Faculty Advisor

Eisenbarth, Thomas


In the era of small devices connected to each other and to the cloud, security in both hardware and software is of high importance. This work takes on the market demands of the "Internet of Things" devices and proposes the smallest hardware crypto core that is immune to side-channel attacks, improving the previous by almost 15%. On the software side, this work takes a first step into proving that by simply measuring certain hardware performance counters, or by monitoring certain hardware resources, it is possible to accurately identify codes during their execution. Because of their high criticality, this work uses crypto codes and tries to identify them using machine learning techniques. In a simulated environment, an accuracy of over 90% is achieved in classifying crypto algorithms.


Worcester Polytechnic Institute

Date Accepted

March 2016


Computer Science


Electrical and Computer Engineering

Project Type

Major Qualifying Project



Advisor Department

Electrical and Computer Engineering