Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things

Hawzhin Mohammed; Tolulope A. Odetola; Nan Guo; Syed Rafay Hasan

doi:10.1109/MWSCAS47672.2021.9531765

Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things

Hawzhin Mohammed
, Tolulope A. Odetola
, Nan Guo
, Syed Rafay Hasan

Division of Business, Engineering, and Information Sciences & Technology (Altoona)

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

1 Scopus citations

Abstract

In this paper, dynamic deployment of Convolutional Neural Network (CNN) architecture is proposed utilizing only IoT-level devices. By partitioning and pipelining the CNN, it horizontally distributes the computation load among resource-constrained devices (called as horizontal collaboration), which in turn increases the throughput. Through partitioning, we can decrease the computation and energy consumption on individual IoT devices and increase the throughput without sacrificing accuracy. Also, by processing the data at the generation point, data privacy can be achieved. The results show that throughput can be increased by 1.55x to 1.75x for sharing the CNN into two and three resource-constrained devices, respectively.

Original language	English (US)
Title of host publication	2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	330-333
Number of pages	4
ISBN (Electronic)	9781665424615
DOIs	https://doi.org/10.1109/MWSCAS47672.2021.9531765
State	Published - Aug 9 2021
Event	2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Virtual, East Lansing, United States Duration: Aug 9 2021 → Aug 11 2021

Publication series

Name	Midwest Symposium on Circuits and Systems
Volume	2021-August
ISSN (Print)	1548-3746

Conference

Conference	2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021
Country/Territory	United States
City	Virtual, East Lansing
Period	8/9/21 → 8/11/21

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Electrical and Electronic Engineering

Access to Document

10.1109/MWSCAS47672.2021.9531765

Cite this

Mohammed, H., Odetola, T. A., Guo, N., & Hasan, S. R. (2021). Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things. In 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings (pp. 330-333). (Midwest Symposium on Circuits and Systems; Vol. 2021-August). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MWSCAS47672.2021.9531765

@inproceedings{0a26434f1c3b4de480e8896059a527d3,

title = "Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things",

abstract = "In this paper, dynamic deployment of Convolutional Neural Network (CNN) architecture is proposed utilizing only IoT-level devices. By partitioning and pipelining the CNN, it horizontally distributes the computation load among resource-constrained devices (called as horizontal collaboration), which in turn increases the throughput. Through partitioning, we can decrease the computation and energy consumption on individual IoT devices and increase the throughput without sacrificing accuracy. Also, by processing the data at the generation point, data privacy can be achieved. The results show that throughput can be increased by 1.55x to 1.75x for sharing the CNN into two and three resource-constrained devices, respectively.",

author = "Hawzhin Mohammed and Odetola, \{Tolulope A.\} and Nan Guo and Hasan, \{Syed Rafay\}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 ; Conference date: 09-08-2021 Through 11-08-2021",

year = "2021",

month = aug,

day = "9",

doi = "10.1109/MWSCAS47672.2021.9531765",

language = "English (US)",

series = "Midwest Symposium on Circuits and Systems",

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

pages = "330--333",

booktitle = "2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings",

address = "United States",

}

Mohammed, H, Odetola, TA, Guo, N & Hasan, SR 2021, Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things. in 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings. Midwest Symposium on Circuits and Systems, vol. 2021-August, Institute of Electrical and Electronics Engineers Inc., pp. 330-333, 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021, Virtual, East Lansing, United States, 8/9/21. https://doi.org/10.1109/MWSCAS47672.2021.9531765

Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things. / Mohammed, Hawzhin; Odetola, Tolulope A.; Guo, Nan et al.
2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2021. p. 330-333 (Midwest Symposium on Circuits and Systems; Vol. 2021-August).

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

TY - GEN

T1 - Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things

AU - Mohammed, Hawzhin

AU - Odetola, Tolulope A.

AU - Guo, Nan

AU - Hasan, Syed Rafay

PY - 2021/8/9

Y1 - 2021/8/9

N2 - In this paper, dynamic deployment of Convolutional Neural Network (CNN) architecture is proposed utilizing only IoT-level devices. By partitioning and pipelining the CNN, it horizontally distributes the computation load among resource-constrained devices (called as horizontal collaboration), which in turn increases the throughput. Through partitioning, we can decrease the computation and energy consumption on individual IoT devices and increase the throughput without sacrificing accuracy. Also, by processing the data at the generation point, data privacy can be achieved. The results show that throughput can be increased by 1.55x to 1.75x for sharing the CNN into two and three resource-constrained devices, respectively.

AB - In this paper, dynamic deployment of Convolutional Neural Network (CNN) architecture is proposed utilizing only IoT-level devices. By partitioning and pipelining the CNN, it horizontally distributes the computation load among resource-constrained devices (called as horizontal collaboration), which in turn increases the throughput. Through partitioning, we can decrease the computation and energy consumption on individual IoT devices and increase the throughput without sacrificing accuracy. Also, by processing the data at the generation point, data privacy can be achieved. The results show that throughput can be increased by 1.55x to 1.75x for sharing the CNN into two and three resource-constrained devices, respectively.

UR - https://www.scopus.com/pages/publications/85115630653

UR - https://www.scopus.com/pages/publications/85115630653#tab=citedBy

U2 - 10.1109/MWSCAS47672.2021.9531765

DO - 10.1109/MWSCAS47672.2021.9531765

M3 - Conference contribution

AN - SCOPUS:85115630653

T3 - Midwest Symposium on Circuits and Systems

SP - 330

EP - 333

BT - 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021

Y2 - 9 August 2021 through 11 August 2021

ER -

Mohammed H, Odetola TA, Guo N, Hasan SR. Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things. In 2021 IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2021 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2021. p. 330-333. (Midwest Symposium on Circuits and Systems). doi: 10.1109/MWSCAS47672.2021.9531765

Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things

Abstract

Publication series

Conference

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this