Florida International University

Alexander Perez-Pons

Chongya Song

Sebastian De La Cruz

Systems and methods for transforming program control flow data into space-filling curves for anomaly detection are provided. A system includes emulating a microcontroller (MCU) in a virtual environment, mapping program control flow data onto a Hilbert space-filling curve to preserve execution locality and applying a convolutional neural network (CNN) to identify anomalies in the resulting spatial representation. The approach improves embedded system security by enabling efficient evaluation of control flow behavior.

Transforming program control flow data into space-filling curves for anomaly detection

Systems and methods for optimizing and/or solving objective functions are provided. Angle-based stochastic gradient descent (AG-SGD) can be used to alleviate pattern deviation(s) not resolved by related art systems and methods. AG-SGD can use the angle between the current gradient (CG) and the previous gradient (PG) to determine the new gradient (NG).

Growing community of inventors

Miami, FL, United States of America

Alexander Perez-Pons