The Application Of Ant Colony Algorithm In The Routing Of Wireless Sensor Networks

Since the beginning of the21st century, wireless sensor network technology isdeveloping rapidly. Scientists have developed the wireless sensor which hasauto-sense, data acquisition and integration and data transmission capacity. Thesesensor networks which consist of a number of individual sensors are widely used inall industries and received people’s great attention.Wireless sensor nodes can be randomly deployed and have a strongenvironmental adaptability. Through self-organization, they can collect original datawhich people need in any environment and at any time. The traditional data collectionmethod is that: in the monitoring area of WSN, each sensor node transfers collecteddata directly to the central processing node. However, this C/S data transmissionmode has many defects. Firstly, limited energy. Secondly, some errors probablyhappen because of the large amount of data processing tasks. Thirdly, duplication ofdata transmission makes the network more congested and the transmission delayincreasing.In order to solve the problems of wireless sensor networks in the C/S mode ofdata transmission, in recent years mobile agent technology is applied to the networkdata transmission research. With the combining of mobile agent technology and antcolony algorithm, a mobile agent routing algorithm based on improved ant colonyalgorithm is proposed in this paper. The mobile agents migrate in the network throughthe guidelines of algorithm. They collect the data of each sensor node and executedata fusion, then return to the central processing node to complete the task.The study results in this paper are as following:(1) The algorithm considers the distance between the sensor nodes, and nodeenergy factors. The energy factor is introduced into the calculation of the mobileagent route, to some extent, it balances the energy consumption of the network’ssensor nodes and avoids a single path of data transmission which can make networkshort-lived.(2) By updating local pheromone, increasing pheromone on global optimal pathas "reward" and decreasing pheromone on the worst path as "punish", the algorithmovercomes the defect that basic ant colony algorithm is easily to fall into localoptimal solution. This algorithm saves the time spent by the mobile agent to complete the task distributed by the central sensor node, and shortens mobile agent’smigration path.Comparing the proposed algorithm with the mobile agent routing algorithmbased on ant colony algorithm and genetic algorithm-based mobile agent routingalgorithm through experimental simulation results, the proposed algorithmsignificantly improves convergence speed and globalization of the search. To a certainextent, the proposed algorithm overcomes the problem of massive data transmissiondelay and balances energy consumption of the network, therefore it extends thenetwork lifetime.