A fully analog pipelined deep neural network (DNN) accelerator is proposed, which is constructed by using pipeline registers based on master-slave switched capacitors. The idea of the master-slave switched capacitors is an analog equivalent of the delayed flip-flop (D-FF) which has been used as a digital pipeline register. To estimate the performance of the pipeline register, it is applied to a conventional DNN which performs non-pipeline operation. Compared with the conventional DNN, the cycle time is reduced by 61.5% and data rate is increased by 160%. The accuracy reaches 99.6% in MNIST classification test. The energy consumption per classification is reduced by 88.2% to 0.128µJ, achieving an energy efficiency of 1.05TOPS/W and a throughput of 0.538TOPS in 180nm technology node.
Yaxin MEI
Waseda University
Takashi OHSAWA
Waseda University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yaxin MEI, Takashi OHSAWA, "A Fully Analog Deep Neural Network Inference Accelerator with Pipeline Registers Based on Master-Slave Switched Capacitors" in IEICE TRANSACTIONS on Electronics,
vol. E106-C, no. 9, pp. 477-485, September 2023, doi: 10.1587/transele.2022ECP5049.
Abstract: A fully analog pipelined deep neural network (DNN) accelerator is proposed, which is constructed by using pipeline registers based on master-slave switched capacitors. The idea of the master-slave switched capacitors is an analog equivalent of the delayed flip-flop (D-FF) which has been used as a digital pipeline register. To estimate the performance of the pipeline register, it is applied to a conventional DNN which performs non-pipeline operation. Compared with the conventional DNN, the cycle time is reduced by 61.5% and data rate is increased by 160%. The accuracy reaches 99.6% in MNIST classification test. The energy consumption per classification is reduced by 88.2% to 0.128µJ, achieving an energy efficiency of 1.05TOPS/W and a throughput of 0.538TOPS in 180nm technology node.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.2022ECP5049/_p
Copy
@ARTICLE{e106-c_9_477,
author={Yaxin MEI, Takashi OHSAWA, },
journal={IEICE TRANSACTIONS on Electronics},
title={A Fully Analog Deep Neural Network Inference Accelerator with Pipeline Registers Based on Master-Slave Switched Capacitors},
year={2023},
volume={E106-C},
number={9},
pages={477-485},
abstract={A fully analog pipelined deep neural network (DNN) accelerator is proposed, which is constructed by using pipeline registers based on master-slave switched capacitors. The idea of the master-slave switched capacitors is an analog equivalent of the delayed flip-flop (D-FF) which has been used as a digital pipeline register. To estimate the performance of the pipeline register, it is applied to a conventional DNN which performs non-pipeline operation. Compared with the conventional DNN, the cycle time is reduced by 61.5% and data rate is increased by 160%. The accuracy reaches 99.6% in MNIST classification test. The energy consumption per classification is reduced by 88.2% to 0.128µJ, achieving an energy efficiency of 1.05TOPS/W and a throughput of 0.538TOPS in 180nm technology node.},
keywords={},
doi={10.1587/transele.2022ECP5049},
ISSN={1745-1353},
month={September},}
Copy
TY - JOUR
TI - A Fully Analog Deep Neural Network Inference Accelerator with Pipeline Registers Based on Master-Slave Switched Capacitors
T2 - IEICE TRANSACTIONS on Electronics
SP - 477
EP - 485
AU - Yaxin MEI
AU - Takashi OHSAWA
PY - 2023
DO - 10.1587/transele.2022ECP5049
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E106-C
IS - 9
JA - IEICE TRANSACTIONS on Electronics
Y1 - September 2023
AB - A fully analog pipelined deep neural network (DNN) accelerator is proposed, which is constructed by using pipeline registers based on master-slave switched capacitors. The idea of the master-slave switched capacitors is an analog equivalent of the delayed flip-flop (D-FF) which has been used as a digital pipeline register. To estimate the performance of the pipeline register, it is applied to a conventional DNN which performs non-pipeline operation. Compared with the conventional DNN, the cycle time is reduced by 61.5% and data rate is increased by 160%. The accuracy reaches 99.6% in MNIST classification test. The energy consumption per classification is reduced by 88.2% to 0.128µJ, achieving an energy efficiency of 1.05TOPS/W and a throughput of 0.538TOPS in 180nm technology node.
ER -