On the Estimation of Complex Circuits Functional Failure Rate by Machine Learning Techniques

Abstract

De-Rating or Vulnerability Factors are a major feature of failure analysis efforts mandated by today's Functional Safety requirements. Determining the Functional De-Rating of sequential logic cells typically requires computationally intensive fault-injection simulation campaigns. In this paper a new approach is proposed which uses Machine Learning to estimate the Functional De-Rating of individual flip-flops and thus, optimising and enhancing fault injection efforts. Therefore, first, a set of per-instance features is described and extracted through an analysis approach combining static elements (cell properties, circuit structure, synthesis attributes) and dynamic elements (signal activity). Second, reference data is obtained through first-principles fault simulation approaches. Finally, one part of the reference dataset is used to train the Machine Learning algorithm and the remaining is used to validate and benchmark the accuracy of the trained tool. The intended goal is to obtain a trained model able to provide accurate per-instance Functional De-Rating data for the full list of circuit instances, an objective that is difficult to reach using classical methods. The presented methodology is accompanied by a practical example to determine the performance of various Machine Learning models for different training sizes.

PULP Fiction No More—Dependable PULP Systems for Space

“…Previous studies have applied ML to reliability analysis, including using ML algorithms to determine the relationship between fault injection outcomes and the characteristics of applications and platforms [30], predicting functional de-rating (FDR) [31], [32], and anticipating hardware defects at the transistor level [33]. Our work aims to utilize ML to predict critical flip-flops in circuits.…”

Exploring Fault Parameter Space Using Reinforcement Learning-based Fault Injection

“…ML methods can also be used for optimizing FI in different application and perspectives. In [14], a ML algorithm is utilized to reduce the computational cost for Functional De-Rating of individual flip-flops. They trained an algorithm for one basic circuit then extend it to another sequential circuit with more complexity.…”

Failure Identification Using Model-Implemented Fault Injection with Domain Knowledge-Guided Reinforcement Learning

“…Without this information, the test engineer cannot inject a fault in the proper location, and most of the injected faults would not lead to failure. (iii) The literature has mainly studied simple fault models, for example static faults, meaning that the fault parameters are fixed during simulation, such as stuck-at faults and bit flips [ 18 , 19 ]. In the case of dynamic faults, the fault’s parameters change over the simulation-like noise of the signal.…”