Is this project an undergraduate, graduate, or faculty project?
Undergraduate
individual
What campus are you from?
Daytona Beach
Authors' Class Standing
Jaden Caradine, Senior
Lead Presenter's Name
Jaden Caradine
Faculty Mentor Name
Bryan Watson
Abstract
"This project aims to develop an Artificial Immune System inspired intruder detection system based on current state-of-the-art strategies. Traditional intruder detection systems are not equipped to handle the evolving landscape of cyber security. Signature based detection systems require a database of known signatures and traditional anomaly detection systems are plagued with false positives. By contrast, Artificial Immune Systems can respond to threats never encountered. They are adaptive, proactive, and resilient to interference. This makes them ideal for determining when a system is behaving abnormally. Researchers are developing a novel intruder detection system, but they need a baseline to compare it to state-of-the-art technology. This Intruder Detection System will utilize Clonal Selection Theory, Negative Selection Theory and signature-based discrimination to establish a unique, decentralized, noise tolerant system capable of optimizing over time. The current state of this system can differentiate normal data traffic from anomalous activity by creating a detector field in two-dimensional space but has yet to be applied to any existing systems. This detector field uses a negative selection algorithm and training data sets to define where normal data exists and everything else is deemed an intruder. Finally, the system measures false positive, false negatives and overall detection rates. Future iterations of this project will evaluate n-dimensional space and utilize clonal selection theory to optimize the position of the detectors over time. This project will result in a local, scalable, state-of-the-art intruder detection system. It will serve as a valuable baseline for evaluating novel intruder detection strategies. "
Did this research project receive funding support from the Office of Undergraduate Research.
No
Included in
Development of an Artificial Immune System Based Intruder Detection Algorithm for Comparison with a Novel Intruder Detection Strategy.
"This project aims to develop an Artificial Immune System inspired intruder detection system based on current state-of-the-art strategies. Traditional intruder detection systems are not equipped to handle the evolving landscape of cyber security. Signature based detection systems require a database of known signatures and traditional anomaly detection systems are plagued with false positives. By contrast, Artificial Immune Systems can respond to threats never encountered. They are adaptive, proactive, and resilient to interference. This makes them ideal for determining when a system is behaving abnormally. Researchers are developing a novel intruder detection system, but they need a baseline to compare it to state-of-the-art technology. This Intruder Detection System will utilize Clonal Selection Theory, Negative Selection Theory and signature-based discrimination to establish a unique, decentralized, noise tolerant system capable of optimizing over time. The current state of this system can differentiate normal data traffic from anomalous activity by creating a detector field in two-dimensional space but has yet to be applied to any existing systems. This detector field uses a negative selection algorithm and training data sets to define where normal data exists and everything else is deemed an intruder. Finally, the system measures false positive, false negatives and overall detection rates. Future iterations of this project will evaluate n-dimensional space and utilize clonal selection theory to optimize the position of the detectors over time. This project will result in a local, scalable, state-of-the-art intruder detection system. It will serve as a valuable baseline for evaluating novel intruder detection strategies. "