•   When: Friday, April 28, 2017 from 12:00 PM to 02:00 PM
  •   Speakers: Ajay Nagarajan
  •   Location: ENGR 3507
  •   Export to iCal

Abstract

The current approach to security is based on perimeter defense and relies on firewalls, intrusion detection systems (IDS), and intrusion prevention systems (IPS). These systems require a priori information about attack patterns and system vulnerabilities. With ever-increasing bandwidth and thousands of unique malware signatures coming out every day, it is becoming impractical to prevent every intrusion. And so, intrusion tolerance assumes that intrusions are inevitable and focuses efforts on minimizing the impact of intrusions. The variety and complexity of cyber-attacks is increasing.  Various industry data breach investigation reports suggest that customized malware are difficult to detect and data ex-filtration often occurs over a period of days, weeks and months. The attackers’ strong motivation leads to organized and targeted cyber-attacks.  The current intrusion detection and prevention approaches are reactive in nature and inadequate to prevent all attacks. 

Given the clear need to design intrusion tolerant architectures, my work focuses on extension and application of recovery driven intrusion tolerance systems that make the attacker work harder by reducing the server’s exposure time to the internet. This approach relies on using hybrid architectures that combine reactive and proactive systems to protecting the cyber infrastructure. My research framework entails a) Proposing hybrid architectures founded on SCIT, a recovery driven intrusion tolerance approach; b) determining the influencing factors of each hybrid strategy and studying the impact of their variations within the context of an integrated intrusion defense strategy; c) defining economic models to assess the efficacy of proposed hybrid architectures; d) using mathematical models to evaluate proposed hybrid architectures and assess optimal operational parameters; and e) validating research using test bed experiments and simulations outlining impact of proposed architectures on system attack surface and intruder work factor.

To system architects and executive management alike, this work can constitute as the basis for making informed decisions while piling layers of security as part of defense-in-depth strategy.

Posted 6 years, 11 months ago