Coordinate rotation based low complexity K-means clustering architecture
Coordinate rotation based low complexity K-means clustering architecture
In this brief, we propose a low-complexity architectural implementation of the K-means-based clustering algorithm used widely in mobile health monitoring applications for unsupervised and supervised learning. The iterative nature of the algorithm computing the distance of each data point from a respective centroid for a successful cluster formation until convergence presents a significant challenge to map it onto a low-power architecture. This has been addressed by the use of a 2-D Coordinate Rotation Digital Computer-based low-complexity engine for computing the n-dimensional Euclidean distance involved during clustering. The proposed clustering engine was synthesized using the TSMC 130-nm technology library, and a place and route was performed following which the core area and power were estimated as 0.36 mm2 and 9.21 mW at 100 MHz, respectively, making the design applicable for low-power real-time operations within a sensor node.
k-means algorithm, hardware design, CORDIC, Signal Processing, Low-power architecture
1568-1572
Adapa, Bhagyaraja
2ac6f05f-9d36-40ad-abc7-e4ddee737922
Biswas, Dwaipayan
314a210f-c293-4d18-8b07-ddaaf57a1707
Bhardwaj, Swati
f3d77c2d-151e-4262-879a-8a0085ed84ac
Raghuraman, Shashank
019ef9f9-9252-4284-8245-61a3f3630d3c
Acharyya, Amit
f7c95a87-04ac-4d13-a74c-0c4d89b1c79c
Maharatna, Koushik
93bef0a2-e011-4622-8c56-5447da4cd5dd
April 2017
Adapa, Bhagyaraja
2ac6f05f-9d36-40ad-abc7-e4ddee737922
Biswas, Dwaipayan
314a210f-c293-4d18-8b07-ddaaf57a1707
Bhardwaj, Swati
f3d77c2d-151e-4262-879a-8a0085ed84ac
Raghuraman, Shashank
019ef9f9-9252-4284-8245-61a3f3630d3c
Acharyya, Amit
f7c95a87-04ac-4d13-a74c-0c4d89b1c79c
Maharatna, Koushik
93bef0a2-e011-4622-8c56-5447da4cd5dd
Adapa, Bhagyaraja, Biswas, Dwaipayan, Bhardwaj, Swati, Raghuraman, Shashank, Acharyya, Amit and Maharatna, Koushik
(2017)
Coordinate rotation based low complexity K-means clustering architecture.
IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 25 (4), .
(doi:10.1109/TVLSI.2016.2633543).
Abstract
In this brief, we propose a low-complexity architectural implementation of the K-means-based clustering algorithm used widely in mobile health monitoring applications for unsupervised and supervised learning. The iterative nature of the algorithm computing the distance of each data point from a respective centroid for a successful cluster formation until convergence presents a significant challenge to map it onto a low-power architecture. This has been addressed by the use of a 2-D Coordinate Rotation Digital Computer-based low-complexity engine for computing the n-dimensional Euclidean distance involved during clustering. The proposed clustering engine was synthesized using the TSMC 130-nm technology library, and a place and route was performed following which the core area and power were estimated as 0.36 mm2 and 9.21 mW at 100 MHz, respectively, making the design applicable for low-power real-time operations within a sensor node.
Text Coordinate Rotation Based Low Complexity
- Accepted Manuscript
More information
Accepted/In Press date: 2 November 2016
e-pub ahead of print date: 10 January 2017
Published date: April 2017
Keywords:
k-means algorithm, hardware design, CORDIC, Signal Processing, Low-power architecture
Organisations:
Electronics & Computer Science, Electronic & Software Systems
Identifiers
Local EPrints ID: 411121
URI: https://eprints.soton.ac.uk/id/eprint/411121
ISSN: 1063-8210
PURE UUID: f1e4d82e-88a9-4e15-884a-7c9bd471d06b
Catalogue record
Date deposited: 14 Jun 2017 16:31
Last modified: 01 Dec 2017 17:32
Export record
Altmetrics
Contributors
Author:
Bhagyaraja Adapa
Author:
Dwaipayan Biswas
Author:
Swati Bhardwaj
Author:
Shashank Raghuraman
Author:
Amit Acharyya
University divisions
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics