Research on WSN Malicious Node Removal Algorithm Based on Secure Data Fusion

WSN, which is regarded as the second largest network after it, closely links the logical information world with the objective physical world, changes the way of communication between man and nature, and brings greater convenience to people's daily life. Due to the important position of the mobile wireless sensor network, the security threats have increased, especially through certain means to invade the mobile wireless sensor network and take measures such as data collection, information forgery and paralysis attack, which can easily lead to serious consequences such as abnormal operation of large-scale networks. Therefore, based on secure data fusion, this paper studies the malicious node removal algorithm of WSN, which reduces the influence of malicious nodes on the fusion result. Compared with this, the fusion result of ANN (Artificial Neural Network) algorithm is closer to the actual value. In our future work, we will expand our method and consider the rotation of converged nodes and multi-hop routing between converged nodes and base stations. Nodes in abnormal state are classified by node aggregation, which prevents malicious nodes from being missed.