Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks

Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks

This article was published in the Eurasip Journal on Wireless Communications and Networking [©2016 Springer International Publishing.] and the definite version is available at: http://dx.doi.org/10.1186/s13638-015-0515-y. The article website is at: http://jwcn.eurasipjournals.springeropen.com/articl...

Szczegółowa specyfikacja

Opis bibliograficzny
1. autor: Khan, Muhidul Islam
Kolejni autorzy: Department of Computer Science and Engineering, BRAC University
Format: Artykuł
Język:English
Wydane: © 2016 Springer International Publishing 2016
Hasła przedmiotowe:
Adversarial bandit solvers
Cooperative reinforcement learning
Independent reinforcement learning
Resource-awareness
Task scheduling
Wireless sensor networks
Dostęp online:http://hdl.handle.net/10361/6642
http://dx.doi.org/10.1186/s13638-015-0515-y
id 10361-6642
record_format dspace
spelling 10361-66422017-01-04T05:31:11Z Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks Khan, Muhidul Islam Department of Computer Science and Engineering, BRAC University Adversarial bandit solvers Cooperative reinforcement learning Independent reinforcement learning Resource-awareness Task scheduling Wireless sensor networks This article was published in the Eurasip Journal on Wireless Communications and Networking [©2016 Springer International Publishing.] and the definite version is available at: http://dx.doi.org/10.1186/s13638-015-0515-y. The article website is at: http://jwcn.eurasipjournals.springeropen.com/articles/10.1186/s13638-015-0515-y A wireless sensor network (WSN) is composed of a large number of tiny sensor nodes. Sensor nodes are very resource-constrained, since nodes are often battery-operated and energy is a scarce resource. In this paper, a resource-aware task scheduling (RATS) method is proposed with better performance/resource consumption trade-off in a WSN. Particularly, RATS exploits an adversarial bandit solver method called exponential weight for exploration and exploitation (Exp3) for target tracking application of WSN. The proposed RATS method is compared and evaluated with the existing scheduling methods exploiting online learning: distributed independent reinforcement learning (DIRL), reinforcement learning (RL), and cooperative reinforcement learning (CRL), in terms of the tracking quality/energy consumption trade-off in a target tracking application. The communication overhead and computational effort of these methods are also computed. Simulation results show that the proposed RATS outperforms the existing methods DIRL and RL in terms of achieved tracking performance. © 2016, Khan. Published 2016-10-31T06:13:48Z 2016-10-31T06:13:48Z 2016 Article Muhidulislam, K. (2016). Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks. Eurasip Journal on Wireless Communications and Networking, 1–17. doi:doi.org/10.1186/s13638-015-0515-y 16871472 http://hdl.handle.net/10361/6642 http://dx.doi.org/10.1186/s13638-015-0515-y en http://jwcn.eurasipjournals.springeropen.com/articles/10.1186/s13638-015-0515-y © 2016 Springer International Publishing
institution Brac University
collection Institutional Repository
language English
topic Adversarial bandit solvers
Cooperative reinforcement learning
Independent reinforcement learning
Resource-awareness
Task scheduling
Wireless sensor networks
spellingShingle Adversarial bandit solvers
Cooperative reinforcement learning
Independent reinforcement learning
Resource-awareness
Task scheduling
Wireless sensor networks
Khan, Muhidul Islam
Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
description This article was published in the Eurasip Journal on Wireless Communications and Networking [©2016 Springer International Publishing.] and the definite version is available at: http://dx.doi.org/10.1186/s13638-015-0515-y. The article website is at: http://jwcn.eurasipjournals.springeropen.com/articles/10.1186/s13638-015-0515-y
author2 Department of Computer Science and Engineering, BRAC University
author_facet Department of Computer Science and Engineering, BRAC University
Khan, Muhidul Islam
format Article
author Khan, Muhidul Islam
author_sort Khan, Muhidul Islam
title Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
title_short Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
title_full Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
title_fullStr Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
title_full_unstemmed Resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
title_sort resource-aware task scheduling by an adversarial bandit solver method in wireless sensor networks
publisher © 2016 Springer International Publishing
publishDate 2016
url http://hdl.handle.net/10361/6642
http://dx.doi.org/10.1186/s13638-015-0515-y
work_keys_str_mv AT khanmuhidulislam resourceawaretaskschedulingbyanadversarialbanditsolvermethodinwirelesssensornetworks
_version_ 1814309683176931328

Podobne zapisy