Satinder Singh
Satinder Singh has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Kim Khoa Nguyen
Cole de Technologie Suprieure, University of Quebec, Montreal, QC, Canada
2 shared publications - Michael P. Wellman
Computer Science and Engineering, University of Michigan, Ann Arbor, MI, USA
2 shared publications - Arun Anand
Department of Veterinary Surgery and Radiology, College of Veterinary Science, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana, India
1 shared publication - Do Van Dang
VNU University of Science, 334 Nguyen Trai, Thanh Xuan, Hanoi, Vietnam
1 shared publication - Edmund H. Durfee
University of Michigan
1 shared publication Krishan Gauba
Oral Health Sciences Centre, Post Graduate Institute of Medical Education & Research (PGIMER), Chandigarh, India
1 shared publication
Latest Publications
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 787-792
Efficient Link-State Multicast Routing by Optimizing Link-Weight with MARL
Networks deployed in contested environments typically have limited resources and face many security threads including information leakage. Multicast traffic engineering with carefully considering eavesdropping attacks can improve network performance while avoiding the information leakage problem. In this paper, we address the problem of achieving multicast traffic engineering based on link-state routing protocols according to the offered traffic while taking the problem of alleviating eavesdropping attacks into consideration. We first provide an Integer Linear Programming (ILP) formulation that finds the optimal link weights for Shortest Path Tree (SPT)-based multicast routing protocols to minimize total network cost. The problem is NP-Hard, and obtaining a solution to the problem in real-time to adapt to highly dynamic traffic demands is very challenging. To meet the real-time requirement, we design a Multi-Agent Reinforcement Learning (MARL) solution to the problem of optimizing link weights to achieve efficient multicast routing in a distributed fashion. In our design, the agents collaborate and communicate with the others in the local region and learn from their experiences to determine the best action to minimize the overall network cost. Our proposed solution is evaluated on a simulation of various traffic profiles and compared with the conventional manually configured link weights and a Genetic Algorithm (GA)-based heuristic solution. Experimental results show the advantages of our solutions in reducing network cost and suggest the potential of using MARL in achieving efficient multicast traffic engineering.
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 787-792
Efficient Link-State Multicast Routing by Optimizing Link-Weight with MARL
Networks deployed in contested environments typically have limited resources and face many security threads including information leakage. Multicast traffic engineering with carefully considering eavesdropping attacks can improve network performance while avoiding the information leakage problem. In this paper, we address the problem of achieving multicast traffic engineering based on link-state routing protocols according to the offered traffic while taking the problem of alleviating eavesdropping attacks into consideration. We first provide an Integer Linear Programming (ILP) formulation that finds the optimal link weights for Shortest Path Tree (SPT)-based multicast routing protocols to minimize total network cost. The problem is NP-Hard, and obtaining a solution to the problem in real-time to adapt to highly dynamic traffic demands is very challenging. To meet the real-time requirement, we design a Multi-Agent Reinforcement Learning (MARL) solution to the problem of optimizing link weights to achieve efficient multicast routing in a distributed fashion. In our design, the agents collaborate and communicate with the others in the local region and learn from their experiences to determine the best action to minimize the overall network cost. Our proposed solution is evaluated on a simulation of various traffic profiles and compared with the conventional manually configured link weights and a Genetic Algorithm (GA)-based heuristic solution. Experimental results show the advantages of our solutions in reducing network cost and suggest the potential of using MARL in achieving efficient multicast traffic engineering.
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 450-455
Joint Jamming Alleviation for Mixed RF/FSO Relay Networks: Optimization and Learning Approaches
In the mixed radio frequency/free-space optical (RF/FSO) relay networks, the effective coordination between the two subsystems is crucial for maintaining system throughput. However, the challenge intensifies when both subsystems are simultaneously intercepted by enemy forces. In this paper, we propose a strategy for alleviating jamming to safeguard the RF/FSO relay system against active jammers. Specifically, we focus on the joint optimization of RF user transmit power and FSO Field-of-View (FoV) angle to maximize system throughput while adhering to jamming alleviation constraints. The strategy is formulated as an optimization problem, which is subsequently solved using an iterative method. Due to the intractable form, the solution for FoV is obtained through a closed-form expression derivation, while the transmit power solution is achieved by using an advanced technique known as the first-order Taylor approximation with the difference of convex functions (D.C). To address the computational complexity associated with the optimization method, we design a Multi-Agent Deep Reinforcement Learning (MADRL) based jamming alleviation algorithm to solve the problem. The numerical results show that the performance of the MADRL-based algorithm is comparable to that of the optimization method, while also being suitable for real-time deployment.
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 450-455
Joint Jamming Alleviation for Mixed RF/FSO Relay Networks: Optimization and Learning Approaches
In the mixed radio frequency/free-space optical (RF/FSO) relay networks, the effective coordination between the two subsystems is crucial for maintaining system throughput. However, the challenge intensifies when both subsystems are simultaneously intercepted by enemy forces. In this paper, we propose a strategy for alleviating jamming to safeguard the RF/FSO relay system against active jammers. Specifically, we focus on the joint optimization of RF user transmit power and FSO Field-of-View (FoV) angle to maximize system throughput while adhering to jamming alleviation constraints. The strategy is formulated as an optimization problem, which is subsequently solved using an iterative method. Due to the intractable form, the solution for FoV is obtained through a closed-form expression derivation, while the transmit power solution is achieved by using an advanced technique known as the first-order Taylor approximation with the difference of convex functions (D.C). To address the computational complexity associated with the optimization method, we design a Multi-Agent Deep Reinforcement Learning (MADRL) based jamming alleviation algorithm to solve the problem. The numerical results show that the performance of the MADRL-based algorithm is comparable to that of the optimization method, while also being suitable for real-time deployment.
