EAGER: Collaborative: Algorithmic Framework for Anomaly Detection in Interdependent Networks

点击翻译按钮获取中文摘要

Modern critical infrastructure relies on successful interdependent function among many different types of networks. For example, the Internet depends on access to the power grid, which in turn depends on the power-grid communication network and the energy production network. For this reason, network science researchers have begun examining the robustness of critical infrastructure as a network of networks, or a multilayer network. Research in network anomaly detection systems has focused on single network structures (specifically, the Internet as a single network). Among these methods, some promising detection algorithms rely on decentralized and distributed coordination among many participants, improving meaningfully over results from independent parallel and centralized algorithms. The project involves rigorous analysis of the different challenges and opportunities for anomaly detection posed by multilayer networks relative to single network structures, with a particular focus on how cross-layer information can be effectively used to improve both efficiency and detection as well as how cross-layer threats can create vulnerabilities. The project develops a general framework that can be used in multiple applications to detect large-scale threats to information flow for enhanced security. This has the potential for significant benefit to society through its contribution to enhanced resiliency in the nation's cyber infrastructure and other interdependent critical infrastructure such as the power grid. The combination of concepts and ideas from the cybersecurity community with the network science community will help researchers in both fields to better understand the realistic problems and be aware of each other's problems, results, and techniques.