Research On Data Aggregation In Wireless Sensor Networks

Wireless sensor networks(WSN) usually consist of a large number of tiny and finite-energy sensor nodes. These sensors deployed densely. As a result, there may be multiple sensor nodes monitoring in the same area, which will collect a lot of duplicate data. In addition, it will consume unnecessary energy if collect data more times from the same area. the data of the same area for multiple data collection will be unnecessary node energy consumption. In this context,the technology of data aggregation arises at the historic moment.The technology of data aggregation can solve a lot of problems, such as aggregation function,node energy, latency, and the accuracy of the data and so on. This paper focuses on the problem latency-minimizing data aggregation, that is minimizing the time slot of the transmission of the network as much as possible, which under the precondition of guarantee the data on the sensor can be received correctly, so as to ensure the effectiveness of data transmission. In this paper, we study the problem of latency-minimizing data aggregation respectively under the interference model(Protocol Interference Model and Physical Interference Model) in wireless sensor network.The protocol interference model is a relatively simple model in the aspect of compute.The study of the problem of data aggregation under the model has more mature. This paper analyzed the algorithm of latency-minimizing data aggregation under protocol interference model. And explain the shortcoming of the model. But physical interference model overcomes the shortcomings. It considers all of interference which concurrent transmission with one node in the network. Making the results more accurate. Therefore, the research is more significant under the physical interference model.Considering that latency-minimizing data aggregation in the physical interference model, this paper proposed a Hexagon-AS algorithm. First of all, our algorithm coverage the network with the hexagon cells; Second divided the network into layer and segment; Then coloring the cells of the network; Last schedule the cell with the same color. By the way, we got a data aggregation tree and a link scheduling set. The sensors can transmission and gathering data according to the structure of the data aggregation tree and the set of the link scheduling in the network. Through the analysis we found that the algorithm got an upper bound of the latency O(m), m is the layer of the hexagon in the networks. In order to verify the validity of the algorithm, we carried out the simulation experiment. The platform of the simulation is matlab7.0, we use C language to write.In the experiments, we set the number of sensor nodes in a network between 100 and 1000. We compared the latency of data aggregation for the different parameters of the experiment.We assigned different power to different nodes according to the transmission distance betweennodes. By this way, not only save the energy of the nodes, but also can guarantee the successful transmission of data. Through the analysis and simulation found that our Hexagon- AS algorithm performs better than the existing algorithm of Cell- AS in latency.