Dynamically Differentiated Multipath Security in Fixed Bandwidth Networks

Abstract

Networks can be secured using a secure quality of service approach in which a sender disperses data along multiple secure paths. In this secure multipath approach, a portion of the data from the sender is transmitted over each path on fixed bandwidth networks, and the receiver assembles the data fragments that arrive. The research presented explores the effects of cloud infrastructure attack scenarios, and gauges the threat levels along each path. Optimal sampling and compression via compressed sensing is employed in order to minimize the possibility of dropped packets and missed attacks. The probability of the presence of specific attack signatures along each network path is determined using statistical learning techniques. Path information assurance levels are derived using these probabilities and encryption strengths are dynamically increased along those paths found to be most vulnerable. As compared to statically employed path encryption schemes, the devised methods in this research significantly increase data throughput on fixed bandwidth networks.

Managing risk in high assurance systems by optimizing topological resources

“…Increasingly, more and more tools and algorithms exist to summarize and characterize the content of both the people associated with the nodes of the graph and well as the communication flowing on the edges of the graph. Two examples in the mobile networking space are given in Truong et al and Wei et al In Obert et al, techniques to determine the probability of attack between node pairs using machine learning are discussed. However, there is a critical need to add ways to actuate high assurance systems with authorities and concepts to allow for the “soft” information from these other sources to flow into the “hard” operations of the network.…”