NMCu-CNN : a scalable near-memory computing co-processor on a RISC-V MCU in 22 nm FDSOI
(2025)- Abstract
- This paper presents NMCu-CNN, a near-memory computing (NMC) co-processor for the hardware acceleration of convolutional neural networks (CNNs) on a low-power, flexible MCU platform. The scalable architecture and the selected platform enable adaptability to the rapidly evolving Edge AIoT landscape, where energy and performance requirements are constrained. Application-tailored NMC units, equipped with an optimized CNN dataflow, are integrated into the shared memory address space of a RISC-V-based MCU. The proposed architecture supports highly flexible runtime configurability, achieving $94 \%$ computational efficiency. Fabricated in 22 nm FDSOI technology, NMCu-CNN delivers a performance of 203 GOPS and an energy efficiency of 1716 GOPS/W... (More)
- This paper presents NMCu-CNN, a near-memory computing (NMC) co-processor for the hardware acceleration of convolutional neural networks (CNNs) on a low-power, flexible MCU platform. The scalable architecture and the selected platform enable adaptability to the rapidly evolving Edge AIoT landscape, where energy and performance requirements are constrained. Application-tailored NMC units, equipped with an optimized CNN dataflow, are integrated into the shared memory address space of a RISC-V-based MCU. The proposed architecture supports highly flexible runtime configurability, achieving $94 \%$ computational efficiency. Fabricated in 22 nm FDSOI technology, NMCu-CNN delivers a performance of 203 GOPS and an energy efficiency of 1716 GOPS/W (1.7 TOPS/W) benchmarked on convolutional layers of a CNN model, outperforming the processing capabilities of other state-of-the-art techniques. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/65fa6acd-ab06-4bd0-8405-89863f1d4fe8
- author
- Prieto, Arturo LU ; Nouripayam, Masoud LU ; Westring, Kristoffer LU ; Castillo Mohedano, Sergio LU ; Allfjord, Alex LU ; Svensson, Linus LU ; Andersson, Per LU and Rodrigues, Joachim LU
- organization
- publishing date
- 2025-11-27
- type
- Contribution to conference
- publication status
- published
- subject
- pages
- 5 pages
- DOI
- 10.1109/SiPS66314.2025.11261229
- language
- English
- LU publication?
- yes
- id
- 65fa6acd-ab06-4bd0-8405-89863f1d4fe8
- date added to LUP
- 2025-10-28 19:11:07
- date last changed
- 2025-12-09 09:59:48
@misc{65fa6acd-ab06-4bd0-8405-89863f1d4fe8,
abstract = {{This paper presents NMCu-CNN, a near-memory computing (NMC) co-processor for the hardware acceleration of convolutional neural networks (CNNs) on a low-power, flexible MCU platform. The scalable architecture and the selected platform enable adaptability to the rapidly evolving Edge AIoT landscape, where energy and performance requirements are constrained. Application-tailored NMC units, equipped with an optimized CNN dataflow, are integrated into the shared memory address space of a RISC-V-based MCU. The proposed architecture supports highly flexible runtime configurability, achieving $94 \%$ computational efficiency. Fabricated in 22 nm FDSOI technology, NMCu-CNN delivers a performance of 203 GOPS and an energy efficiency of 1716 GOPS/W (1.7 TOPS/W) benchmarked on convolutional layers of a CNN model, outperforming the processing capabilities of other state-of-the-art techniques.}},
author = {{Prieto, Arturo and Nouripayam, Masoud and Westring, Kristoffer and Castillo Mohedano, Sergio and Allfjord, Alex and Svensson, Linus and Andersson, Per and Rodrigues, Joachim}},
language = {{eng}},
month = {{11}},
title = {{NMCu-CNN : a scalable near-memory computing co-processor on a RISC-V MCU in 22 nm FDSOI}},
url = {{http://dx.doi.org/10.1109/SiPS66314.2025.11261229}},
doi = {{10.1109/SiPS66314.2025.11261229}},
year = {{2025}},
}