Recent Projects
Automated Wordlength Optimization Method for Design and Implementation of a Fixed-Point Pipeline FFT Processor
In this project, we focus on the widely-used radix-22 Decimation-In-Frequency (DIF) FFT algorithm. A matrix representation of the quantization error propagation model is thoroughly investigated and the quantitative relationships among wordlength configuration, fixed-point FFT architecture and output SQNR (Signal-to-Quantization-Noise Ratio) is disclosed by assessing the analytical form of the FFT output SQNR. Accordingly, a novel automated wordlength optimization method is proposed. Under a group of constraints: input wordlength, output SQNR and FFT length, the method can automatically generate a group of optimal wordlength configurations.
Intelligent Analysis Engine and Event Correlation Model for Cyber Threat Discrimination
Intrusion detection and prevention system (IDPS) that is dedicated to detect malicious cyber threats is regarded as a promising solution for protecting SCADA systems. However, the detection accuracy and event storms are two major problems for IDPS. Our project proposes a distributive implemented event correlation system, which is the key component for an intelligent cyber-threat analysis engine. In this project, we design and implement a set of algorithms and optimized processing flows that can improve the detection accuracy, simplify event correlation, and thereby reduce overall operating costs. Spatial-Temporal logic is an intuitive way of expressing correlation patterns.
Virtual SCADA Lab
In this project, we built a Virtual SCADA Software Infrastructure (VSSI) and analyzing tools with configurable software-defined emulated sensors, actuators and/or Programmable Logic Controllers (PLCs), network emulators for communication simulation, and SCADA system with data analytics, cyber security assessment, and vulnerability investigation. The proposed VSSI will be constructed by a set of software modules, which will simulate the behavior of various sensors, actuators, and control system, and the various SCADA communication protocols. You can see more details in our project webstie.
Cyber Security Vulnerabilities and Possible Cures for Smart Grid Transmissions Subsystem
In this project, we studied the smart grid security vulnerabilities with respect to transmissions subsystem. Construct threat models and methods to circumvent the threats germane to such threat models. We achieved our long-term goal by completing four stages. First, we completed an extensive survey on smart grid security literature and published a survey paper on cyber security issues on smart grid. Our investigation convinced us that the cyber security issues in transmission subsystems are critical with big impact. Second, we constructed a malicious data injection attack model. This model is premised on adversaries accessing a number of measurement devices installed throughout the transmission network so as to tamper the measurement data without been detected. Third, we proposed to use a wide area measurement system based on phasor measurement units (PMUs) to establish a minimal security infrastructure to circumvent the effects of such attacks. Finally, we used a co-simulation platform GECO which combines a power system simulator and a communication network simulator into an integral platform to validate the feasibility and validity of our attack model and counter measure.