Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads

Shang Kai Wei; Steven D. Gardner; Mohammad R. Haider; J. Iwan D. Alexander; Yehia Massoud

doi:10.1109/MWSCAS.2019.8884854

Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads

Shang Kai Wei, Steven D. Gardner, Mohammad R. Haider, J. Iwan D. Alexander, Yehia Massoud

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

With the increase of the number of sensors in an Internet-of-Things (IoT) environment, there is a critical need of in-situ signal processing for local intelligence. This work presents a ring-oscillator based neural network (ONN) for sensor signal processing. A series of 36 ring oscillators are coupled serially and the network performs computations by synchronization. The coupling weights are modulated by sensor inputs which in turn affect the stored memory by indication of synchronization time and frequency of the network. This work studies the effects of synchronization time and frequency of the ONN for different supply voltages, coupling weights and number of oscillating nodes. The 36 oscillator ring neural network scheme has been designed and simulated using 130 nm SiGe BiCMOS process. Simulation results show power consumption to be less than 5.6 mW for the entire ONN and pattern recognition based on synchronization frequency and time differences between the training and input patterns.

Original language	English (US)
Title of host publication	Midwest Symposium on Circuits and Systems
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	61-64
Number of pages	4
ISBN (Print)	9781728127880
DOIs	https://doi.org/10.1109/MWSCAS.2019.8884854
State	Published - Aug 1 2019
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2022-09-13

Access to Document

10.1109/MWSCAS.2019.8884854

Cite this

@inproceedings{7ab73757eddd4c7e834f28dc574fbe31,

title = "Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads",

abstract = "With the increase of the number of sensors in an Internet-of-Things (IoT) environment, there is a critical need of in-situ signal processing for local intelligence. This work presents a ring-oscillator based neural network (ONN) for sensor signal processing. A series of 36 ring oscillators are coupled serially and the network performs computations by synchronization. The coupling weights are modulated by sensor inputs which in turn affect the stored memory by indication of synchronization time and frequency of the network. This work studies the effects of synchronization time and frequency of the ONN for different supply voltages, coupling weights and number of oscillating nodes. The 36 oscillator ring neural network scheme has been designed and simulated using 130 nm SiGe BiCMOS process. Simulation results show power consumption to be less than 5.6 mW for the entire ONN and pattern recognition based on synchronization frequency and time differences between the training and input patterns.",

author = "Wei, {Shang Kai} and Gardner, {Steven D.} and Haider, {Mohammad R.} and Alexander, {J. Iwan D.} and Yehia Massoud",

note = "Generated from Scopus record by KAUST IRTS on 2022-09-13",

year = "2019",

month = aug,

day = "1",

doi = "10.1109/MWSCAS.2019.8884854",

language = "English (US)",

isbn = "9781728127880",

pages = "61--64",

booktitle = "Midwest Symposium on Circuits and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

}

Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads. / Wei, Shang Kai; Gardner, Steven D.; Haider, Mohammad R. et al.
Midwest Symposium on Circuits and Systems. Institute of Electrical and Electronics Engineers Inc., 2019. p. 61-64.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads

AU - Wei, Shang Kai

AU - Gardner, Steven D.

AU - Haider, Mohammad R.

AU - Alexander, J. Iwan D.

AU - Massoud, Yehia

N1 - Generated from Scopus record by KAUST IRTS on 2022-09-13

PY - 2019/8/1

Y1 - 2019/8/1

N2 - With the increase of the number of sensors in an Internet-of-Things (IoT) environment, there is a critical need of in-situ signal processing for local intelligence. This work presents a ring-oscillator based neural network (ONN) for sensor signal processing. A series of 36 ring oscillators are coupled serially and the network performs computations by synchronization. The coupling weights are modulated by sensor inputs which in turn affect the stored memory by indication of synchronization time and frequency of the network. This work studies the effects of synchronization time and frequency of the ONN for different supply voltages, coupling weights and number of oscillating nodes. The 36 oscillator ring neural network scheme has been designed and simulated using 130 nm SiGe BiCMOS process. Simulation results show power consumption to be less than 5.6 mW for the entire ONN and pattern recognition based on synchronization frequency and time differences between the training and input patterns.

AB - With the increase of the number of sensors in an Internet-of-Things (IoT) environment, there is a critical need of in-situ signal processing for local intelligence. This work presents a ring-oscillator based neural network (ONN) for sensor signal processing. A series of 36 ring oscillators are coupled serially and the network performs computations by synchronization. The coupling weights are modulated by sensor inputs which in turn affect the stored memory by indication of synchronization time and frequency of the network. This work studies the effects of synchronization time and frequency of the ONN for different supply voltages, coupling weights and number of oscillating nodes. The 36 oscillator ring neural network scheme has been designed and simulated using 130 nm SiGe BiCMOS process. Simulation results show power consumption to be less than 5.6 mW for the entire ONN and pattern recognition based on synchronization frequency and time differences between the training and input patterns.

UR - https://ieeexplore.ieee.org/document/8884854/

UR - http://www.scopus.com/inward/record.url?scp=85074970853&partnerID=8YFLogxK

U2 - 10.1109/MWSCAS.2019.8884854

DO - 10.1109/MWSCAS.2019.8884854

M3 - Conference contribution

SN - 9781728127880

SP - 61

EP - 64

BT - Midwest Symposium on Circuits and Systems

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Energy-Efficient Ring-Oscillator Signal Processing Scheme for Pattern Recognition and Reduced Computational Loads

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this