Research On Multiple Agents Routing Planning For Wireless Sensor Networks

Nowdays，the use of wireless sensor network is getting more and more widely,it can not only be used in battlefield monitoring, but also in home automation,to alarge extent, we can say wireless sensor network becomes pervasive.In the past, client/server model is the base of data collection and fusion inwireless sensor network, with the burgeon of large and complicated sensornetwork, this kind of model could not meet people’s needs any more. Somepeople brought mobile agent into the wireless sensor network, this model leads torapid developmet in sensor network. In the sensor network, the problem ofItinerary Planning is always tough and complex, but is also essential, afterintrouducing the mobile agent, the relevant problems also come.For single mobile-agent model，the existent method and theory is perfectlydeveloped，in these years，some ambitious scientists have proposed the conceptof Multi-agent in wireless sensor network, and put the problem of multi-agentitineray planning on the agenda,lots of people try hard to solve thess kinds ofproblems, until now,the ponderable method is still based on four steps.In this thesis, the following work has been done, firstly, implemented theavailable three representative multiple mobile agents routing planning methods inwireless sensor network,and then also proposed a new routing planning based onnovel genetic method, and through the OPNET simulation, we find that the newgenetic algorithm method to multiple mobile agents routing planning is muchmore efficient, although it still may have some drawbacks in some respects,anyway, it changes the traditiol four-step method, and provide us with a newview, which is serviceable.