Automatic modulation classification for software defined radios : a deep learning approach

dc.contributor.advisorÁlvarez Meza, Andrés Marino
dc.contributor.advisorCollazos Huertas, Diego Fabián
dc.contributor.authorMosquera Trujillo, Carlos Enrique
dc.contributor.cvlacMosquera Trujillo, Carlos Enrique [0001997126]
dc.contributor.orcidMosquera Trujillo, Carlos Enrique [0009000857193516]
dc.contributor.orcidMosquera Trujillo, Carlos Enrique [0000000303089576]
dc.contributor.orcidMosquera Trujillo, Carlos Enrique [0000000204343444]
dc.contributor.researchgroupGrupo de Control y Procesamiento Digital de Señales
dc.date.accessioned2026-02-25T22:04:01Z
dc.date.available2026-02-25T22:04:01Z
dc.date.issued2025
dc.descriptiongraficas, tablasspa
dc.description.abstractThe exponential growth of wireless communications, driven by technologies such as 5G, 6G, and the Internet of Things (IoT), has led to unprecedented radio spectrum congestion. In this environment, Automatic Modulation Classification (AMC) has emerged as a critical technology for intelligent and adaptive spectrum management. However, the practical implementation of Deep Learning (DL) models, which dominate the state of the art, is hindered by three fundamental challenges: their high computational complexity, their lack of robustness to dynamic channel conditions, and their "black-box" nature, which limits their reliability in critical applications. This document presents a comprehensive solution to these challenges: an AMC pipeline based on a novel lightweight neural network architecture, named ComplexMobileNet-AMC. To address robustness, the pipeline incorporates an adaptive pre-processing stage based on Random Fourier Features (RFF) for signal reconstruction and a modulation-aware, stochastic data augmentation strategy. For efficiency, the model architecture combines complex-valued convolutions, depthwise separable blocks, and self-attention mechanisms to minimize computational load without sacrificing accuracy. Finally, for interpretability, an "interpretable-by-design" approach is adopted, using the model’s internal attention weights to generate physics-grounded explanations in the time, phase, and frequency domains. Experimental results demonstrate that the proposed model is not only competitive with stateof-the-art architectures in terms of accuracy but significantly surpasses them in efficiency. In deployments on edge hardware such as Software-Defined Radios (SDR) and devices like the NVIDIA Jetson Nano, the model is between 2 and 6 times faster than reference models. The interpretability analysis confirms that the model learns to identify physically relevant features, such as the symbol rate of digital modulations. The entire system is validated in a functional over-the-air (OTA) transmission and reception prototype with a real-time monitoring GUI. This holistic approach positions the proposed solution as a robust, efficient, and transparent methodology for the practical deployment of AMC in next-generation communication systems.eng
dc.description.abstractEl crecimiento exponencial de las comunicaciones inalámbricas, impulsado por tecnologías como 5G, 6G y el Internet of Things (IoT), ha generado una congestión sin precedentes en el espectro radioeléctrico. En este entorno, la Automatic Modulation Classification (AMC) se ha convertido en una tecnología crítica para la gestión inteligente y adaptativa del espectro. Sin embargo, la implementación práctica de los modelos de Deep Learning (DL), que dominan el estado del arte, se ve obstaculizada por tres desafíos fundamentales: su elevada complejidad computacional, su falta de robustez ante las condiciones dinámicas del canal y su naturaleza de "caja negra", que limita su fiabilidad en aplicaciones críticas. Este documento presenta una solución integral a estos desafíos: un pipeline de AMC basado en una novedosa arquitectura de red neuronal ligera, denominada ComplexMobileNet-AMC. Para abordar la robustez, el pipeline incorpora una etapa de