The Route Optimization Of WSN Based On The Combination Of Ant Colony Optimization And Patricle Swarm Optimization

The link of routing plays a very important part in energy conservation of Wireless SensorNetworks. Because the routing protocol determines the path of nodes communication, and thecommunication path affects the energy consumption of the communication. Considering theperformance of hierarchical routing protocol is better than flat routing protocols, this paperfocuses on the research of hierarchical routing protocol.(1) The improved Ant Colony Optimization algorithm is used to optimize the multi-hoppath of cluster-heads in WSN: it uses DCHS algorithm to set up clusters, then it uses theimproved Ant Colony Optimization to find out the optimal path that travels through all thecluster heads from the cluster head node which is nearest to the sink to sink node. This is animproved TSP(Traveling Salesman Problem) model. As the distance between eachcluster-head node is very short, this algorithm greatly reduces the energy consumption of thecommunication between each single cluster-head node and sink node. And the optimum pathin this algorithm is the global optimal path. It doesn’t mean that each cluster-head node hasone optimal path from it to the sink node. The improved ant colony optimization algorithm isreflected by that the heuristic function in selection probability formula is optimized.(2) The special particle swarm algorithm is used to optimize the selection of thecluster-heads in WSN: it uses DCHS algorithm to set up clusters in advance, followed by ituses special Particle Swarm Optimization algorithm to find out the optimal nodes which arebest suitable to be cluster-heads in each cluster. In the special Particle Swarm Optimizationalgorithm, every particle corresponds to a cluster and only jumps in the corresponding cluster,and each particle doesn’t repeat the nodes which have been through by. Therefore, theiteration number decreases considerably. And because the global extremum factor doesn’thave reference value to the selection of the cluster-heads, this algorithm doesn’t have theglobal extremum factor. This algorithm is characterized by less number of iterations and highefficiency, it also can prolong the cycle when the first dead node appears effectively.(3) These two algorithms mentioned above are combined together, they respectively acton the cluster set-up phase and the phase of the communication between cluster-head nodesand sink node. The PSO-ACO algorithm which is the combination of these two algorithms,compares with these two algorithms at performance by the simulation diagram. Accordingtentative verification, the PSO-ACO algorithm greatly balances the network energyconsumption, and it has a considerable improvement in prolonging the cycle when the firstdead node appears and the lifecycle of wireless sensor network.