Internet QoS, Fault-Tolerance, and Security    Research Publications

Internet infrastructure Security:

The phenomenal growth of the Internet coupled with growing concerns for cyber warfare demand for immediate solutions for securing the Internet infrastructure. Although the security research has witnessed sophisticated algorithms and protocols for information assurance, the issue of securing the Internet infrastructure has not been addressed adequately and is gaining momentum only in recent years. Our research focus has been on developing a comprehensive understanding of the security threats to Internet infrastructure and develop robust counter-measures; specifically focusing on DDoS attack prevention, mitigation, and traceback schemes, and detecting and mitigating botnet attacks, and some work in secure routing and multicasting protocols.

QoS and Overlay Routing and Multicasting:

The proliferation and increasing importance of QoS-aware applications coupled with the advancement in high-speed networking are driving the need for scalable and deployable routing and multicasting architectures, algorithms, and protocols over the Internet. In this context, the focus of our research has been design and analysis of (a) overlay algorithms for routing, multicast, and fault detection (b) measurement algorithms for overlay networks, (c) QoS constrained routing algorithms, (c) reliability constrained routing in QoS networks, (d) edge-based fault detection and recovery algorithms, (e) algorithms for multicast tree construction, (f) algorithms for multicast tree maintenance balancing the tradeoff between tree cost and service disruption, and (g) overlay-routing assisted transport protocols for wireline-wireless networks.

---

Real-Time Systems    Research Publications

Our research focuses on “energy-aware resource management for (wireless) networked real-time embedded systems,” wherein the goal is to minimize energy consumption while satisfying timeliness and resource requirements of real-time tasks. The primary focus is on the design, analysis, and implementation of cross-layer algorithms that exploit synergy across system layers (compiler, OS, and networking) so as to significantly improve the energy savings of embedded systems much beyond what is known in the state-of-the art research and design.

Past Work:

Feedback-based resource management employing feedback control strategies for real-time systems with applications to industrial automation and precision farming.

Resource management in parallel and distributed real-time systems focusing on dynamic scheduling, fault-tolerant scheduling, resource reclaiming, and real-time WAN and LAN networks.

Our past research in this areas is reported in the Ph.D Dissertation: G. Manimaran, "Resource management with dynamic scheduling in parallel and distributed real-time systems", (pdf) Ph.D Dissertation, Indian Institute of Technology (IIT), Madras, India, July 1998.(Synopsis).
 

---

Critical Infrastructure Systems (Electric Power Grid)   Research Publications                PowerCyber Lab

Cyber Security for Electric Power Systems: Our work focuses on a cyber-physical systems approach for cyber security risk assessment, anomaly detection, and mitigation algorithms to counter intrusion based and DoS based attacks on the SCADA control system. This is an integrated approach that captures both cyber vulnerabilities and their resulting impacts on the physical power system in terms of capacity loss (electric load loss) and stability. Collaborator: Prof. Chen-Ching Liu

Embedded Sensor Network for Robust Electric Grid: Our current work focuses on embedded sensor network design and associated decision algorithms for data aggregation, fault diagnosis, state estimation, and control with the goal of  significantly improving the performance and resiliency of energy infrastructure. Collaborator: Prof. Vijay Vittal

Also, research on robust computing and communication infrastructure support for electric power grid - focusing on fault detection, location, containment, restoration, real-time processing - to restore the power grid following catastrophic failures. 

---