pre-procesamiento adaptativo basado en Random Fourier Features (RFF) para la reconstrucción de la señal y una estrategia de aumento de datos estocástico y consciente de la modulación. Para la eficiencia, la arquitectura del modelo combina convoluciones de valores complejos, bloques separables en profundidad y mecanismos de auto-atención para minimizar la carga computacional sin sacrificar la precisión. Finalmente, para la interpretabilidad, se adopta un enfoque "interpretable por diseño", utilizando los pesos de atención internos del modelo para generar explicaciones físicamente fundamentadas en los dominios del tiempo, la fase y la frecuencia. Los resultados experimentales demuestran que el modelo propuesto no solo es competitivo con arquitecturas de última generación en términos de precisión, sino que las supera significativamente en eficiencia. En despliegues sobre hardware de borde como Software-Defined Radios (SDR) y dispositivos como la NVIDIA Jetson Nano, el modelo es entre 2 y 6 veces más rápido que los modelos de referencia. El análisis de interpretabilidad confirma que el modelo aprende a identificar características físicamente relevantes, como la tasa de símbolo de las modulaciones digitales. El sistema completo se valida en un prototipo funcional de transmisión y recepción over-the-air (OTA) con una GUI de monitoreo en tiempo real. Este enfoque holístico posiciona la solución propuesta como una metodología robusta, eficiente y transparente para el despliegue práctico de AMC en sistemas de comunicación de próxima generación.spa
dc.description.curricularareaEléctrica, Electrónica, Automatización Y Telecomunicaciones.Sede Manizales
dc.description.degreelevelMaestría
dc.description.degreenameMagíster en Ingeniería - Automatización Industrial
dc.format.extentxv, 153 páginas
dc.format.mimetypeapplication/pdf
dc.identifier.instnameUniversidad Nacional de Colombiaspa
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombiaspa
dc.identifier.repourlhttps://repositorio.unal.edu.co/spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/89685
dc.language.isoeng
dc.publisherUniversidad Nacional de Colombia
dc.publisher.branchUniversidad Nacional de Colombia - Sede Manizales
dc.publisher.facultyFacultad de Ingeniería y Arquitectura
dc.publisher.placeManizales, Colombia
dc.publisher.programManizales - Ingeniería y Arquitectura - Maestría en Ingeniería - Automatización Industrial
dc.relation.indexedAgrosavia
dc.relation.indexedBireme
dc.relation.indexedRedCol
dc.relation.indexedLaReferencia
dc.relation.indexedAgrovoc
dc.relation.referencesAbdel-Aty, A.; Abdel-Rahman, M. & Elsayed, H.: , 2024; Alwnn: Automatic modulation classification via adaptive lightweight wavelet neural network; en 2024 IEEE Wireless Communications and Networking Conference (WCNC); doi:10.1109/WCNC60147.2024.10423498
dc.relation.referencesAkyildiz, I. F.; Lee, W.-Y.; Vuran, M. C. & Mohanty, S.: , 2022; Cognitive radio networks: A survey; Physical Communication; 50: 101512
dc.relation.referencesAl-Fuqaha, A.; Guizani, M.; Mohammadi, M.; Aledhari, M. & Ayyash, M.: , 2021; A comprehensive survey on the internet of things: Applications, challenges, and security; IEEE Communications Surveys & Tutorials; 17 (4): 2347–2378
dc.relation.referencesAli, E. M.; Salama, G. M.; Khalil, A. A. & Ezz-Eldin, M.: , 2025; Automatic modulation classification for enhanced cognitive radio for iot systems based on deep learning; Journal of Advanced Engineering Trends; 44 (1): 106–113
dc.relation.referencesAlshamrani, F. & Shamsi, J.: , 2023; Automatic modulation classification for low-power iot applications: A lightweight deep learning approach; Telecom; 5 (2): 459–474
dc.relation.referencesAnsari, S.; Al-Fuqaha, A. et al.: , 2025a; A novel lstm architecture for automatic modulation recognition: Comparative analysis with conventional machine learning and rnn-based approaches; IEEE Access; 13: 72527–72540; doi: 10.1109/ACCESS.2025.3564032
