Research On Secure Routing Algorithms For Energy Harvesting Wireless Sensor Network

The Internet of Things(Io T)technology is already becoming the fastest-developing and potentially one of the most important fields in the world’s information industry,guiding the world economy towards a new era—the era of smart manufacturing.Among the kinds of Io T technologies,the wireless sensor network(WSN)is drawing great concern for its ability to integrate many low-cost,low-power sensors.However,wireless sensor network is usually distributed in an exposed environment and is vulnerable to various types of network attacks,which may cause the network monitoring data to be leaked or damaged or even render the network unable to work properly.Sensor nodes using batteries as an energy source have limited energy,especially the nodes closer to the sink nodes,which have higher energy consumption and are prone to producing "energy holes".Recently,the development of energy harvesting and conversion technologies has driven the advent of energy harvesting wireless sensor networks(EH-WSN),which enable the nodes in the network to use the abundant renewable energy sources in nature as a source of energy.Nevertheless,renewable energy sources in the environment are often intermittent and unstable.Therefore,how to improve the ability of EH-WSN to resist different types of attacks,balance the energy consumption of each node in the network,and plan and utilize renewable energy more rationally are the research focuses of EH-WSN secure routing algorithms.Based on the characteristics and security requirements of self-powered wireless sensor networks,this dissertation constructs a multi-factor relay node evaluation model and proposes a secure routing algorithm that takes into account network security and balanced node energy consumption.The main contents of the dissertation are as follows:(1)Analyzing the typical network attacks and security requirements of EH-WSN,addressing the problem that the single evaluation criterion of the trust mechanism leads to the network not working properly under many types of network attacks,and combining the high reliability,proximity,and energy requirements of relay nodes in routing,the dissertation proposed a multi-factor relay node evaluation model.Using Markov decision processes to derive the optimal value function of the multi-factor relay node evaluation model and to address the problem of slow convergence and difficulty in quickly identifying the best routing path due to the greedy algorithm strategy of the traditional Q-Learning algorithm,the dynamic exploration strategy is adopted,which combines the optimal hyperparameters to make the proportion of exploration actions decrease with the state space.Based on the improved QLearning algorithm,establish the mathematical connection between node behavior,node position,node energy,and node value during network routing.Simulation experiments confirm that the cumulative hop count of network transmission is reduced and the average hop count is decreased by about 21% under the dynamic exploration strategy compared with the traditional Q-Learning algorithm,which effectively improves the ability of the network to identify the optimal path.(2)In response to the issue of trust mechanisms being unable to defend against external network attacks,this dissertation designs a lightweight authenticated encryption mechanism and proposes a secure routing algorithm based on the authenticated encryption mechanism by combining the multi-factor relay node evaluation model.The research idea of this algorithm is that based on the physical unclonable function to extract the node’s physical fingerprint as node identity information,the sink node stores and constructs the identity information database,and when the node sends a connection request to the sink node,it decides whether to allow the node to access the network via comparing the identity information,effectively avoiding the unauthorized node from accessing the network data.The objective is to select a reliable routing path and complete the calculation of the comprehensive evaluation value of relay nodes.Also,the algorithm considers the energy consumption of nodes,sets the node energy threshold,and dynamically adjusts the routing path according to the node energy,which effectively protects the low-energy nodes and solves the "energy hole" caused by the nodes with a high evaluation value or the nodes close to the sink nodes.Simulation results show that the algorithm can effectively resist different types of network attacks and improve the packet delivery rate,the detection rate of selfish nodes,and the accuracy rate.The algorithm also balances the energy consumption of each node and effectively utilizes the renewable energy harvested by the nodes.