Abstract
The problem of cooperative intrusion detection in battery-powered wireless mesh and sensor networks is challenging, primarily because of the limited resources available to participating nodes. Although the problem has received some attention from the research community, little is known about the tradeoffs among different objectives, such as high network performance, low power consumption, low delay in information collection and high security effectiveness. This article proposes, to the best of our knowledge for the first time, cooperative intrusion detection functions that take into account multiple objectives simultaneously. We formulate the problem of identifying the type of intrusion detection function each node runs, as a multi-objective optimization problem, and propose solutions based on genetic algorithms. Through extensive simulations we demonstrate that our solutions are scalable to large networks, and are characterized by a small variance in the normalized fitness value of individual/single objectives and by a small attack detection/reporting delay. In a real implementation/evaluation we demonstrate that our cooperative intrusion detection system achieves a higher detection rate (93%) than state of art solutions.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have