RCNet: $ΔΣ$ IADCs as Recurrent AutoEncoders

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Verdant, Arnaud, Guicquero, William, Chossat, Jérôme
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918066119507968
author Verdant, Arnaud
Guicquero, William
Chossat, Jérôme
author_facet Verdant, Arnaud
Guicquero, William
Chossat, Jérôme
contents This paper proposes a deep learning model (RCNet) for Delta-Sigma ($ΔΣ$) ADCs. Recurrent Neural Networks (RNNs) allow to describe both modulators and filters. This analogy is applied to Incremental ADCs (IADC). High-end optimizers combined with full-custom losses are used to define additional hardware design constraints: quantized weights, signal saturation, temporal noise injection, devices area. Focusing on DC conversion, our early results demonstrate that $SNR$ defined as an Effective Number Of Bits (ENOB) can be optimized under a certain hardware mapping complexity. The proposed RCNet succeeded to provide design tradeoffs in terms of $SNR$ ($>$13bit) versus area constraints ($<$14pF total capacitor) at a given $OSR$ (80 samples). Interestingly, it appears that the best RCNet architectures do not necessarily rely on high-order modulators, leveraging additional topology exploration degrees of freedom.
format Preprint
id arxiv_https___arxiv_org_abs_2506_16903
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RCNet: $ΔΣ$ IADCs as Recurrent AutoEncoders
Verdant, Arnaud
Guicquero, William
Chossat, Jérôme
Hardware Architecture
Machine Learning
This paper proposes a deep learning model (RCNet) for Delta-Sigma ($ΔΣ$) ADCs. Recurrent Neural Networks (RNNs) allow to describe both modulators and filters. This analogy is applied to Incremental ADCs (IADC). High-end optimizers combined with full-custom losses are used to define additional hardware design constraints: quantized weights, signal saturation, temporal noise injection, devices area. Focusing on DC conversion, our early results demonstrate that $SNR$ defined as an Effective Number Of Bits (ENOB) can be optimized under a certain hardware mapping complexity. The proposed RCNet succeeded to provide design tradeoffs in terms of $SNR$ ($>$13bit) versus area constraints ($<$14pF total capacitor) at a given $OSR$ (80 samples). Interestingly, it appears that the best RCNet architectures do not necessarily rely on high-order modulators, leveraging additional topology exploration degrees of freedom.
title RCNet: $ΔΣ$ IADCs as Recurrent AutoEncoders
topic Hardware Architecture
Machine Learning
url https://arxiv.org/abs/2506.16903