Title:  Predicting Vehicular Collisions in Vehicle-to-vehicle Networks using Physical Layer Techniques

Committee:

Dr. John Copeland, ECE, Advisor, Chair , Advisor

Dr. George Riley, ECE

Dr. Gregory Durgin, ECE

Dr. Raheem Beyah, ECE

Dr. Ellen Zegura, CoC

Dr. Yusun Chang, ECE

Abstract:

This dissertation presents novel physical layer (PHY) techniques for providing collision avoidance services to drivers and future autonomous systems participating in Vehicle-to-Vehicle (V2V) networks. To date, predicting car collisions by observing PHY characteristics of DSRC radios in V2V networks is not well investigated nor validated. The periodic broadcast of either SM or non-safety network traffic could be leveraged to identify physically observable events indicative of a collision. The advancements presented in this work will allow future investigators the opportunity to design new collision avoidance methods and to develop additional services for both drivers and autonomous systems. With this body of work, car accidents could be reduced and V2V enhanced to provide collision avoidance not just at the application layer, but directly from the PHY of the communication stack.