Reference : Short-Packet Communications in Multi-Hop WPINs: Performance Analysis and Deep Learnin...
Scientific congresses, symposiums and conference proceedings : Unpublished conference
Engineering, computing & technology : Electrical & electronics engineering
Short-Packet Communications in Multi-Hop WPINs: Performance Analysis and Deep Learning Design
Nguyen, Toan-Van []
Nguyen, van Dinh mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SigCom >]
Costa, Daniel Benevides da []
An, Beongku []
IEEE Global Communications Conference (GLOBECOM)
7-12-2021 to 12-12-2021
[en] Block error rate ; deep neural network ; energy harvesting
[en] In this paper, we study short-packet communications (SPCs) in multi-hop wireless-powered Internet-of-Things networks (WPINs), where IoT devices transmit short packets to multiple destination nodes by harvesting energy from multiple power beacons. To improve system block error rate (BLER) and throughput, we propose a best relay-best user (bR-bU) selection scheme with an accumulated energy harvesting mechanism. Closed-form expressions for the BLER and throughput of the proposed scheme over Rayleigh fading channels are derived and the respective asymptotic analysis is also carried out. To support real-time settings, we design a deep neural network
(DNN) framework to predict the system throughput under different channel settings. Numerical results demonstrate that the proposed bR-bU selection scheme outperforms several baseline ones in terms of the BLER and throughput, showing to be an efficient strategy for multi-hop SPCs. The resulting DNN can estimate accurately the throughput with low execution time. The effects of message size on reliability and latency are also evaluated and discussed.

File(s) associated to this reference

Fulltext file(s):

Open access
SPC_Multihop_GC21.pdfAuthor postprint1.12 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.