| Wireless Sensor Networks (WSNs) are one of the key technologies of Internet of Things. Asan emerging network technology and computing model, WSNs cross computer, communications,electronics and other fields, and have been widely used in many domains, such as scientificobservation, disaster relief, environmental monitoring, and industrial control. Usually, peopledeploy various WSNs in different environment to obtain variety of perceptive data. Then theyextract useful information from them, which may improves the understanding about the objectiveworld. During the whole process, the data processing technique is the key. In WSNs, the designand optimization of data processing protocols is constrained by two factors. Firstly, WSNs areresource-constrained networks, so any technology should follow the prerequisite of usingresources efficiently. Secondly, WSNs are application-related networks, so the design andoptimization of these protocols should be consistent with the characteristics and requirements ofapplications. Starting from the two concepts, this dissertation researched several issues related todata processing, and discussed the status and progress of them. Then their deficiencies and defectswere discussed and some improvements were given. Overall, this dissertation made the followingcontributions to the WSNs.(1) The issue of data collection in wireless sensor networks was studied. In this dissertation, amobile sink node was introduced and a data collection scheme with regular track (DCSR) wasproposed and realized. DCSR consisted of two phases. Firstly, a number of collection sites wereselected according to the sensor distribution. And then a quantum genetic algorithm was performedto calculate the shortest loop across the sites. Simulation tests were carried out. Simulation resultsshowed that DCSR had better performance in network throughput and energy efficiency, and itcould collets more data.(2) The issue of data transmission in wireless sensor networks was studied. The video datacollection was chosen as the object of the study, and the quality of service (QoS) problem of videotransmission was analyzed. Video applications were generally characterized by large data size andmany QoS requirements, which made the transmission of video always a hard problem. In thisdissertation, a cross-layer and multipath based video transmission scheme (CMVT) was presented.CMVT introduced the idea of differentiated service and multipath routing, and operated in bothapplication layer and network layer. In application layer, different types of video frames weredistinguished and marked with different tags. Then in network layer these frames were identified by the tags and CMVT forwarded them in different paths. In this way, the key frames could besent to the sink node reliably. Simulation tests were carried out, and simulation results showedthat CMVT extended the network lifetime greatly and provided better assurance for quality ofvideo applications.(3) The issue of data aggregation in wireless sensor networks was studied. Data aggregationwas an efficient in-network data processing method. It reduced data redundancy and improvedinformation quality, which may save communication energy and increase collection efficiency. Inthis dissertation, a temporal and spatial correlation based data aggregation scheme (TSDA) wasproposed and realized. TSDA used a cluster structure based network, and it had three parts. Firstly,the cluster head would not submit any packets derived from the cluster members. Instead, it fusedthem into an outcome packet and then sent it to the sink node. Secondly, the cluster headscheduled the nodes with low energy level into sleep. Thirdly, the cluster head would use acombined forecasting algorithm to estimate the data of sleeping nodes. Simulation tests werecarried out, and simulation results showed that TSDA had good performance. It not only extendedthe network lifetime greatly, but also provided better assurance of data quality.(4) The issue of data storage in wireless sensor networks was studied. This dissertationmainly discussed the distributed storage technique, and focused on the problem that differentnodes had different collection rate. Through analysis, this problem was converted into a k-medianproblem in non-metric space, and three distributed storage solutions were proposed: randomstrategy based data storage scheme (RDS), reverse greedy strategy based data storage scheme(GDS) and SQGA （small world model based quantum genetic algorithm）based data storagescheme (SDS). In the first strategy, k nodes were selected randomly, and then each node chose thenode with least cost as its storage node. In the second strategy, a method used in k-medianproblem of metric space was applied to the selection of storage nodes. In the third strategy, animproved quantum genetic algorithm SQGA was used to compute the data storage problem. RDS,GDS and SDS were tested in this dissertation. Simulation results showed that GDS, SDS hadgood performance. Furthermore, taking the balance of node energy consumption into account,SDS performed better. |
PDF Full Text Request