Funding

Agency: Pacific Northwest National Laboratory
Researchers: Medal, H.R. (PI), Bian, L., Zhang, S.
Amount: $400,000 ($200,000 Medal share)

From the International Working Groups Program, International Institute, Mississippi State University

Researchers: Grala, R.K, Gordon, J., Medal, H.R.

Amount: $2,500

Agency: Pacific Northwest National Laboratory (via the Distributed Analytics and Security Institute)
Researchers: Medal, H.R. (PI), Bian, L., Hu, M., Marufuzzaman, M., Zhang, S.
Amount: $600,000
Abstract: The main goal of this project is to develop a new cyber-node classification approach that utilizes both graph theory and probability modeling. We will pursue our project goal via four objectives: 1) develop an ensemble of models for the probability that a node is malicious or compromised and that an edge exists or not; 2) use advanced model selection techniques to recommend the best probability model; 3) develop novel graph analytic methods that can incorporate uncertainty and scale up to solve very large datasets, and 4) develop new approaches for computing graph uncertainty.

Paper:

1. Botnet Detection Using Graph-Based Feature Clustering
   Chowdhury, S., Khanzadehdaghalian, M., Akula, R., Zhang, F., S. Zhang, Medal, H., M. Marufuzzaman, and L. Bian. To appear in Journal of Big Data.
   Publication Year: 2017
   Abstract
   Detecting botnets in a network is crucial because bots impact numerous areas such as cyber security, finance, health care, law enforcement, and more. Botnets are becoming more sophisticated and dangerous day-by-day, and most of the existing rule based and flow based detection methods may not be capable of detecting bot activities in an efficient and effective manner. Hence, designing a robust and fast botnet detection method is of high significance. In this study, we propose a novel botnet detection methodology based on topological features of nodes within a graph: in degree, out degree, in degree weight, out degree weight, clustering coefficient, node betweenness, and eigenvector centrality. A self-organizing map clustering method is applied to establish clusters of nodes in the network based on these features. Our method is capable of isolating bots in clusters of small sizes while containing the majority of normal nodes in the same big cluster. Thus, bots can be detected by searching a limited number of nodes. A filtering procedure is also developed to further enhance the algorithm efficiency by removing inactive nodes from consideration. The methodology is verified using the CTU-13 datasets, and benchmarked against a classification-based detection method. The results show that our proposed method can efficiently detect the bots despite their varying behaviors.

Agency: Engineer Research and Development Center (U.S. Army)
Researcher: Medal, H.R. (PI)
Amount: $106,310

Abstract: The main goal of this research is to design new algorithms for use on HPC clusters that will enable stakeholders to rapidly and accurately simulate the effect that jamming attacks will have on the performance of a given wireless network design. This goal will be pursued through the following two objectives: 1) design new algorithms for rapidly and accurately computing the throughput of a wireless network under normal conditions (no jamming) and 2) design new algorithms for rapidly and accurately computing the throughput of a wireless network after a jamming attack. These two objectives will contribute to the public good by 1) providing network designers with tools for simulating the throughput of a hypothetical wireless network design and 2) providing designers with tools for simulating the security of a hypothetical design.

Paper:

1. The wireless network jamming problem subject to protocol interference
   Hugh R. Medal. Networks 67(2), 111–125
   Publication Year: 2016
   [PDF]   [Link to Article] Abstract
   We study the following questions related to wireless network security: Which jammer placement configuration during a jamming attack results in the largest degradation of network throughput? and Which network design strategies are most effective in mitigating a jamming attack? Although others have studied similar jammer placement problems, this article is the first to optimize network throughput subject to radio wave interference. We formulate this problem as a bi-level mixed-integer program, and solve it using a cutting plane approach that is able to solve networks with up to 81 transmitters, which is a typical size for studies in wireless network optimization. Experiments with the algorithm also yielded the following insights into wireless network jamming: (1) increasing the number of channels is the best strategy for designing a network that is robust against jamming attacks, and (2) increasing the range of the jammer is the best strategy for the attacker.

Benefits and costs of implementing fuel treatments on nonindustrial private forest (NIPF) lands in Mississippi
Agency: Joint Fire Science Program (Department of the Interior, Bureau of Land Management)
Researchers: Grala, R.K, Varner, J.M., Medal, H.R. (Co-PI), Munn, I.A., Grado, S.C., Cooke III, W.H.
Amount: $218,000 ($46,513 Medal share)

From the Bagley College of Engineering, Mississippi State University

Researchers: Medal, H.R., Li, X.

Amount: $2,000

Agency: National Center for Intermodal Transportation and Economic Competitiveness (funded by US Department of Transportation)Researchers: Medal, H.R. (PI) and Li, X.
Amount: $99,954 ($55,036 Medal share) with $102,988 cost share from Mississippi State University
Abstract: In the project, we investigate the effect of traffic calming measures, such as crosswalks and sidewalks on the overall cost and safety of a multimodal transportation network system design. Our design problem includes auto, transit, and walking as modes of transportation. We propose a new method for multimodal user equilibrium (UE) traffic assignment with network reconstruction, which allows for mode switching. We propose a bi- level mathematical programming model that integrates multimodal user equilibrium traffic assignment in the lower level and the network design in the upper level. The model tries to optimally implement and locate sidewalks and crosswalks considering limited financial resources to provide city planners with a comprehensive tool for planning. Due to the complexity of the problem, it requires a large amount of computational resources and therefore cannot be solved efficiently for large scale problems using state of the art solvers; hence we develop a greedy heuristic and a simulated annealing algorithm to solve large problems. The algorithms use a nonlinear complimentary algorithm to solve the UE traffic assignment. The computational results show that implementing sidewalks and crosswalks both reduces the overall transportation cost and improves pedestrians’ safety.

Paper: MISSING PAPER