Improvement of pre-partitioning on reinforcement learning based spectrum sharing

Tao Jiang; David Grace; Paul D Mitchell

Improvement of pre-partitioning on reinforcement learning based spectrum sharing

Electronic Engineering

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

Abstract

In this paper, we present a novel technique for reinforcement learning based spectrum sharing schemes. It shows how the potential action space of users is reduced by randomly partitioning the spectrum for different users. Cognitive agents are therefore able to finish their exploration stage faster than more basic reinforcement learning based schemes. Simulation results show how the communication system is more stable if all users reserve a subset of resources at the beginning and then learn to select the best channels in the reserved spectrum set. The system performance by using pre-partitioning is investigated and comparison with the traditional reinforcement learning based schemes is given.

Original language	English
Title of host publication	IET International Communication Conference on Wireless Mobile & Computing (CCWMC), Shanghai, China, December 2009
Publisher	IEEE
Pages	299 -302
Number of pages	4
Volume	2009
Edition	562 CP
Publication status	Published - 1 Dec 2009
Event	International Communication Conference on Wireless Mobile and Computing (CCWMC 2009) - Shanghai, China Duration: 7 Dec 2009 → 9 Dec 2009

Conference

Conference	International Communication Conference on Wireless Mobile and Computing (CCWMC 2009)
Country/Territory	China
City	Shanghai
Period	7/12/09 → 9/12/09

Bibliographical note

Uploaded in accordance with the publisher's self-archiving policy.

Access to Document

Improvement of Pre-partitioning on Reinforcement Learning Based Spectrum SharingFinal published version, 119 KB

Cite this

@inproceedings{7b6f0d73760245e29623c78bb0fcbe43,

title = "Improvement of pre-partitioning on reinforcement learning based spectrum sharing",

abstract = "In this paper, we present a novel technique for reinforcement learning based spectrum sharing schemes. It shows how the potential action space of users is reduced by randomly partitioning the spectrum for different users. Cognitive agents are therefore able to finish their exploration stage faster than more basic reinforcement learning based schemes. Simulation results show how the communication system is more stable if all users reserve a subset of resources at the beginning and then learn to select the best channels in the reserved spectrum set. The system performance by using pre-partitioning is investigated and comparison with the traditional reinforcement learning based schemes is given.",

author = "Tao Jiang and David Grace and Mitchell, {Paul D}",

note = "Uploaded in accordance with the publisher's self-archiving policy.; International Communication Conference on Wireless Mobile and Computing (CCWMC 2009) ; Conference date: 07-12-2009 Through 09-12-2009",

year = "2009",

month = dec,

day = "1",

language = "English",

volume = "2009",

pages = "299 --302",

booktitle = "IET International Communication Conference on Wireless Mobile & Computing (CCWMC), Shanghai, China, December 2009",

publisher = "IEEE",

address = "United States",

edition = "562 CP",

}

Jiang, T , Grace, D & Mitchell, PD 2009, Improvement of pre-partitioning on reinforcement learning based spectrum sharing. in IET International Communication Conference on Wireless Mobile & Computing (CCWMC), Shanghai, China, December 2009. 562 CP edn, vol. 2009, IEEE, pp. 299 -302, International Communication Conference on Wireless Mobile and Computing (CCWMC 2009), Shanghai, China, 7/12/09.

Improvement of pre-partitioning on reinforcement learning based spectrum sharing. / Jiang, Tao ; Grace, David ; Mitchell, Paul D.
IET International Communication Conference on Wireless Mobile & Computing (CCWMC), Shanghai, China, December 2009. Vol. 2009 562 CP. ed. IEEE, 2009. p. 299 -302.

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

TY - GEN

T1 - Improvement of pre-partitioning on reinforcement learning based spectrum sharing

AU - Jiang, Tao

AU - Grace, David

AU - Mitchell, Paul D

N1 - Uploaded in accordance with the publisher's self-archiving policy.

PY - 2009/12/1

Y1 - 2009/12/1

N2 - In this paper, we present a novel technique for reinforcement learning based spectrum sharing schemes. It shows how the potential action space of users is reduced by randomly partitioning the spectrum for different users. Cognitive agents are therefore able to finish their exploration stage faster than more basic reinforcement learning based schemes. Simulation results show how the communication system is more stable if all users reserve a subset of resources at the beginning and then learn to select the best channels in the reserved spectrum set. The system performance by using pre-partitioning is investigated and comparison with the traditional reinforcement learning based schemes is given.

AB - In this paper, we present a novel technique for reinforcement learning based spectrum sharing schemes. It shows how the potential action space of users is reduced by randomly partitioning the spectrum for different users. Cognitive agents are therefore able to finish their exploration stage faster than more basic reinforcement learning based schemes. Simulation results show how the communication system is more stable if all users reserve a subset of resources at the beginning and then learn to select the best channels in the reserved spectrum set. The system performance by using pre-partitioning is investigated and comparison with the traditional reinforcement learning based schemes is given.

UR - http://www.scopus.com/inward/record.url?scp=78649810376&partnerID=8YFLogxK

M3 - Conference contribution

VL - 2009

SP - 299

EP - 302

BT - IET International Communication Conference on Wireless Mobile & Computing (CCWMC), Shanghai, China, December 2009

PB - IEEE

T2 - International Communication Conference on Wireless Mobile and Computing (CCWMC 2009)

Y2 - 7 December 2009 through 9 December 2009

ER -

Improvement of pre-partitioning on reinforcement learning based spectrum sharing

Abstract

Conference

Bibliographical note

Access to Document

Other files and links

Cite this