Research On Multi-sensor Information Fusion Technology Of WSN Node

Wireless sensor networks?WSNs?technology has comprehensive processing capabilities such as sensing,data processing,storage,and wireless communication.It has the characteristics of multiple disciplines,broad knowledge,and wide application scope.It is the main technical support for building an intelligent sensing environment.With the rapid development of sensor technology,wireless communication technology and Internet of Things technology,the research on WSN technology has been further deepened.As the forefront of the network,the WSN node's data processing capabilities have a significant impact on the security,efficiency,and intelligence of the network.The thesis takes WSN nodes as research objects,and focuses on the research of multi-sensor information fusion technology related theories and algorithms.The main work includes:?1?The related knowledge of WSN and multi-sensor information fusion is briefly introduced,and a multi-sensor information decision algorithm based on distance vector is proposed.This algorithm uses the characteristics of sensor information decision-making,combined with the system evaluation criteria,through the normalization process of the support degree matrix,to obtain the dominance degree function.After normalization,the decision results are ranked,and the maximum value is the optimal solution.At the same time,the discrimination function is defined,which provides a scientific basis for algorithm evaluation.Simulation analysis shows that the method has a recognition discrimination of more than 0.5 in sensor information decision-making.At the same time,because it avoids the probabilistic assignment of sensor data in the calculation process,compared with other methods that require probabilistic assignment of sensor data in decision-making,Algorithm has obvious advantages.?2?A data fusion algorithm using the power mean operator and D-S evidence theory is proposed.Use the multi-sensor information redundancy to remove outliers from multi-sensor collected data.In order to reduce the increase in calculation caused by excessive use of redundant information,a dynamic sliding window method is proposed to reduce the data usage,combined with the power mean operator Data fusion with DS evidence theory.At the same time,using the structural characteristics of wireless sensor networks,a hierarchical decision-making idea is proposed,and conflicts of DS evidence theory are set up by setting evidence thresholds.Simulation results show that this method can effectively avoid the conflict problem of D-S evidence theory,and the recognition rate of the result H₁ reaches 0.81,which has the characteristics of good decision-making effect and simple calculation process.