Modeling the Dependency of Analog Circuit Performance Parameters on Manufacturing Process Variations With Applications in Sensitivity Analysis and Yield Prediction

Sandru, Elena-Diana; David, Emilian; Kovacs, Ingrid; Buzo, Andi; Burileanu, Corneliu; Pelz, Georg

doi:10.1109/tcad.2021.3054804

Cited by 6 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lot of papers, 2,4,6,11,[13][14][15][16][17][18][19][20][21][22][23] however, investigate yield prediction with supervised regression models. They are summarised in Table 2.…”

Section: Regression Modelsmentioning

confidence: 99%

Wafer yield prediction using AI: potentials and pitfalls

Busch

Wahl

Choubey

2023

Metrology, Inspection, and Process Control XXXVII

View full text Add to dashboard Cite

Yield prediction is highly beneficial in semiconductor manufacturing and hence has attracted significant research interest. A number of recent techniques to do so have utilized machine-learning techniques to improve prediction accuracy. However, the diversity of algorithms used in the wide application domains of semiconductors fabrication make it difficult to compare them as well as to identify key models to use in future. In this paper, we navigate this diversity by conducting an extensive literature research with the aspects of the different data used and the different algorithms applied. Various ML algorithms have their unique strengths and weaknesses, which makes them differently suited for different aspects of yield prediction. In addition, we also consider whether regression analyses or classification analyses are used to nd potentials and pitfalls in this area. We will analyse the type of data being used and various pre-processing techniques proposed. In addition, we will also analyse the practical implementation of these models. A lot has already been achieved on yield prediction, but every future improvement that enables significant cost saving by making it easier to optimally coordinate the individual processes and perhaps detect errors earlier is more than welcome. This review paper provides a foundation for model and data selection for yield prediction based on current state of the art.

show abstract

“…A lot of papers, 2,4,6,11,[13][14][15][16][17][18][19][20][21][22][23] however, investigate yield prediction with supervised regression models. They are summarised in Table 2.…”

Section: Regression Modelsmentioning

confidence: 99%

Wafer yield prediction using AI: potentials and pitfalls

Busch

Wahl

Choubey

2023

Metrology, Inspection, and Process Control XXXVII

View full text Add to dashboard Cite

show abstract

“…One can also rely on dedicated on-chip test structures for extracting useful features. For example, one can rely on the SymBIST principle [84], Process Control Monitors (PCMs) at die-level [132,2,3,28,133], amplitude detectors [119], current sensors [29], digital signatures of analog waveforms [14], etc. Another approach is to optimize the features towards a low generalization error for the machine learning models [120,6].…”

Section: Extractionmentioning

confidence: 99%

Machine Learning Support for Diagnosis of Analog Circuits

Stratigopoulos

2022

Machine Learning Support for Fault Diagnosis of System-on-Chip

View full text Add to dashboard Cite

We discuss the state-of-the-art on fault diagnosis for analog circuits with a focus on techniques that leverage machine learning. For a chip that has failed either in post-manufacturing testing or in the field of operation, fault diagnosis is launched to identify the root-cause of failure at sub-block level and transistor-level. In this context, machine learning can be used to build a smart system that predicts the fault that has occurred from diagnostic measurements extracted on the chip. We discuss the different elements of a diagnosis flow for analog circuits, including fault modeling, fault simulation, diagnostic measurement extraction and selection, and the machine learning algorithms that compose the prediction system. We also demonstrate a machine learning-based diagnosis flow on an industrial case study.

show abstract

“…Industrial analogue Integrated Circuits (ICs) design is not limited to fully optimised nominal design solutions, especially in nanometre Complementary Metal-Oxide-Semiconductor (CMOS) technologies [1]. The ratio of circuits meeting all the desired specifications in a system among all the fabricated circuits is called yield [2].…”

Section: Introductionmentioning

confidence: 99%

A four‐stage yield optimization technique for analog integrated circuits using optimal computing budget allocation and evolutionary algorithms

Yaseri

Maghami

Radmehr

2022

IET Computers & Digital Tech

View full text Add to dashboard Cite

A high yield estimation is necessary for designing analogue integrated circuits. In the Monte‐Carlo (MC) method, many transistor‐level simulations should be performed to obtain the desired result. Therefore, some methods are needed to be combined with MC simulations to reach high yield with high speed at the same time. In this paper, a four‐stage yield optimisation approach is presented, which employs computational intelligence to accelerate yield estimation without losing accuracy. Firstly, the designs that met the desired characteristics are provided using critical analysis (CA). The aim of utilising CA is to avoid unnecessary MC simulations repeating for non‐critical solutions. Then in the second and third stages, the shuffled frog‐leaping algorithm and the Non‐dominated Sorting Genetic Algorithm‐III are proposed to improve the performance. Finally, MC simulations are performed to present the final result. The yield value obtained from the simulation results for two‐stage class‐AB Operational Transconductance Amplifer (OTA) in 180 nm Complementary Metal‐Oxide‐Semiconductor (CMOS) technology is 99.85%. The proposed method has less computational effort and high accuracy than the MC‐based approaches. Another advantage of using CA is that the initial population of multi‐objective optimisation algorithms will no longer be random. Simulation results prove the efficiency of the proposed technique.

show abstract

Modeling the Dependency of Analog Circuit Performance Parameters on Manufacturing Process Variations With Applications in Sensitivity Analysis and Yield Prediction

Cited by 6 publications

References 39 publications

Wafer yield prediction using AI: potentials and pitfalls

Wafer yield prediction using AI: potentials and pitfalls

Machine Learning Support for Diagnosis of Analog Circuits

A four‐stage yield optimization technique for analog integrated circuits using optimal computing budget allocation and evolutionary algorithms

Contact Info

Product

Resources

About