A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators

Weison Lin; Tughrul Arslan

doi:10.1109/ICECS53924.2021.9665510

A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators

Weison Lin, Tughrul Arslan

School of Engineering

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

Abstract / Description of output

Edge AI accelerators have been emerging as a solution for near customers' applications in areas such as image recognition sensors, remote sensing satellites, robotics, wearable devices, and drones. These applications require meeting performance targets and strict area and power constraints due to their portable mobility feature and limited power sources. As a result, a column streaming-based convolution engine has been proposed in this paper that includes column sets of processing elements design for flexibility in terms of the applicability for different CNN algorithms in edge AI accelerators. Compared to a commercialized CNN accelerator, the key results reveal that the column streaming-based convolution engine requires similar execution cycles to process a 227 × 227 feature map and avoid zero-padding penalties.

Original language	English
Title of host publication	2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers
Pages	1-6
Number of pages	6
ISBN (Electronic)	9781728182810
DOIs	https://doi.org/10.1109/ICECS53924.2021.9665510
Publication status	Published - 10 Jan 2022
Event	28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Dubai, United Arab Emirates Duration: 28 Nov 2021 → 1 Dec 2021

Conference

Conference	28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021
Country/Territory	United Arab Emirates
City	Dubai
Period	28/11/21 → 1/12/21

Keywords / Materials (for Non-textual outputs)

CNN mapping algorithm
convolution engine
edge AI accelerators
PERFORMANCE EVALUATION
satellites
Convolution
Wearable computers
AI accelerators
Streaming media
Robot sensing systems

Access to Document

10.1109/ICECS53924.2021.9665510

Cite this

@inproceedings{ab43b7cdf07f497eb388bea694c7df1a,

title = "A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators",

abstract = "Edge AI accelerators have been emerging as a solution for near customers' applications in areas such as image recognition sensors, remote sensing satellites, robotics, wearable devices, and drones. These applications require meeting performance targets and strict area and power constraints due to their portable mobility feature and limited power sources. As a result, a column streaming-based convolution engine has been proposed in this paper that includes column sets of processing elements design for flexibility in terms of the applicability for different CNN algorithms in edge AI accelerators. Compared to a commercialized CNN accelerator, the key results reveal that the column streaming-based convolution engine requires similar execution cycles to process a 227 × 227 feature map and avoid zero-padding penalties. ",

keywords = "CNN mapping algorithm, convolution engine, edge AI accelerators, PERFORMANCE EVALUATION, satellites, Convolution, Wearable computers, AI accelerators, Streaming media, Robot sensing systems",

author = "Weison Lin and Tughrul Arslan",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 ; Conference date: 28-11-2021 Through 01-12-2021",

year = "2022",

month = jan,

day = "10",

doi = "10.1109/ICECS53924.2021.9665510",

language = "English",

pages = "1--6",

booktitle = "2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Lin, W & Arslan, T 2022, A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators. in 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings., 9665510, Institute of Electrical and Electronics Engineers, pp. 1-6, 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021, Dubai, United Arab Emirates, 28/11/21. https://doi.org/10.1109/ICECS53924.2021.9665510

A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators. / Lin, Weison; Arslan, Tughrul.
2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings. Institute of Electrical and Electronics Engineers, 2022. p. 1-6 9665510.

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

TY - GEN

T1 - A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators

AU - Lin, Weison

AU - Arslan, Tughrul

PY - 2022/1/10

Y1 - 2022/1/10

N2 - Edge AI accelerators have been emerging as a solution for near customers' applications in areas such as image recognition sensors, remote sensing satellites, robotics, wearable devices, and drones. These applications require meeting performance targets and strict area and power constraints due to their portable mobility feature and limited power sources. As a result, a column streaming-based convolution engine has been proposed in this paper that includes column sets of processing elements design for flexibility in terms of the applicability for different CNN algorithms in edge AI accelerators. Compared to a commercialized CNN accelerator, the key results reveal that the column streaming-based convolution engine requires similar execution cycles to process a 227 × 227 feature map and avoid zero-padding penalties.

AB - Edge AI accelerators have been emerging as a solution for near customers' applications in areas such as image recognition sensors, remote sensing satellites, robotics, wearable devices, and drones. These applications require meeting performance targets and strict area and power constraints due to their portable mobility feature and limited power sources. As a result, a column streaming-based convolution engine has been proposed in this paper that includes column sets of processing elements design for flexibility in terms of the applicability for different CNN algorithms in edge AI accelerators. Compared to a commercialized CNN accelerator, the key results reveal that the column streaming-based convolution engine requires similar execution cycles to process a 227 × 227 feature map and avoid zero-padding penalties.

KW - CNN mapping algorithm

KW - convolution engine

KW - edge AI accelerators

KW - PERFORMANCE EVALUATION

KW - satellites

KW - Convolution

KW - Wearable computers

KW - AI accelerators

KW - Streaming media

KW - Robot sensing systems

UR - http://www.scopus.com/inward/record.url?scp=85124626763&partnerID=8YFLogxK

U2 - 10.1109/ICECS53924.2021.9665510

DO - 10.1109/ICECS53924.2021.9665510

M3 - Conference contribution

AN - SCOPUS:85124626763

SP - 1

EP - 6

BT - 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings

PB - Institute of Electrical and Electronics Engineers

T2 - 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021

Y2 - 28 November 2021 through 1 December 2021

ER -

A Column Streaming-Based Convolution Engine and Mapping Algorithm for CNN-based Edge AI Accelerators

Abstract / Description of output

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Other files and links

Fingerprint

Cite this