The Key Management Scheme For Wireless Sensor Networks Based On Convertible Ternary Polynomial

Wireless sensor networks used in military and commercial areas have a high demand onsafety performance of the network.However the wireless sensor nodes are limited by thecomputing power, storage space and energy, wireless sensor networks can not adopt directlysecurity solution ripe applied to the wireless network. Therefore, the studies of securitysolutions for wireless sensor networks are imperative. This thesis analyzes the threatsand countermeasures in wireless sensor network protocol stack layers, and the researchstatus of the wireless sensor network key management.With the wide application of wireless sensor networks, wireless sensor networks scalewill gradually become larger. Key management scheme designed for large-scale wirelesssensor networks is imminent. Combining exclusion basic system with symmetric bivariatepolynomial key management schemes, the paper proposed an novel wireless sensor networkkey management scheme based on convertible ternary polynomial.Variables take differentvalues, convertible ternary polynomial can generate the management key and the sessionkey. Taking into account the sensitivity of the chaotic function on the initial, convertible ternarypolynomial used chaotic function generator to product its coefficients, so communicationprocess only need to pass the initial variable to reduce the communication cost. The thesisestablished a mathematical model to deal with the collusion problems between nodes, useddiscrete particle swarm optimization algorithm to optimize the model,increase the length of theshortest collusion chain and enhanced the ability to resist collusion attack. Forsome variable assignment, convertible ternary polynomial can be transformed into binarysymmetric polynomial, The nodes used the binary symmetric polynomial to complete thethe negotiation and generate the session key. In the thesis, two-level EBS structure was not onlyconvenient to get rid of a node, increase a node and updat key, while reducing the number ofpolynomial used.Finally, the scheme was compared with two other options at five aspects including the safety performance, storage space, communication overhead, the costs to get rid of anode and update key and calculation cost. The results show that the proposed scheme byadding a small amount of computational cost, improved safety performance, reduced storagespace and communication overhead. The scheme is suitable for large scale wireless sensornetworks.