Research On Intrusion Tolerant Technique In Wireless Sensor Networks

The Internet of Things has come into our lives. Related technology has begun to applications in agriculture, transportation, military, medical, business and many other fields. With practical applications such as smart sprinkler systems, intelligent traffic monitoring systems, residential security systems, monitoring and other rare animals footprint. Large number of nodes of Things perception layer constituting the sensor network, and usually communication by wireless between nodes. Are wireless sensor networks secure, can network guarantee the confidentiality, integrity, reliability, safety goals becomes a problem need to consider.WSN has limited computing power and energy, and is usually located in open spaces, vulnerable to malicious attacks, so there are many researchs of wireless sensor network intrusion tolerance technology. However, the existing intrusion tolerant network routing protocols only consider whether to tolerate the invasion, and did not consider how to ensure that the network has good QoS performance issue. The network node authentication needs to use public-key cryptography. Multivariate public key cryptographic calculations more efficient,but there is vulnerable to attack the problem of linear equations. To solve these problems, specific research work as follows:For most of the multivariate public key cryptosystem is vulnerable to attacks of linear equations, this thesis proposed MPKCRM public key encryption scheme to solve the problem. Most multivariate public key cryptography are based on extension field methods to generate the public key polynomial, this approach implies the k-linear nature, which is the reason of be vulnerable to the attack of linear equations. MPKCRM program introduces the RSA method in the program to generate public-key polynomial, MPK CRM has the advantage of not easily be compromised and high computational efficiency of private-key, and the disadvantage of complex public-key polynomials, and the program gives a suitable application scenarios.For existing intrusion tolerant protocol considered only network Intrusion Tolerance performance, and did not consider the network QoS performance issue, this thesis proposed performance feedback mechanism. Using this mechanism flag exceptions node, select a cluster head node as reliable. This thesis designed PFITP which is based on this mechanism. PFITP both in the network have tolerance for the main types of attacks, but also continue to optimize network performance. It can bring good robustness and practicality to the sensor network.Designed and implemented IoT perception layer test bed system (IoT Testbed) of the model attacks and IEEE802.15.4 protocol testing framework. Use Leach to do a test example, the platform showed improved functionality and better stability.