A Location Based Data Gathering Scheme For Mobile Sensing Network

Due to the explosive popularization of mobile terminal devices which are embedded with various sensors like smart phone,tablet personal computer and so on,Men have become sensors,and together with mobile devices constitutes mobile sensor networks,with the mobile devices as basic perception unit,and collecting sensor data as the most important task of mobile sensor networks.This paper focuses on the design of efficient data collection method.In the previous data collection methods in opportunistic sensor network,sensor data was finally uploaded to the central server with the help of the communication opportunities provided by body movement and multiple forwarded,cooperative and hop-by-hop trans-missive methods supported by mobile devices.But the brief encounter of mobile devices by chance can not enable the construction of a reliable connection among each other and the transmission success rate of sensor data is relatively low.Considering the fact that in actual surroundings many places have been deployed Wi-Fi AP,and once men meet with AP,sensor data can be upload successfully,therefore we can compare these AP to Sink.It is different from ordinary mobile opportunistic network,its mobile devices didn't know the target node of the sensor data,and the goal of the transmission opportunity can be any Sink.The main contents of the paper can be summarized as follows:(1)We proposed an optimal Sink selection method.In regard to the problem of the selection of optimal Sink among many Sink mobile sensor network,definitions and assumptions will be conducted in the first place,and secondly analysis of the attributes of space,time and encounter probability between mobile sensor nodes and Sink will be carried out.The problem of selecting the optimal Sink will be compared to the solving of multi-objective optimization problem,and analytic hierarchy process of multiple objectives decision will be introduced to find out the optimal Sink or determine the target node of senor data.(2)We proposed an data forwarding method based on location.Through the analysis of human social behavior with the kind of social mobile sensor network,the phenomenon that human movement is closely linked to the location was found out.Inspired by PeopleRank algorithm,and with the help of mobile devices which can build stable connection to exchange data completely in certain distance,this paper has built a mobile sensor network model of HUNAN university,and related definitions have been given to sensor data,preference location and popularity,etc.And on this basis,forwarding method of sensor data based on the location has been proposed to optimize the next hop selective strategy.(3)At last,the feasibility and effectiveness of the proposed methods in this paper was validated through a series of experiments.In order to conduct quantitative comparison,the following three indicators were used as measurement basis:transmission rate,forwarding cost,forwarding delay.The experimental results showed that this method has not only greatly improved the transmission success rate of senor data,but also has lowered the cost and delay of forwarding significantly.