dc.relation.referencesAnsari, S.; Mahmoud, S.; Majzoub, S.; Almajali, E.; Jarndal, A. & Bonny, T.: , 2025b; A novel lstm architecture for automatic modulation recognition: Comparative analysis with conventional machine learning and rnn-based approaches; IEEE Access; 13: 72526–72537; doi:10.1109/ACCESS.2025.3564032
dc.relation.referencesBaloch, U.; Khan, A. & Iqbal, M.: , 2024; Deep learning-based automatic modulation classification: Recent advances and challenges; International Journal of Communication Systems
dc.relation.referencesCho, K.; Van Merri"enboer, B.; Gulcehre, C.; Bahdanau, D.; Bougares, F.; Schwenk, H. & Bengio, Y.: , 2014; Learning phrase representations using rnn encoder-decoder for statistical machine translation; arXiv preprint arXiv:1406.1078
dc.relation.referencesDai, W. & Ji, C.: , 2024; Modulation recognition system of electromagnetic interference signal based on sdr; Telecom; 5 (3): 928–940; doi:10.3390/ telecom5030046
dc.relation.referencesDonoho, D. L. & Johnstone, I. M.: , 1995; Adapting to unknown smoothness via wavelet shrinkage; Journal of the american statistical association; 90 (432): 1200–1224
dc.relation.referencesEricsson: , 2025; Ericsson mobility report; Informe técnico; Ericsson; publicly available report providing data and forecasts on mobile traffic, subscriptions, and market trends. Page 8
dc.relation.referencesGautama, T.; Mandic, D. P. & Van Hulle, M. M.: , 2004; A benchmark of standard and kernel based methods for time series prediction; Neurocomputing; 56: 357–375
dc.relation.referencesGCPDS and Agencia Nacional del Espectro: , 2024; Cost-effective and scalable prototype for radioelectric spectrum monitoring in colombia using software-defined radio and deep learning; Research and Development Project; hermes Code: 69909, Agreement No. 163-2024
dc.relation.referencesGuo, L.; Wang, Y.; Lin, Y.; Zhao, H. & Gui, G.: , 2024a; Ultralight convolutional neural network for automatic modulation classification in internet of unmanned aerial vehicles; IEEE Internet of Things Journal; 11 (11): 20831–20839; URL https://ieeexplore.ieee.org/document/10460305
dc.relation.referencesGuo, L.; Wang, Y.; Liu, Y.; Lin, Y.; Zhao, H. & Gui, G.: , 2024b; Ultralight convolutional neural network for automatic modulation classification in internet of unmanned aerial vehicles; IEEE Internet of Things Journal; 11 (11): 20831–20839; doi:10.1109/JIOT.2024.3373497
dc.relation.referencesHameed, S.; Alzahrani, M. & Chaudhry, S.: , 2024; Adaptive and interpretable neural network for signal modulation classification; Applied Sciences; 14 (6): 6302; doi:10.3390/app14066302
dc.relation.referencesHamidi-Rad, S. & Jain, S.: , 2021; Mcformer: A transformer based deep neural network for automatic modulation classification; en 2021 IEEE Global Communications Conference (GLOBECOM); IEEE; págs. 1–6
dc.relation.referencesHan, J.; Yu, Z. & Yang, J.: , 2022; Multimodal attention-based deep learning for automatic modulation classification; Frontiers in Energy Research; 10: 1041862
dc.relation.referencesHassan, M. & Rahman, R. M.: , 2022; Automatic modulation classification using convolutional neural network with batch normalization: A novel approach; en 2022 IEEE 11th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON); IEEE; págs. 0392–0396; doi: 10.1109/IEMCON55942.2022.9998193
dc.relation.referencesHaykin, S.: , 2005; Cognitive radio: brain-empowered wireless communications; IEEE journal on selected areas in communications; 23 (2): 201–220
dc.relation.referencesHicks, B.; Biswas, S.; Ball, J. & Gurbuz, A.: , 2025; Automatic classification of radar and communication waveforms through interpretable deep learning; pág. 28; doi:10.1117/12.3054111
dc.relation.referencesHoward, A.; Sandler, M.; Chu, G.; Chen, L.-C.; Chen, B.; Tan, M.; Wang, W.; Zhu, Y.; Pang, R.; Vasudevan, V. et al.: , 2019; Searching for mobilenetv3; en Proceedings of the IEEE/CVF international conference on computer vision; págs. 1314–1324
dc.relation.referencesHu, J.; Shen, L. & Sun, G.: , 2018; Squeeze-and-excitation networks; en Proceedings of the IEEE conference on computer vision and pattern recognition; págs. 7132–7141
