Otimização de características para classificação automática de modulação de sinais afetados pelos canais Awgn e Rayleigh.

• Autor(a) principal: Tavares, Samuel Alves
• Data de Publicação: 2023
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/37275
• topic: Aprendizado de máquina; Machine learning; Classificação automática de modulações; Automatic modulation classification; Inteligência artificial; Artificial intelligence; Canal de rayleigh; Rayleigh channel; Correlação; Correlation; Seleção de características; Feature selection; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::TELECOMUNICACOES

In the last years, lot of new mobile network technologies have emerged. So, Telecommunications networks need to be more autonomous, identifying all user parameters, including the type of modulation. The objective of this work is to generate a database with the main types of digital modulation and build a system that automatically classifies digital modulations of a signal that passed through the AWGN and Rayleigh channels, select and optimize the best characteristics of the signal so that the classification be done more quickly and efficiently. The selection of features using the correlation and RFE methods, proved to be very promising, decreasing by about 80% of the initially proposed features, in addition to having reached a high accuracy of correct classification.