Cal Poly Pomona Cyber Collaborative

Cybersecurity Research

Undergraduate and graduate students participate in research projects under expert faculty advisors to provide important information to innovate and strengthen cybersecurity, including detecting system weaknesses, analyzing new attacks and potential solutions, and creating important datasets. For the most current research, visit the webpages of the PolySec Lab and Research Experience for Undergraduates in Big Data Security and Privacy.

Student Research

Josh Silva researched four main areas of cyber attacks, which include two common types: malware – software that damages a computer, server or computer network; and phishing – obtaining private information such as credit card numbers and passwords through spam. He researched how cyber attacks can pass through systems undetected depending on the computer network layer on which information is sent and received. Silva then tagged popular attacks for models to recognize if the information is harmful to the device or compromises private information.

“Analyzing details and applying everything in my classes is really beneficial to me because I get to see what people do in the industry. Cal Poly Pomona gave me opportunities, including this research project, and they have connections with Northrup Grumman and the Jet Propulsion Laboratory, and with companies in networking and software engineering. I got a taste of everything and I realized that maybe networking is for me.” – Silva

Bronco Digital Magazine: Cyber Security

Faculty Research

Tamer Omar leads groups of students in developing a pilotless autonomous vehicles command and control center. Students control the movement of planes and automotive carts from a remote station, using a VRX simulator with wide-view screens to navigate and view the environment. Autonomous vehicles have multiple security and safety applications, allowing greater assessment of damaged areas, including finding individuals trapped in difficult-to-reach areas and sending rescue and mission critical signals.

Omar uses application programmable interfaces as systems to house the programming needed to make the vehicles function and a high-speed, 4G wireless network to communicate the information from the systems to the vehicle and driver. Students participating in his projects learn the importance of data security and how systems can be hacked and autonomous vehicles taken over by outside attackers.

“Part of what we’re doing as related to cybersecurity is that there is a lot of traffic going everywhere around us. We send emails, use social media, you use your phone, your computer – there’s a huge amount of data that we produce every day. Now, it’s not a matter of producing the data, it’s more of the matter of, what is the value of this data and how do we secure it?” – Omar

Frontiers Technologies Cybersecurity Research Lab
Bronco Digital Magazine: Out of the Driver's Seat

Tingting Chen is an advisor on multiple projects in the Research Experience for Undergraduates in Big Data Security and Privacy. Her specialty is machine learning security and privacy, and medical data security.

For machine learning security and privacy projects, Chen’s team aims to: solve data privacy issues when a deep learning model is trained in a cloud by developing a new, fully homomorphic encryption based protocol preserving the privacy of training and test data, guaranteeing high accuracy and reasonable efficiency; and improve the robustness of current deep learning models with the presence of adversarial examples, which appear normal to humans, but reliably fool deep classifiers.

Medical data security has two challenges: the large scale of data, specifically genomic information; and how to share usable and secure data between agencies without exposing the sensitive information attached to it. Chen’s solution is to provide homomorphic encryption to preserve patterns of the original data set, allowing extraction of data with matching attributes. A specific graphics processing unit program will be used in the encryption and decryption of big genomic data on a parallel processing platform to speed up the process.