Predicting Device-to-Device Channels From Cellular Channel Measurements: A Learning Approach

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F20%3A00344305" target="_blank" >RIV/68407700:21230/20:00344305 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/TWC.2020.3008303" target="_blank" >https://doi.org/10.1109/TWC.2020.3008303</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TWC.2020.3008303" target="_blank" >10.1109/TWC.2020.3008303</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Predicting Device-to-Device Channels From Cellular Channel Measurements: A Learning Approach

Popis výsledku v původním jazyce

Device-to-device (D2D) communication, which ena- bles a direct connection between users while bypassing the cellular channels to base stations (BSs), is a promising way to offload the traffic from conventional cellular networks. In D2D communication, optimizing the resource allocation requires the knowledge of D2D channel gains. However, such knowledge is hard to obtain at reasonable signaling costs. In this paper, we show this problem can be circumvented by tapping into the information provided by the estimated cellular channels between the users and surrounding BSs as these channels are estimated anyway for a normal operation of the network. While the cellular and D2D channel gains exhibit independent fast fading behavior, we show that average gains of the cellular and D2D channels share a non-explicit relation, which is rooted into the network topology, terrain, and buildings setup. We propose a deep learning approach to predict the D2D channel gains from seemingly independent cellular channels. Our results show a high degree of convergence between the true and predicted D2D channel gains. Moreover, we demonstrate the robustness of the proposed scheme against environment changes and inaccuracies during the offline training. The predicted gains allow to reach a near-optimal capacity in many radio resource management algorithms.

Název v anglickém jazyce

Predicting Device-to-Device Channels From Cellular Channel Measurements: A Learning Approach

Popis výsledku anglicky

Device-to-device (D2D) communication, which ena- bles a direct connection between users while bypassing the cellular channels to base stations (BSs), is a promising way to offload the traffic from conventional cellular networks. In D2D communication, optimizing the resource allocation requires the knowledge of D2D channel gains. However, such knowledge is hard to obtain at reasonable signaling costs. In this paper, we show this problem can be circumvented by tapping into the information provided by the estimated cellular channels between the users and surrounding BSs as these channels are estimated anyway for a normal operation of the network. While the cellular and D2D channel gains exhibit independent fast fading behavior, we show that average gains of the cellular and D2D channels share a non-explicit relation, which is rooted into the network topology, terrain, and buildings setup. We propose a deep learning approach to predict the D2D channel gains from seemingly independent cellular channels. Our results show a high degree of convergence between the true and predicted D2D channel gains. Moreover, we demonstrate the robustness of the proposed scheme against environment changes and inaccuracies during the offline training. The predicted gains allow to reach a near-optimal capacity in many radio resource management algorithms.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20203 - Telecommunications

Projekt

<a href="/cs/project/LTT18007" target="_blank" >LTT18007: Spolupráce s mezinárodním výzkumným centrem v oblasti komunikačních systémů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

IEEE Transactions on Wireless Communications

ISSN

1536-1276

e-ISSN

1558-2248

Svazek periodika

19

Číslo periodika v rámci svazku

11

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

15

Strana od-do

7124-7138

Kód UT WoS článku

000589218700009

EID výsledku v databázi Scopus

2-s2.0-85096243504