dc.relation.referencesHuynh-The, T.; Hua, C.-H.; Pham, Q.-V. & Kim, D.-S.: , 2020; Mcnet: An efficient cnn architecture for robust automatic modulation classification; IEEE Communications Letters; 24 (4): 811–815; doi:10.1109/LCOMM.2020.2968030; URL https://ieeexplore.ieee.org/abstract/document/8963964
dc.relation.referencesJi, Z.; Wang, S.; Yang, K.; Zhang, Q. & Ye, P.: , 2024; Transfer learning guided noise reduction for automatic modulation classification; arXiv preprint arXiv:2411.08376 ; URL https://arxiv.org/abs/2411.08376
dc.relation.referencesKarim, F. et al.: , 2017; Automatic modulation classification using recurrent neural networks; en IEEE; URL https://ieeexplore.ieee.org/abstract/ document/8322633
dc.relation.referencesKhan, M. U. & Raza, S. Z.: , 2023; Automatic modulation classification using cnn and support vector machine; Journal of Applied Engineering and Technology; 44 (1): 106–113
dc.relation.referencesKocijan, J.; Girard, A.; Banko, B. & Murray-Smith, R.: , 2004; Gaussian processes for system identification; IFAC Proceedings Volumes; 37 (8): 689–694
dc.relation.referencesLazarus, A.; Mairal, J.; Bach, F. & Vert, J.-P.: , 2018; Skpr: A kernel-based method for sparse online signal recovery; en 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); IEEE; págs. 4164– 4168
dc.relation.referencesLe, Q. N.; Huynh, T. Q.; Ta, H. Q.; Vo Tan, P. & Nguyen, L. L.: , 2024; Modified receiver architecture in software-defined radio for real-time modulation classification; EURASIP Journal on Advances in Signal Processing ; 2024 (77): 1–16; doi:10.1186/s13634-024-01176-6
dc.relation.referencesLi, H.; Zhang, P. & Wang, X.: , 2023a; Amcnet: Efficient automatic modulation classification network for low-complexity devices; Sensors; 23 (4): 1802; doi:10.3390/s23041802
dc.relation.referencesLi, J. et al.: , 2019; Automatic modulation classification using parallel fusion of convolutional neural networks; URL https://scholar.google.com/scholar?hl= en&as_sdt=0%2C5&q=Automatic+Modulation+Classification+Using+Parallel+ Fusion+of+Convolutional+Neural+Networks&btnG=
dc.relation.referencesLi, Y.; Wang, X. & Zhao, Q.: , 2023b; Automatic modulation classification using deep residual neural network with masked modeling; Drones; 7 (6): 390
dc.relation.referencesLi, Y.; Yuan, L.; Zhou, F.; Wu, Q.; Al-Dhahir, N. & Wong, K.-K.: , 2024a; Kgamc: A novel knowledge graph driven automatic modulation classification scheme; arXiv preprint arXiv:2402.19188v2 ; URL https://arxiv.org/ abs/2402.19188
dc.relation.referencesLi, Y.; Zhao, X. & Chen, H.: , 2024b; A survey on modulation classification using deep learning techniques; Physical Communication; 62: 101443
dc.relation.referencesLiu, F.; Huang, X.; Chen, Y. & Suykens, J. A.: , 2021; Random features for kernel approximation: A survey on algorithms, theory, and beyond; IEEE Transactions on Pattern Analysis and Machine Intelligence ; 44 (10): 7128–7148
dc.relation.referencesLiu, Q. et al.: , 2020; Robust and fast automatic modulation classification with cnn under multipath fading channels; IEEE; URL https://ieeexplore.ieee. org/abstract/document/9128408
dc.relation.referencesLiu, X.; Song, Y.; Zhu, J.; Shu, F. & Qian, Y.: , 2024a; An efficient deep learning model for automatic modulation classification; Radioengineering; 33 (4): 713–720; doi:10.13164/re.2024.0713
dc.relation.referencesLiu, Z.; Wang, H. & Chen, S.: , 2024b; Ultralight convolutional neural network for automatic modulation classification in internet of unmanned aerial vehicles; IEEE Transactions on Aerospace and Electronic Systems; doi:10.1109/TAES. 2024.3382068
dc.relation.referencesMitola III, J.: , 1999; Software radio architecture: A mathematical perspective; IEEE Journal on Selected Areas in Communications; 17 (4): 514–538
dc.relation.referencesMondino-Llermanos, Y. R. & CorralBriones, G.: , 2024; Automatic modulation classification for low-power iot applications; IEEE Latin America Transactions; 22 (3): 204–212; doi:10.1109/TLA.2024.3163537
