The paper is titled "RF-PUF: IoT Security Enhancement through Authentication of Wireless Nodes using In-situ Machine Learning.” Humans can recognize the identity of a known person based on their voice signatures. During Radio-Communication, unique process-variation signatures are inadvertently imparted on the communicated EM signals, but are typically rejected at the receiver. These signatures could be utilized to detect the transmitter identity, using Machine-Learning at the receiver, without any extra hardware at the resource-constrained IoT Transmitter. This work demonstrated how a human-like ‘voice signature recognition’ capability could be brought to asymmetric IoT Radios to improve IoT last-mile Communication Security for resource-constrained IoT Nodes using Machine Learning (ML) in the Gateway receiver.  

 

This is the second year in a row Prof. Sen’s group has received an awards at HOST. In 2017, the group won a Best Student Paper Award.