Research On WSN Compressed Data Collection Method Based On Mobile Agent

Wireless Sensor Network is a resource-limited functional network that performs monitoring tasks for a long time by transmitting sensing information to Sink node.The transmission,calculation and storage capabilities of sensor nodes are very limited.Once the energy of the sensor node is exhausted,it is difficult to supply it.Therefore,how to efficiently and energy-efficiently perform WSN data collection tasks,and balancing network energy consumption and extending network lifetime is an important challenge in the WSN project.From the perspective of optimizing network performance,combined with compressed sensing theory and mobile agent technology,this paper designs a WSN compressed data collection method based on mobile agent,and uses the advantages that mobile agents can migrate independently among network nodes,a scheme for matching the WSN compressed data collection model with the routing mechanism is given.The main works of the thesis are as follows:(1)Firstly,studied the mobile agent model,and proposed a mobile agent itinerary planning algorithm based on tree(ITMIP).Adaptively partitioning multiple sub-areas according to mobile agent migration characteristics and nodes distribution,and the optimal number of partitions is fitted,the mobile agent is assigned to the sub-area,and the optimal itinerary of the mobile agent is obtained through a tree-based greedy strategy.In comparison,the algorithm has less network energy consumption and shorter service delays;(2)Based on the ITMIP algorithm,a mobile agent-based compressed data collection model is designed,which fully considers the problems that may be applied to the WSN data collection in the application of compressed sensing.The idea of group projection is innovatively proposed,and the number of nodes in a single projection is increased.The mobile agent is allocated for compressed collection.Compared with the existing methods,the method effectively reduces the network transmission volume and has better energy consumption balance;(3)Finally,in view of the quality of the compressed data collected,a statistical model is proposed to update the sparse basis and measurement matrix periodically and adaptively,to meet the requirement of data reconstruction accuracy.The algorithm has better data reconstruction performance,and further improves the proposed model.