Abstract
When it comes to selecting an optimal defense strategy for an intrusion detection system of a wireless sensor network, such challenges as those brought about by the diversification of the attack methods and the expanded database of the attack patterns have to be dealt with. To overcome those challenges, this paper combines realistic bounded rationality with the incomplete information of the attack-defense players by employing evolutionary game theory as a tool. Firstly, an attack-defense evolutionary game model considering three types of population, in which attackers are subdivided by the source of the threat into external attackers and selfish nodes, is proposed. The sets of player types and the game strategies in our model can be extended from 2×2 to n× m × l. The sensitivity of the evolutionary population to similar strategies, which reflects the efficiency change in the multi-agent learning process, is depicted by a replicator dynamic equation especially optimized for this purpose by the introduction of an enhanced cooperation mechanism. In essence, an optimal defense strategy selection algorithm is provided by calculating the evolutionary stable equilibrium and a description of the evolutionary trajectory of the players over time is obtained. Moderate security and proactive defense in the form of support decisions have been provided by our method for wireless sensor networks. Experimental results have verified the validity of our method. Moreover, the optimized algorithm has solved the problem that an excessively large database of attack patterns affects the speed of switching to the optimal defense decision strategy and the learning efficiency of the evolutionary game replicator dynamic mechanism is not fast enough.
Highlights
T HE characteristics of Wireless Sensor Networks (WSNs), such as low power consumption, selforganization, cheap, convenient, and high node redundancy, have helped them enjoy widespread usage in areas such as military reconnaissance, environmental monitoring, smart home, medical health etc
As network attacks diversified and complicated increasingly, the security challenge of WSNs caused the full attention of researchers and has obtained specific research results, such as encryption technology, firewall technology, intrusion detection system (IDS), etc
Considering the actual network scenario, it is proposed that different types of attackers have different influences on IDS decisions, in which attackers are segmented into external attackers and internal selfish nodes, both of which affect WSN’s decisions
Summary
T HE characteristics of Wireless Sensor Networks (WSNs), such as low power consumption, selforganization, cheap, convenient, and high node redundancy, have helped them enjoy widespread usage in areas such as military reconnaissance, environmental monitoring, smart home, medical health etc. The most critical role of WSNs is the continuous collection of all kinds of data required by the observers in the monitoring area. Ensuring the security of the sensor network is the premise to ensure the authenticity of the acquired data [2]. As network attacks diversified and complicated increasingly, the security challenge of WSNs caused the full attention of researchers and has obtained specific research results, such as encryption technology, firewall technology, intrusion detection system (IDS), etc. Many existing security schemes do not apply to WSNs due to the limited node energy and weak computing capacity [3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
