NetBrane - Detection and Mitigation of DDoS

#DDoS,FPCA+KMeans
Share

The DHS funded NetBrane project has developed highly efficient techniques to detect anomalies within network data as we attempt to mitigate Distributed Denial of Service attacks.  Using functional principal component analysis and K-means clustering (FPCA+KMeans), an unsupervised machine learning data-driven approach, we analyze observed activity to reveal several categories of outliers ranging from benign friendly actors to malicious attackers.  With eigenfunction scores, clustering, and individual behavior summary statistics, we assign risk probabilities to these.  We can mitigate a 10gbps attack within 2 minutes and are shooting for 100gbps within 2 minutes by the date of this presentation.

Stephen C. Hayne is a Full Professor of Computer Information Systems in the College of Business at Colorado State University. He received his Ph.D. from the University of Arizona (1990). He has received more than $8M in grants from NSF, ONR, and DHS as well as an IBM Faculty Fellowship. He has published 70+ papers in journals such as Journal of Management Information Systems, Decision Sciences, Database, Journal of Information and Management, Journal of Computer Supported Collaborative Work, IBM Systems Journal, Electronic Markets, Journal of Organizational Computing & Electronic Commerce, and International Journal of Human Computer Studies, and in major conferences. His research is based in the desire to use innovative technologies to solve real business problems. He has developed theories and implemented systems to help groups with communication and decision-making, using concepts such as shared cognition, collaborative drawing, group brainstorming, concurrent issue surfacing/consolidation, consensus building, choice, pattern recognition and team bidding in auctions. Dr. Hayne’s other research interests lie with “social” networks theory/analytics, distributed systems, big data, auctions and ecommerce (reputation and pricing effects). His most recent research award from DHS involves detecting and mitigating large-scale distributed denial of service attacks (DDOS).

Dr. Hayne’s teaching in the subjects of network/systems security, database, strategy, electronic commerce, and software engineering, have led to nominations as a Best Teacher at CSU by the Alumni Association (2011, 2013, 2014 and 2017). He has been General Chair and Treasurer for many of the GROUP and CSCW conferences and was co-Chair of the 2007 America’s Conference on Information Systems.



  Date and Time

  Location

  Hosts

  Registration



  • Date: 17 Oct 2019
  • Time: 06:00 PM to 09:00 PM
  • All times are (UTC-06:00) Mountain Time (US & Canada)
  • Add_To_Calendar_icon Add Event to Calendar
  • 201 East 4th Street
  • Loveland, Colorado
  • United States
  • Building: Desk Chair
  • Room Number: Cleveland

  • Contact Event Host
  • Starts 14 September 2019 11:18 AM
  • Ends 17 October 2019 03:00 PM
  • All times are (UTC-06:00) Mountain Time (US & Canada)
  • No Admission Charge