Deep Learning for Primary Sector Prediction in FR2 New Radio Systems

Millimeter wave communications technology is an essential component of the <italic>new radio</italic> (NR) standard for standalone 5G networks, i.e., <italic>frequency range</italic> 2 (FR2) bands. This technology provides contiguous bandwidth at the detriment of high path lo...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Adel Aldalbahi, Mohammed A. Jasim, Farzad Shahabi, Asim Mazin, Nazli Siasi, Diogo Oliveira
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Beamforming convolutional neural networks deep learning initial access long short-term memory millimeter wave communications Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/72e04e7ff8304e8ab32aab7f31e6db97
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:72e04e7ff8304e8ab32aab7f31e6db97
record_format	dspace
spelling	oai:doaj.org-article:72e04e7ff8304e8ab32aab7f31e6db972021-12-03T00:00:46ZDeep Learning for Primary Sector Prediction in FR2 New Radio Systems2169-353610.1109/ACCESS.2021.3128432https://doaj.org/article/72e04e7ff8304e8ab32aab7f31e6db972021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9615238/https://doaj.org/toc/2169-3536Millimeter wave communications technology is an essential component of the <italic>new radio</italic> (NR) standard for standalone 5G networks, i.e., <italic>frequency range</italic> 2 (FR2) bands. This technology provides contiguous bandwidth at the detriment of high path loss and blockage sensitivity. Hence beamforming architectures are leveraged to compensate for the channel impairments. However, beamforming introduces significant challenges in terms of initial beam access and beam adaptation requirements. Namely, the <italic>base station</italic> (BS) and <italic>mobile station</italic> (MS) are compelled to search the entire spatial directions to specify the pointing directions (optimum beamforming and combining vectors) that yield in the strongest impinged signal levels. This search process results in a high computational complexity, extended delay periods, high power consumption and energy inefficiency. Hence this paper proposes a novel sector (beam) prediction scheme that leverages the synergistic combination of <italic>convolutional neural networks</italic> (CNN) and <italic>long short-term memory</italic> (LSTM). The goal is to propose ultra-low access times for FR2-bsaed 5G networks, thus enhancing mmWave bands to operate independently as a standalone network without reliance on <italic>Frequency Range</italic> 1 (FR1) bands, e.g., dual-connectivity. The proposed scheme here predicts the primary sector with the highest popularity class at the BS, which is affiliated with the mostly used beamforming vector. This retrieves information about the sector locations with the highest MS traffic, thereby the BS can eliminate the spatial search over locations of scarce MS densities. Consequently, this process reduces the beam scanning search at the BS, while performing conventional search at the MS. The proposed scheme yields in reduced complexity and access times as compared to prominent existing methods.Adel AldalbahiMohammed A. JasimFarzad ShahabiAsim MazinNazli SiasiDiogo OliveiraIEEEarticleBeamformingconvolutional neural networksdeep learninginitial accesslong short-term memorymillimeter wave communicationsElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 157522-157539 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Beamforming convolutional neural networks deep learning initial access long short-term memory millimeter wave communications Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Beamforming convolutional neural networks deep learning initial access long short-term memory millimeter wave communications Electrical engineering. Electronics. Nuclear engineering TK1-9971 Adel Aldalbahi Mohammed A. Jasim Farzad Shahabi Asim Mazin Nazli Siasi Diogo Oliveira Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
description	Millimeter wave communications technology is an essential component of the <italic>new radio</italic> (NR) standard for standalone 5G networks, i.e., <italic>frequency range</italic> 2 (FR2) bands. This technology provides contiguous bandwidth at the detriment of high path loss and blockage sensitivity. Hence beamforming architectures are leveraged to compensate for the channel impairments. However, beamforming introduces significant challenges in terms of initial beam access and beam adaptation requirements. Namely, the <italic>base station</italic> (BS) and <italic>mobile station</italic> (MS) are compelled to search the entire spatial directions to specify the pointing directions (optimum beamforming and combining vectors) that yield in the strongest impinged signal levels. This search process results in a high computational complexity, extended delay periods, high power consumption and energy inefficiency. Hence this paper proposes a novel sector (beam) prediction scheme that leverages the synergistic combination of <italic>convolutional neural networks</italic> (CNN) and <italic>long short-term memory</italic> (LSTM). The goal is to propose ultra-low access times for FR2-bsaed 5G networks, thus enhancing mmWave bands to operate independently as a standalone network without reliance on <italic>Frequency Range</italic> 1 (FR1) bands, e.g., dual-connectivity. The proposed scheme here predicts the primary sector with the highest popularity class at the BS, which is affiliated with the mostly used beamforming vector. This retrieves information about the sector locations with the highest MS traffic, thereby the BS can eliminate the spatial search over locations of scarce MS densities. Consequently, this process reduces the beam scanning search at the BS, while performing conventional search at the MS. The proposed scheme yields in reduced complexity and access times as compared to prominent existing methods.
format	article
author	Adel Aldalbahi Mohammed A. Jasim Farzad Shahabi Asim Mazin Nazli Siasi Diogo Oliveira
author_facet	Adel Aldalbahi Mohammed A. Jasim Farzad Shahabi Asim Mazin Nazli Siasi Diogo Oliveira
author_sort	Adel Aldalbahi
title	Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
title_short	Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
title_full	Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
title_fullStr	Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
title_full_unstemmed	Deep Learning for Primary Sector Prediction in FR2 New Radio Systems
title_sort	deep learning for primary sector prediction in fr2 new radio systems
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/72e04e7ff8304e8ab32aab7f31e6db97
work_keys_str_mv	AT adelaldalbahi deeplearningforprimarysectorpredictioninfr2newradiosystems AT mohammedajasim deeplearningforprimarysectorpredictioninfr2newradiosystems AT farzadshahabi deeplearningforprimarysectorpredictioninfr2newradiosystems AT asimmazin deeplearningforprimarysectorpredictioninfr2newradiosystems AT nazlisiasi deeplearningforprimarysectorpredictioninfr2newradiosystems AT diogooliveira deeplearningforprimarysectorpredictioninfr2newradiosystems
_version_	1718373994057433088

Deep Learning for Primary Sector Prediction in FR2 New Radio Systems

Ejemplares similares