node energy

摘要:: 目前无线传感器网络中存在蚁群环路现象和网络节点能量分布不均衡等问题，容易导致节点过早休眠、网络生存周期短。本文以现有蚁群算法为基础，在蚂蚁数据包中增加前向蚂蚁数据包序列号、前向蚂蚁数据包源地址和蚂蚁数据包生存时间等数据，并引入自适应信息素蒸发系数，将路由跳数修正为多跳消耗的能量值，提高信息素更新公式的准确性，使网络中各节点的能量消耗更均衡。通过仿真实验对比，本文算法平均剩余能量提高了9.6%，证明该算法可有效削弱蚂蚁环路效应，平衡网络节点能量，延长网络寿命。

Abstract:: At present, the ant colony loop phenomenon and uneven energy distribution of nodes in the wireless sensor networks could cause nodes to go dormant prematurely and shorten the network lifetime. To improve the accuracy of the pheromone update formula and further balance nodes energy consumption, the following improvements could be made based on the original ant colony algorithm: add data into ant data packet, such as the sequence number of the forward ant data packet, the source address on the packet of the forward ant, number of packet path nodes, energy consumed by relay nodes, length of path, survival time of the ant data packet, initial energy and average remaining energy of packet. The adaptive evaporation coefficient could be introduced into the pheromone update formula whose number of routing hops was altered to the energy consumption of multiple hops; the pheromone increment formula could be improved, the number of nodes visited by packets was redefined as the node energy loss function. In experimental results, there was a 5.7 percent reduce in the shortest path. It was obvious that this algorithm could effectively mitigate the ant loop effect, ba-lance nodes energy consumption and extend the network lifetime.