Power System Security Under False Data Injection Attacks With Exploitation and Exploration Based on Reinforcement Learning

Resilient Configuration of Distribution System versus False Data Injection Attacks Against State Estimation

State estimation is used in power systems to estimate grid variables based on meter measurements. Unfortunately, power grids are vulnerable to cyber-attacks. Reducing cyber-attacks against state estimation is necessary to ensure power system safe and reliable operation. False data injection (FDI) is a type of cyber-attack that tampers with measurements. This paper proposes network reconfigurati...

Exploration and Exploitation in Reinforcement Learning

False Data Injection Attacks in Control Systems

This paper analyzes the effects of false data injection attacks on Control System. We assume that the system, equipped with a Kalman filter and LQG controller, is used to monitor and control a discrete linear time invariant Gaussian system. We further assume that the system is equipped with a failure detector. An attacker wishes to destabilize the system by compromising a subset of sensors and ...

Learning Exploration/Exploitation Strategies for Single Trajectory Reinforcement Learning

We consider the problem of learning high-performance Exploration/Exploitation (E/E) strategies for finite Markov Decision Processes (MDPs) when the MDP to be controlled is supposed to be drawn from a known probability distribution pM(·). The performance criterion is the sum of discounted rewards collected by the E/E strategy over an infinite length trajectory. We propose an approach for solving...

A Security Method for Multiple Attacks in Sensor Networks: against False-report Injection, False-vote Injection, and Wormhole Attacks

In a large-scale wireless sensor network, damage spreads rapidly in the network when under false report injection, false votes injection, or wormhole attacks. These attacks cause finite energy resources to be drained, legitimate reports to be dropped, and data to be intercepted by adversary nodes. A probabilistic voting-based filtering scheme (PVFS) and localized encryption and authentication p...