Research On Wireless Sensor Networks Security And Data Aggregation Technology

The industry of wireless sensor network has an extensive application.However,due to limitations of hardware and environment,the WSN nodes have been found with the following problems,including unsustainability of energy and weak operational capability.These problems give rise to a higher requirement of the WSN in terms of the network protocol design,safety evaluation model building,and energy algorithm design.The data aggregation technology can first eliminate the redundant information from the data monitored by the adjacent sensor nodes and then send the simplified valid data to the sink nodes.It is of vital importance to improve data aggregation efficiency,increase data collection accuracy,reduce the overall network energy consumption,and extend the network life cycle.How to design a good network system model and fully evaluate its overall security before its operation so as to avoid data damage or loss or even system crashes is of equal research significance.This paper mainly studies the wireless sensor data aggregation technology and its network security evaluation technology.The research has been approved as the National Natural Science Foundation of China(NSFC)Project(No.61371071),the NSFC "Youth Foundation" Project(No.61201159),the Beijing Natural Science Foundation Project(No.4132057),the Beijing Science and Technology Plan Project(Z121100007612003),and the China Electronics Technology Group Corporation(CETC)Key Lab Open Foundation Project.The main contributions and innovations of the dissertation are as follows:(1)The basic characteristics and system model of the wireless heterogeneous network are studied.A method to test sensitive events in the data aggregation of the heterogeneous network is proposed.Though the traditional event test methods have improved data aggregation efficiency,the loss rate of key data is high during the aggregation process.As a result,some sensitive events are not reported.This paper introduces the genetic machine learning algorithm to data aggregation,and makes use of the algorithm's guiding idea for data screening.The simulation results show that the sensitive event test method put forward by this paper for the heterogeneous network can effectively improve the sensitive event test efficiency and accuracy.(2)Currently,in coping with redundancy reduction,many data aggregation algorithms fail to effectively identify the boundary threshold value.According to the idea of distributed clustering,this paper studies the sensor data quick grouping method of the K-means clustering algorithm.On the basis,an intranet data aggregation algorithm based on the dynamic optimal weight distribution(abbreviated as DOWA)is proposed.According to the optimal condition of the minimum total mean square error,this algorithm uses the minimum mean square error method to realize the dynamic optimal weight allocation,which can then guarantee optimization of data aggregation.Simulation experiment shows that the DOWA algorithm can provide efficient decision-making for data aggregation,reduce the data redundancy,and improve the data measurement precision.(3)The current mainstream data aggregation tree technology and its functional model are studied.Though the existing data aggregation tree algorithms can effectively conduct data aggregation,their energy consumption is high and imbalanced,which might lead to premature failure of the cluster nodes and shortening of the system network life cycle.In view of the current tree data aggregation algorithms,this paper proposes a low-energy-consumption data aggregation algorithm based on the improved polymerized tree model.In this algorithm,data aggregation is realized through the idea of the support matrix.Experimental results demonstrate that the algorithm can not only effectively bring down the system energy consumption,but also improve the network life cycle.(4)At present,many WSNs,after being designed and ready for operation,do not undergo a comprehensive evaluation of their security.If the overall security of the system is not high,loss of important system data or even system crashes might be caused.This paper studies the intuitionistic fuzzy algorithm.Combining characteristics of the WSN and grading of the network security,an WSN security evaluation algorithm based on the interval-valued intuitionistic fuzzy information model is proposed.The algorithm statistically collects the interval-valued intuitionistic fuzzy data based on the interval-valued intuitionistic fuzzy hybrid geometric(IVIFHG)operators.Then,the MADM problem model is adopted to evaluate the wireless sensor security problems.Finally,an example is adopted for algorithm verification.