- 17 May 2023
- Published by Springer Nature
- p. 237-250
On Surface Features of Canine Strategic Teeth by Metal Additive Manufacturing and Investment Casting—A Comparative Study
- 17 May 2023
- Published by Springer Nature
- p. 237-250
On Surface Features of Canine Strategic Teeth by Metal Additive Manufacturing and Investment Casting—A Comparative Study
- 1 April 2020
- Published by Elsevier in Journal of Endodontics
- Vol. 46 (4) , 471-474
Is Bilateral Mental Incisive Nerve Block Better than Unilateral Mental Incisive Nerve Block during the Endodontic Management of Mandibular Incisors with Symptomatic Irreversible Pulpitis? A Prospective Single-blind Randomized Clinical Trial
- 1 April 2020
- Published by Elsevier in Journal of Endodontics
- Vol. 46 (4) , 471-474
Is Bilateral Mental Incisive Nerve Block Better than Unilateral Mental Incisive Nerve Block during the Endodontic Management of Mandibular Incisors with Symptomatic Irreversible Pulpitis? A Prospective Single-blind Randomized Clinical Trial
- 18 January 2020
- Published by Springer Nature in Autonomous Agents and Multi-Agent Systems
- Vol. 34 (1) , 19
Semantics and algorithms for trustworthy commitment achievement under model uncertainty
- 18 January 2020
- Published by Springer Nature in Autonomous Agents and Multi-Agent Systems
- Vol. 34 (1) , 19
Semantics and algorithms for trustworthy commitment achievement under model uncertainty
- Open Access
- 13 December 2018
- Published by Wiley in Security and Communication Networks
- Vol. 2018, 1-28
Multistage Attack Graph Security Games: Heuristic Strategies, with Empirical Game-Theoretic Analysis
We study the problem of allocating limited security countermeasures to protect network data from cyber-attacks, for scenarios modeled by Bayesian attack graphs. We consider multistage interactions between a network administrator and cybercriminals, formulated as a security game. This formulation is capable of representing security environments with significant dynamics and uncertainty and very large strategy spaces. We propose parameterized heuristic strategies for the attacker and defender and provide detailed analysis of their time complexity. Our heuristics exploit the topological structure of attack graphs and employ sampling methods to overcome the computational complexity in predicting opponent actions. Due to the complexity of the game, we employ a simulation-based approach and perform empirical game analysis over an enumerated set of heuristic strategies. Finally, we conduct experiments in various game settings to evaluate the performance of our heuristics in defending networks, in a manner that is robust to uncertainty about the security environment.
- 13 December 2018
- Published by Wiley in Security and Communication Networks
- Vol. 2018, 1-28
Open AccessMultistage Attack Graph Security Games: Heuristic Strategies, with Empirical Game-Theoretic Analysis
We study the problem of allocating limited security countermeasures to protect network data from cyber-attacks, for scenarios modeled by Bayesian attack graphs. We consider multistage interactions between a network administrator and cybercriminals, formulated as a security game. This formulation is capable of representing security environments with significant dynamics and uncertainty and very large strategy spaces. We propose parameterized heuristic strategies for the attacker and defender and provide detailed analysis of their time complexity. Our heuristics exploit the topological structure of attack graphs and employ sampling methods to overcome the computational complexity in predicting opponent actions. Due to the complexity of the game, we employ a simulation-based approach and perform empirical game analysis over an enumerated set of heuristic strategies. Finally, we conduct experiments in various game settings to evaluate the performance of our heuristics in defending networks, in a manner that is robust to uncertainty about the security environment.
- Open Access
- 14 May 2007
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Constraint satisfaction algorithms for graphical games
We formulate the problem of computing equilibria in multi-player games represented by arbitrary undirected graphs as a constraint satisfaction problem and present two algorithms. The first is PureProp: an algorithm for computing approximate Nash equilibria in complete information one-shot games and approximate Bayes-Nash equilibria in one-shot games of incomplete information. PureProp unifies existing message-passing based algorithms for solving these classes of games. We also address repeated graphical games, and present a second algorithm, PureProp-R, for computing approximate Nash equilibria in these games.
- 14 May 2007
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessConstraint satisfaction algorithms for graphical games
We formulate the problem of computing equilibria in multi-player games represented by arbitrary undirected graphs as a constraint satisfaction problem and present two algorithms. The first is PureProp: an algorithm for computing approximate Nash equilibria in complete information one-shot games and approximate Bayes-Nash equilibria in one-shot games of incomplete information. PureProp unifies existing message-passing based algorithms for solving these classes of games. We also address repeated graphical games, and present a second algorithm, PureProp-R, for computing approximate Nash equilibria in these games.