dc.relation.referencesMosquera-Trujillo, C.; Lugo-Rojas, J.; CollazosHuertas, D.; Álvarez Meza, A. & Castellanos-Dominguez, G.: , 2024a; Lightweight and interpretable dl model using convolutional rff for amc; Submitted for publication
dc.relation.referencesMosquera-Trujillo, C.; Lugo-Rojas, J.; CollazosHuertas, D.; Álvarez Meza, A. & Castellanos-Dominguez, G.: , 2024b; A lightweight and interpretable rff-based neural network for automatic modulation classification; Preprint submitted to Elsevier
dc.relation.referencesO’Shea, T. J. & West, N.: , 2016; Radio machine learning dataset generation with gnu radio; en Proceedings of the GNU Radio Conference, tomo 1
dc.relation.referencesO’Shea, T. J. & West, N. E.: , 2017; Deep neural network architectures for modulation classification; en IEEE; URL https://ieeexplore.ieee. org/abstract/document/8335483
dc.relation.referencesO’Shea, T. J.; Corgan, J. & Clancy, T. C.: , 2016; Convolutional radio modulation recognition networks; en Engineering Applications of Neural Networks; Springer; URL https://link.springer.com/chapter/10.1007/ 978-3-319-44188-7_16
dc.relation.referencesO’Shea, T. J.; Corgan, J. & Clancy, T. C.: , 2017a; Deep architectures for modulation recognition; en IEEE; URL https://ieeexplore.ieee. org/abstract/document/7920754
dc.relation.referencesO’Shea, T. J.; Roy, T. & Clancy, T. C.: , 2017b; Over-the-air deep learning based radio signal classification; arXiv preprint arXiv:1712.04578
dc.relation.referencesO’Shea, T. J.; Roy, T. & Clancy, T. C.: , 2017b; Over-the-air deep learning based radio signal classification; arXiv preprint arXiv:1712.04578
dc.relation.referencesOuamna, H.; Kharbouche, A.; Madini, Z. & Zouine, Y.: , 2025; Deep learning-assisted automatic modulation classification using spectrograms; Engineering, Technology & Applied Science Research; 15 (1): 19925–19932; doi:10. 48084/etasr.9334
dc.relation.referencesRahimi, A. & Recht, B.: , 2007; Random features for large-scale kernel machines; en Advances in Neural Information Processing Systems 20
dc.relation.referencesRao, X.; Yan, X.; Wang, Q.; Wu, H.-C. & Wu, Y.: , 2024; Novel cooperative automatic modulation classification based on received-signal quality; en 2024 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB); doi:10.1109/BMSB62888.2024.10608244
dc.relation.referencesRasmussen, C. E. & Williams, C. K.: , 2006; Gaussian processes for machine learning; tomo 1; MIT press Cambridge
dc.relation.referencesRehman, F. U.; Abbas, Q. & Shehzad, M. K.: , 2025; Dl-amc: Deep learning for automatic modulation classification; arXiv preprint arXiv:2504.08011 ; URL https://arxiv.org/abs/2504.08011
dc.relation.referencesRiaz, A.; Hassan, S. A.; Mahmood, A.; Zikria, Y. B. & Ghafoor, K. Z.: , 2023; Physical layer security in iot networks: A survey; IEEE Internet of Things Journal; 10 (12): 10185–10210
dc.relation.referencesSadhukhan, P. & Bhaumik, J.: , 2024; Automatic modulation classification using convolutional neural network with batch normalization: A novel approach; en 2024 4th International Conference on Computer, Communication, Control & Information Technology (C3IT); doi:10.1109/C3IT60531.2024.10829439
dc.relation.referencesSaleh, M.; Khan, J. & Iqbal, Z.: , 2024; High-reliability envelopebased automatic modulation classification using deep learning; IEEE Access; doi: 10.1109/ACCESS.2024.3376812
dc.relation.referencesStoica, P. & Moses, R. L.: , 2005; Spectral analysis of signals; Pearson/Prentice Hall
dc.relation.referencesTariq, A.; Shakir, M. Z. et al.: , 2024; A transformer-based architecture for automatic modulation classification under realistic wireless channels; arXiv preprint arXiv:2411.08376; URL https://arxiv.org/abs/2411.08376
dc.relation.referencesTian, Y.; Xu, D.; Tong, E.; Sun, R.; Chen, K.; Li, Y.; Baker, T.; Niu, W. & Liu, J.: , 2024; Toward learning model-agnostic explanations for deep learning-based signal modulation classifiers; IEEE Transactions on Reliability; doi:10.1109/TR.2024.3367780
