BMW Lab @ Wright State

Projects @ BMW LAB

Ongoing Projects

SNMiner
Modeling faults and malicious activities in sensor networks can be challenging. Designing and re-evaluating a 'good' classifier to detect abnormalities imposes yet another challenge once the sensor network is deployed in the field. Common approaches among researchers involve obtaining publicly accessible labeled datasets, generating synthetic sensor data, or collecting sensor readings from a real-world deployment. Then, an anomaly detection/classification algorithm is applied over the dataset and its accuracy is further validated (after labeling data instances of unknown classes). These approaches, however, do not directly take into account the emergence of new types of abnormalities under the dynamics of a deployed sensor network. In order to incorporate new fault models, it is essential to reconfigure the network without interrupting its operation. To this end, we introduce SNMiner, a tool that facilitates fault modeling, and constantly re-evaluates the accuracy of existing classifiers or detection algorithms over real-time sensor streams.








NSF CCLI: Evolvable Wireless Laboratory Design and Implementation for Enhancing Undergraduate Wireless Engineering Education

This project develops an easily replicable model of evolvable, low cost, software defined radio (SDR)-based wireless communication and networking laboratories and associated teaching and learning materials that can be adopted or adapted to impact national engineering education practices. Furthermore, the PIs will demonstrate the effectiveness of enhancing student learning and skills in three existing undergraduate courses, wireless communication, mobile computing, and wireless sensor networks, which form a part of an integral wireless engineering curriculum. This exploratory project will last for two years, and will be a collaborative effort between faculty from Computer Science & Engineering and Electrical Engineering departments. Evaluation and assessment of project outcomes will be conducted by the Center for Teaching and Learning at Wright State University in collaboration with the PIs.

The proposed SDR-based labs are tailored to the need of individual courses, yet serve as a catalyst for the integration of core courses. The outcomes of the project include (1) a set of pilot wireless course laboratories based on the Universal Software Radio Peripheral (USRP) boards that employ GNU software radio, lab development user manuals, lab teaching manuals, proven methods of effective lab instruction, and evaluation & assessment materials, (2) a pilot test for evaluating the developed labs that demonstrates the effectiveness of motivating, engaging, and enhancing student learning and skills as prescribed by the ABET engineering criteria; and (3) transformative educational approaches in wireless communication through the innovative blend of prescribed and open-ended lab structures to advance the current approach, in which most undergraduate wireless communication courses are being taught without a laboratory. The proposed labs will create a space where students can learn by working with tangible signals, wireless channels, and communication systems, which reinforces mathematics and simulation examples, and helps integrate concepts by building a working system.

Future engineers will need to be trained with fundamental principles as well as emerging technologies. The evolution of wireless communication and networking presents such a need and a unique opportunity to integrate undergraduate education across the Electrical Engineering and Computer Science curricula. After the funding period ends, the project will be self-sustainable because the developed labs and courses will be smoothly integrated as a part of a wireless core curriculum, based on which an undergraduate program in wireless engineering across Electrical Engineering and Computer Science departments will be built at Wright State University. The project will benefit a diverse population of students by motivating, engaging, enhancing their learning and skills as prescribed by the ABET. The technology on which the proposed lab development is based is cutting edge, demonstrating a viable example of adopting new technology and research to enhance undergraduate STEM education. The platform employed for development, USRP boards, is of low cost; and the software used, GNU software radio, is free and has a large supporting community that provides unlimited innovation. Therefore, the lab environment can be easily portable to even the most cash-starved institution. No additonal expensive testing equipment is needed, which provides a very low entry point for adaptation and/or adoption nationwide. The project will benefit the society by producing more technology innovators to meet the demand of a new wave of wireless IT. The success of the exploratory project will lead to a full scale implementation for a national model of SDR lab-based wireless communication and networking courses.

NSF CRI: IAD Instrumentation of a Measurement and Test System for Open Spectrum Wireless Communication and Networking

Past Projects

Network Security Visualization

Network simulators can produce huge traces of simulated traffic. These traces, usually presented as text, cannot be easily used to detect any malicious activity. Security visualization techniques have been developed over the decades as a result of various research efforts in the industry, academia, and even research carried out by individual hackers. These techniques can be powerful when employed in the field of network security where the visual recognition capability of human eyes can be exploited to allow an early detection of malicious acts. In this research project, we develop a new security visualization tool called "SecVizer". QualNet network simulator is used to simulate various security scenarios and produce traffic traces that would serve as inputs to our developed visualization tool. Employing a combination of techniques, our tool can render a QualNet network topology in 3-D, play animation of the traffic traces, and generate statistics. In particular, it combines both topology visualization in a 3-dimensional perspective and the parallel coordinate plot technique to obtain a faster and more effective identification of network vulnerabilities. By observing image patterns of the parallel coordinate plot, one can identify the different security attacks while at the same time exploring the network traffic volume and the topology being deployed. The tool has shown success in detecting simulated DDoS, port scan and host scan attacks. It was also able to detect real traffic attacks; real-time traffic is converted to QualNet-like traces which is compatible with SecVizer input format.