No abstract

Recent studies demonstrate that Mobile Agent (MA) approach could be more effective than conventional client-server model in Wireless Sensor Network (WSN). Particularly, itinerary planning for MAs is a significant aspect of these approaches. Tree-based methods have been widely used for this purpose where many agents are dispatched simultaneously. Most of tree-based methods try to construct an optimal/suboptimal tree in terms of an objective function. In contrast, in this paper we introduce a new approach that tries to separate the MA dispatching and data fusion operations by considering two itinerary trees which results in more flexibility to regulate the itinerary plan. Based on this idea, we propose an algorithm called Two Trees Clone-based Itinerary (TTCI) which constructs two trees, one serves to distribute MA in the network and another to fuse back the sensed data. By experimental results, we demonstrate the performance improvement of TTCI algorithm in terms of overall energy consumption in comparison with the previous schemes. Meanwhile, the TTCI keeps the delay low.