Utilize este identificador para referenciar este registo:
http://hdl.handle.net/10071/28846
Autoria: | Farkhari, H. Viana, J. Sebastião, P. Bernardo, L. Kahvazadeh, S. Dinis, R. |
Data: | 2023 |
Título próprio: | Accurate and reliable methods for 5G UAV jamming identification with calibrated uncertainty |
Título e volume do livro: | RCIS: The 17th International Conference on Research Challenges in Information Science |
Título do evento: | 17th International Conference on Research Challenges in Information Science |
Referência bibliográfica: | Farkhari, H., Viana, J., Sebastião, P., Bernardo, L., Kahvazadeh, S., & Dinis, R. (2023). Accurate and reliable methods for 5G UAV jamming identification with calibrated uncertainty. In RCIS: The 17th International Conference on Research Challenges in Information Science. http://hdl.handle.net/10071/28846 |
ISSN: | 1613-0073 |
Palavras-chave: | Unmanned Aerial Vehicle Deep neural networks Calibration Uncertainty Reliability Jamming identification 5G 6G |
Resumo: | This research highlights the negative impact of ignoring uncertainty on DNN decision-making and Reliability. Proposed combined preprocessing and post-processing methods enhance DNN accuracy and Reliability in time-series binary classification for 5G UAV security dataset, employing ML algorithms and confidence values. Several metrics are used to evaluate the proposed hybrid algorithms. The study emphasizes the XGB classifier's unreliability and suggests the proposed methods' potential superiority over the DNN softmax layer. Furthermore, improved uncertainty calibration based on the Reliability Score metric minimizes the difference between Mean Confidence and Accuracy, enhancing accuracy and Reliability. |
Arbitragem científica: | yes |
Acesso: | Acesso Aberto |
Aparece nas coleções: | IT-CRI - Comunicações a conferências internacionais |
Ficheiros deste registo:
Ficheiro | Tamanho | Formato | |
---|---|---|---|
conferenceobject_96444.pdf | 678,34 kB | Adobe PDF | Ver/Abrir |
Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.