Online fault detection for Networks-on-Chip interconnect

Abstract

A key requirement for modern Networks-on-Chip (NoC) is the ability to detect and diagnose faults and failures. A novel approach is proposed which addresses the challenge of fault detection using an online mechanism. The approach minimises online intrusion by employing dynamic rates of testing to maximize NoC throughput while still ensuring sufficient testing. This is achieved using a novel Monitor Module based on the back-off algorithm. The paper presents results on the minimal impact on the intrusion of the NoC for a range of test conditions and also highlights the low area/power overheads for scalability.