A Markov Decision Process-based Collision Avoidance in IoT Applications

Fatima Djiroun, Miloud Bagaa, Tarik Taleb

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

130 Downloads (Pure)

Abstract

IoT covers various scales and types of wireless networks. The first constraint to be respected, for an efficient application, is to reduce as much as possible the amount of energy consumption. The idle listening process in existing Medium Access Control (MAC) protocols is a very energy consuming task. Recently, a new emerging technology based on a low power wake-up radio has shown real benefits by completely eliminating the problem of idle listening. Thanks to the use of this technology, an IoT device keeps its main radio in deep sleep until a wake-up message is received by the wake-up radio that consumes less energy. However, collision can occur among wake-up messages (i.e., wake-up plane). Collision in the wake-up plane, if not handled efficiently, leads to collision at the data plane which is more complicated. In this paper, we address this issue by modeling the wake-up decision using a Markov Decision Process (MDP). The goal is to formulate a decision policy that determines whether to send a wake-up message in the actual time slot or to report it, taking into account the time factor. Experiments have been conducted to determine the decision policies. Results of the proposed approach have been compared against those of RFIDImpulse, a CSMA\CA-based wake-up MAC protocol. The obtained results show the efficiency of the proposed approach.
Original languageEnglish
Title of host publication 2017 IEEE Global Communications Conference, GLOBECOM 2017
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-5090-5019-2
DOIs
Publication statusPublished - Dec 2017
MoE publication typeA4 Article in a conference publication
EventIEEE Global Communications Conference - Singapore, Singapore
Duration: 4 Dec 20178 Dec 2017

Publication series

NameIEEE Global Communications Conference
ISSN (Print)2334-0983

Conference

ConferenceIEEE Global Communications Conference
Abbreviated titleGLOBECOM
CountrySingapore
CitySingapore
Period04/12/201708/12/2017

Fingerprint

Dive into the research topics of 'A Markov Decision Process-based Collision Avoidance in IoT Applications'. Together they form a unique fingerprint.

Cite this