The signature characteristic of a WMD attack is massive but localized faults. However, the usual metric for evaluating the fault-tolerance capability of a communication network, the connectivity of the network graph, has no way of capturing the notion of locality of faults. In order to capture the notion of locality of faults, the PI of this proposal (Sen) introduced the notion of region-based faults and region-based connectivity in . Since the inception of this project in 2009, we have studied the impact of region-based faults on critical infrastructures. Since a WMD attack will be conned to a region, it can be modeled as a region-based fault. It may be noted that recognizing the appropriateness of region-based faults as the correct modeling tool for WMD in icted faults, several DTRA funded researchers are currently using this model [10, 5, 2]. Our current DTRA grant is due to expire in April 2012. If this request for increase of the Period of Performance of the grant by adding option years is approved, we plan to expand our focus from a single-layer network to multi-layer interdependent networks(MLIN). Study of MLIN is of signicant importance as critical infrastructures, such as electric power distribution network, communication network, transportation network andnancial network, are highly interdependent and failure of one network has inevitable consequences in other networks. A MLIN involving power and communication is shown in Fig. 1(a). One of the objectives of this project is to understand and assess the damage caused by a WMD attack on a MLIN. By damage assessment, we not only imply immediate damage but also the damage as it progresses over time, involvingrst-order, second-order and third-order eects . The immediate impact of a WMD attack may lead to disruption of one of more layers of the MLIN. Fig. 1(b) shows the disruption of long distanceber optic communication network after an attack in central Europe. As a part of research conducted by this team, we have already developed ecient data storage schemes that ensure that no matter where the WMD attack takes place (the red circle in thegure), the disconnected components of the surviving network will have enough data segments (parts of thele), so that the entire le can be reconstructed in the largest connected component. In addition to studying the impact of WMD attack related progressive failure, the objectives of this research include study of the issues related to MLIN hardening - i.e., strengthening of multi-layer network structures (subject to budget constraints) so that impact of progressive failure is minimized progressive recovery from failure due to a WMD attack data resilience to a WMD attack Each objective will comprise of a set of tasks, and we plan to undertake these tasks through a multidisciplinary approach, employing tools and techniques from various mathematical disciplines.
|Effective start/end date||4/8/09 → 12/31/14|
- DOD: Defense Threat Reduction Agency (DTRA): $1,428,647.00