Research Of WSN Routing Protocols Based On The Differential-Ant Colony Algorithm

As one of the most important supporting technology of the Internet of things, Wireless Sensor Network has attracted much attention in current international research field, and it has been widely applied in special areas of remote monitoring,military defense, environmental detection, the remote control of danger zone, etc. At the same time, the routing algorithm as the key support technology of network transmission, will be developed with the appearance of the new technology of network. The sensor nodes have the characteristic of low cost, small volume, its energy is limited because of the using of micro battery power and they are difficult to change because of the remote or hostile environment. So the WSN routing protocol’s design will focus on the realization of the network energy balance and the reduction of energy consumption, then the network energy efficiency will be improved and the network life cycle will be prolonged.As a kind of algorithm to solve the combinatorial optimization problems, Ant Colony Optimization has received much attention by academic circles, which has a strong dynamic adaptability, and it’s characteristics are similar with WSN. In the thesis, several proposed WSN routing protocols based on Ant Colony Optimization at home and abroad are analyzed and compared, then its disadvantages are found, such as long search time, slow convergence speed and easy to fall into local optimum. On this basis, DEACO Wireless Sensor Network routing protocol is proposed, which is combined Ant Colony Optimization with Differential Evolution Algorithm. Firstly, the "pheromone" that the ant colony left in the path is served as the action object of Differential evolution operator, then the optimal solution is got by Differential Evolution Algorithm. Secondly, the problem is solved by the the characteristics of positive feedback and parallelism of Ant Colony Optimization. Finally, the amount of pheromone left before is optimized by the method of disturbing the random deviation in the Differential Evolution Algorithm to generate new individual, in order to obtain better pheromone distribution, and the optimal path is obtained.Differential Evolution Algorithm has the ability of strong global optimization, it is better than other intelligent algorithms in solving the overall performance of combinatorial optimization, and has less controlled parameters. It is expected that this algorithm can be combined with Ant Colony Optimization to make up for the shortage, thus effectively prolong the life-cycle of the WSN and improve the network performance. In order to verify the validity of the algorithm, the thesis makes use of network simulator NS2 to test the DEACO protocol, and compared with ACO routing protocol which is based on the basic Ant Colony Optimization.The thesis adopts three performance indicators to measure the QoS of the two WSN routing protocol,respectively are the network throughput, packet loss rate and average end-to-end transmission delay;In order to fully illustrate the rationality of the WSN routing protocol which has put forward, at the end of simulation,the thesis will present the energy loss curve of the two kinds routing protocol which are showed before.Edit corresponding gawk code aiming at Trace file which is the results of the simulation,and the results of the analysis are three kinds of data under different contract rate, finally using Gnuplot draw the comparison chart curve of the above-mentioned routing protocols.Results show that the presented DEACO optimization strategy is better than the ACO routing protocol, especially in the following aspects :node energy consumption, throughput, average delay and network life cycle. Therefore, the results of the simulation prove the validity of the proposed optimization routing protocol.