A QoS Ensuring Two-Layered Multi-Attribute Auction Mechanism to Mitigate DDoS Attack

Abstract

Incentives are very important to be employed in any defensive mechanism against DDoS attack. Incentive is a major concept abandoned by most of the defensive mechanisms that have been proposed so far. It is a tool that can motivate users to send data wisely into the network. Therefore, in this paper, we have proposed a two layered multi-attribute auction mechanism for incentivising users by imposing payment schemes as well as by providing rewards. Apart from this, we have developed a reputation assessment procedure to identify malicious user by monitoring his credibility score calculated through his marginal utility. Identified malicious users are then mapped to different levels of suspiciousness. Identified legitimate users are forwarded towards first level of auction in which virtual users have been added by service provider to increase the competition among users. Critical values are computed for every user and the users satisfying the criteria are moved towards the second level. In second level, greedy method is utilized for resource allocation. Extensive simulations have been conducted on MatLab to check the validity of the proposed model. Rate of social welfare degradation and user’s satisfaction are utilized to check the appropriateness and validity of the model. Results from experimentation have shown that proposed model is able to generate enough revenue for the service provider and is able to provide acceptable QoS to identified legitimate users when there is an increase in number of malicious users.