Cooperative retransmissions using Markov decision process with reinforcement learning

Ghasem Naddafzadeh Shirazi; Peng Yong Kong; Chen Khong Tham

doi:10.1109/PIMRC.2009.5450098

Cooperative retransmissions using Markov decision process with reinforcement learning

Ghasem Naddafzadeh Shirazi, Peng Yong Kong, Chen Khong Tham

Electrical Engineering

Institute for Infocomm Research

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

4 Scopus citations

Abstract

In cooperative retransmissions, nodes with better channel qualities help other nodes in retransmitting a failed packet to its intended destination. In this paper, we propose a cooperative retransmission scheme where each node makes local decision to cooperate or not to cooperate at what transmission power using a Markov decision process with reinforcement learning. With the reinforcement learning, the proposed scheme avoids solving an Markov decision process with a large number of states. Through simulations, we show that the proposed scheme is robust to collisions, is scalable with regard to the network size, and can provide significant cooperative diversity.

Original language	British English
Title of host publication	2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	652-656
Number of pages	5
ISBN (Print)	9781424451234
DOIs	https://doi.org/10.1109/PIMRC.2009.5450098
State	Published - 2009
Event	2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009 - Tokyo, Japan Duration: 13 Sep 2009 → 16 Sep 2009

Publication series

Name	IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC

Conference

Conference	2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009
Country/Territory	Japan
City	Tokyo
Period	13/09/09 → 16/09/09

Access to Document

10.1109/PIMRC.2009.5450098

Cite this

Shirazi, G. N., Kong, P. Y., & Tham, C. K. (2009). Cooperative retransmissions using Markov decision process with reinforcement learning. In 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009 (pp. 652-656). Article 5450098 (IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PIMRC.2009.5450098

Shirazi, Ghasem Naddafzadeh ; Kong, Peng Yong ; Tham, Chen Khong. / Cooperative retransmissions using Markov decision process with reinforcement learning. 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009. Institute of Electrical and Electronics Engineers Inc., 2009. pp. 652-656 (IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC).

@inproceedings{8e12ffde93ff4e23b61b4526afa6dd43,

title = "Cooperative retransmissions using Markov decision process with reinforcement learning",

abstract = "In cooperative retransmissions, nodes with better channel qualities help other nodes in retransmitting a failed packet to its intended destination. In this paper, we propose a cooperative retransmission scheme where each node makes local decision to cooperate or not to cooperate at what transmission power using a Markov decision process with reinforcement learning. With the reinforcement learning, the proposed scheme avoids solving an Markov decision process with a large number of states. Through simulations, we show that the proposed scheme is robust to collisions, is scalable with regard to the network size, and can provide significant cooperative diversity.",

author = "Shirazi, {Ghasem Naddafzadeh} and Kong, {Peng Yong} and Tham, {Chen Khong}",

year = "2009",

doi = "10.1109/PIMRC.2009.5450098",

language = "British English",

isbn = "9781424451234",

series = "IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC",

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

pages = "652--656",

booktitle = "2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009",

address = "United States",

note = "2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009 ; Conference date: 13-09-2009 Through 16-09-2009",

}

Shirazi, GN, Kong, PY & Tham, CK 2009, Cooperative retransmissions using Markov decision process with reinforcement learning. in 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009., 5450098, IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC, Institute of Electrical and Electronics Engineers Inc., pp. 652-656, 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009, Tokyo, Japan, 13/09/09. https://doi.org/10.1109/PIMRC.2009.5450098

Cooperative retransmissions using Markov decision process with reinforcement learning. / Shirazi, Ghasem Naddafzadeh; Kong, Peng Yong; Tham, Chen Khong.
2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009. Institute of Electrical and Electronics Engineers Inc., 2009. p. 652-656 5450098 (IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC).

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

TY - GEN

T1 - Cooperative retransmissions using Markov decision process with reinforcement learning

AU - Shirazi, Ghasem Naddafzadeh

AU - Kong, Peng Yong

AU - Tham, Chen Khong

PY - 2009

Y1 - 2009

N2 - In cooperative retransmissions, nodes with better channel qualities help other nodes in retransmitting a failed packet to its intended destination. In this paper, we propose a cooperative retransmission scheme where each node makes local decision to cooperate or not to cooperate at what transmission power using a Markov decision process with reinforcement learning. With the reinforcement learning, the proposed scheme avoids solving an Markov decision process with a large number of states. Through simulations, we show that the proposed scheme is robust to collisions, is scalable with regard to the network size, and can provide significant cooperative diversity.

AB - In cooperative retransmissions, nodes with better channel qualities help other nodes in retransmitting a failed packet to its intended destination. In this paper, we propose a cooperative retransmission scheme where each node makes local decision to cooperate or not to cooperate at what transmission power using a Markov decision process with reinforcement learning. With the reinforcement learning, the proposed scheme avoids solving an Markov decision process with a large number of states. Through simulations, we show that the proposed scheme is robust to collisions, is scalable with regard to the network size, and can provide significant cooperative diversity.

UR - http://www.scopus.com/inward/record.url?scp=77952855228&partnerID=8YFLogxK

U2 - 10.1109/PIMRC.2009.5450098

DO - 10.1109/PIMRC.2009.5450098

M3 - Conference contribution

AN - SCOPUS:77952855228

SN - 9781424451234

T3 - IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC

SP - 652

EP - 656

BT - 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009

Y2 - 13 September 2009 through 16 September 2009

ER -

Shirazi GN, Kong PY, Tham CK. Cooperative retransmissions using Markov decision process with reinforcement learning. In 2009 IEEE 20th Personal, Indoor and Mobile Radio Communications Symposium, PIMRC 2009. Institute of Electrical and Electronics Engineers Inc. 2009. p. 652-656. 5450098. (IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC). doi: 10.1109/PIMRC.2009.5450098

Cooperative retransmissions using Markov decision process with reinforcement learning

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this