| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 5.08 MB | Adobe PDF |
Authors
Abstract(s)
Durante as últimas décadas, as redes móveis apresentaram um crescimento constante, acompanhado
do surgimento de novas tecnologias. Atualmente, esse crescimento tende a continuar, principalmente
devido à introdução da próxima 5a geração (5G) de comunicações móveis. Muitas vezes o consumo energético das redes móveis é uma questão frequentemente negligenciada em relação à de conectividade, avaliação do ritmo de transmissão, cobertura e qualidade de serviço. À medida que as redes crescem, o desafio surge em suportar o tráfego, diminuindo ou mantendo o consumo de energia. Além das responsabilidades corporativas na proteção ambiental, os operadores estão também conscientes dos seus gastos com o consumo energético, correspondendo entre 18% e 32% das despesas de Operational Expenditure (OPEX). Nesta perspetiva, a redução de custos e sustentabilidade ambiental podem ser vistos como objetivos convergentes. A avaliação detalhada da quantidade de energia consumida pelos equipamentos da rede de acesso é um passo essencial para determinar soluções otimizadas. O trabalho desenvolvido nesta dissertação visa propor modelos de consumo energético para os equipamentos que constituem uma estação base,nomeadamente as Remote Radio Units (RRU’s) e Baseband Units (BBU’s), para as tecnologias atualmente massificadas no mercado de telecomunicações, 2G, 3G e 4G, e para os dois tipos de fabricantes Ericsson e Huawei. A modelação dos dados baseia-se no método de inferência de regressão linear de efeitos mistos. De modo a avaliar os resultados, foram utilizadas métricas de Akaike Information Criterion (AIC), Mean Absolute Percentage Error (MAPE) e Root Mean Square Error (RMSE). Os resultados dos modelos atingiram um erro percentual máximo de 3.7% para os equipamento de rádio e de 8.49% para os equipamentos de banda base. Dada a evolução das gerações de comunicações móveis e o possível switch-off de tecnologias, apresenta-se uma estimativa da redução do consumo de energia ao desligar as tecnologias 2G e/ou 3G.
During the last decades, the mobile networks presented a constant growth, with the emergence of new technologies. Currently, this growth is likely to continue, mainly due to the introduction of the next fifth generation (5G) of mobile communications. Often the energy consumption of mobile networks is often neglected in favour of connectivity, high data throughput, coverage, and quality of service. As networks grow, the challenge lies in supporting traffic and reducing or maintaining power consumption. In addition to corporate responsibilities for environmental protection, operators are also aware of their energy consumption expenditures, accounting for 18% to 32% of Operational Expenditure (OPEX) expenditures. In this perspective, cost reduction and environmental sustainability can be seen as converging objectives. Accurate assessment of the amount of power consumed by radio access network equipment is an essential step in determining optimized solutions. The work developed in this dissertation aims to propose models of energy consumption for the equipment constituting a base station, namely Remote Radio Units (RRU’s) and Baseband Units (BBU’s), for technologies currently massified in the telecommunications market, 2G, 3G and 4G, and for the two vendors, Ericsson and Huawei. The data modeling is based on the mixed-effects linear regression inference method. In order to evaluate the results, we used metrics of Akaike Information Criterion (AIC), Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The model’s results reached a maximum percentage error of 3.7 % for the radio equipment and 8.49% for the baseband equipment. Given the evolution of mobile communications generations and the possible switch-off of technologies, an estimate of the energy consumption reduction is presented by turn off 2G and/or 3G technologies.
During the last decades, the mobile networks presented a constant growth, with the emergence of new technologies. Currently, this growth is likely to continue, mainly due to the introduction of the next fifth generation (5G) of mobile communications. Often the energy consumption of mobile networks is often neglected in favour of connectivity, high data throughput, coverage, and quality of service. As networks grow, the challenge lies in supporting traffic and reducing or maintaining power consumption. In addition to corporate responsibilities for environmental protection, operators are also aware of their energy consumption expenditures, accounting for 18% to 32% of Operational Expenditure (OPEX) expenditures. In this perspective, cost reduction and environmental sustainability can be seen as converging objectives. Accurate assessment of the amount of power consumed by radio access network equipment is an essential step in determining optimized solutions. The work developed in this dissertation aims to propose models of energy consumption for the equipment constituting a base station, namely Remote Radio Units (RRU’s) and Baseband Units (BBU’s), for technologies currently massified in the telecommunications market, 2G, 3G and 4G, and for the two vendors, Ericsson and Huawei. The data modeling is based on the mixed-effects linear regression inference method. In order to evaluate the results, we used metrics of Akaike Information Criterion (AIC), Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The model’s results reached a maximum percentage error of 3.7 % for the radio equipment and 8.49% for the baseband equipment. Given the evolution of mobile communications generations and the possible switch-off of technologies, an estimate of the energy consumption reduction is presented by turn off 2G and/or 3G technologies.
Description
Dissertação para obtenção do grau de Mestre em Engenharia Eletrónica e Telecomunicações na área de especialização em Telecomunicações
Keywords
Redes móveis Rede de Acesso Rádio Regressão linear de efeitos mistos Mobile networks Radio access network Mixed effects linear regression
Citation
SARAIVA, Thaína de Castro Saraiva - Desenvolvimento de um modelo de consumo energético multi-tecnologia e multi-vendor para nós de acesso rádio em redes móveis. Lisboa: Instituto Superior de Engenharia de Lisboa, 2019. Dissertação de Mestrado