dc.relation.referencesTrabelsi, C.; Bilaniuk, O.; Zhang, Y.; Serdyuk, D.; Subramanian, S.; Santos, J. F.; Mehri, S.; Rostamzadeh, N.; Bengio, Y. & Pal, C. J.: , 2018; Deep complex networks; en International Conference on Learning Representations (ICLR)
dc.relation.referencesTüc, J. & Maršálek, R.: , 2019; Software defined radio: The software communications architecture; Radioengineering; 28 (4): 685–694
dc.relation.referencesVaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A. N.; Kaiser, Ł. & Polosukhin, I.: , 2017; Attention is all you need; en Advances in neural information processing systems, tomo 30
dc.relation.referencesWang, M.; Fang, S.; Fan, Y.; Li, J.; Zhao, Y. & Wang, Y.: , 2024a; An ultra lightweight neural network for automatic modulation classification in drone communications; Scientific Reports; 14 (21540): 1–14; doi: 10.1038/s41598-024-72867-1
dc.relation.referencesWang, Q.; Wu, B.; Zhu, P.; Li, P.; Wang, Z. & Hu, Q.: , 2020; Ecanet: Efficient channel attention for deep convolutional neural networks; en Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; págs. 11534– 11542
dc.relation.referencesWang, X.; Xie, W.; Shi, Q.; Yang, W.; Zhang, X. & Yang, Y.: , 2024b; Robust automatic modulation classification using asymmetric trilinear attention net with noisy activation function; Physical Communication; 62: 102078; doi:10.1016/j.phycom.2024.102078
dc.relation.referencesWang, Y.; Zhou, L.; Yang, Z.; Wu, L.; Yin, Z.; Zhao, Y. & Wu, Z.: , 2023; An improved modulation recognition algorithm based on fine-tuning and feature re-extraction; Electronics; 12 (9); doi:10.3390/electronics12092134; URL https://www.mdpi.com/2079-9292/12/9/2134
dc.relation.referencesWest, N. E. & O’Shea, T. J.: , 2018; Deep learning models for wireless signal classification with distributed low-cost spectrum sensors; IEEE Journal of Selected Topics in Signal Processing ; URL https://ieeexplore.ieee.org/abstract/ document/8357902
dc.relation.referencesXu, B.; Bhatti, U. A.; Tang, H.; Yan, J.; Wu, S.; Sarhan, N.; Awwad, E. M.; S., S. M. & Ghadi, Y. Y.: , 2024a; Towards explainability for ai-based edge wireless signal automatic modulation classification; Journal of Cloud Computing; 13 (1): 10; doi:10.1186/s13677-024-00590-3; URL https://doi.org/10. 1186/s13677-024-00590-3
dc.relation.referencesXu, B.; Bhatti, U. A.; Tang, H.; Yan, J.; Wu, S.; Sarhan, N.; Awwad, E. M.; S., S. M. & Ghadi, Y. Y.: , 2024b; Towards explainability for ai-based edge wireless signal automatic modulation classification; Journal of Cloud Computing; 13 (10); doi:10.1186/s13677-024-00590-3
dc.relation.referencesXu, J.; Luo, C.; Parr, G. & Luo, Y.: , 2020; A spatiotemporal multichannel learning framework for automatic modulation recognition; IEEE Wireless Communications Letters; 9 (10): 1629–1632; doi:10.1109/LWC.2020.2999453; URL https://ieeexplore.ieee.org/abstract/document/9106397
dc.relation.referencesXu, Z.; Fan, Y.; Fang, S.; Fu, Y. & Yi, L.: , 2025; A lightweight dual-branch complex-valued neural network for automatic modulation classification of communication signals; Sensors; 25 (8): 2489; doi:10.3390/s25082489
dc.relation.referencesYadav, A.; Dobre, O. A. & E-Hadi, H. M.: , 2023; 6g wireless communications networks: A comprehensive survey; IEEE Access; 11: 38733–38792
dc.relation.referencesZhang, F.; Luo, C.; Xu, J. & Luo, Y.: , 2021a; An efficient deep learning model for automatic modulation recognition based on parameter estimation and transformation; IEEE Communications Letters; doi:10.1109/LCOMM.2021.3102656; URL https://ieeexplore.ieee.org/abstract/document/9507514
dc.relation.referencesZhang, T.; Liu, H. & Chen, Y.: , 2023; Automatic modulation classification using hybrid data augmentation and lightweight neural network; Sensors; 23 (5): 1234
dc.relation.referencesZhang, X. et al.: , 2020; Cnn-based automatic modulation classification for beyond 5g communications; IEEE; URL https://ieeexplore.ieee.org/ abstract/document/8977561
dc.relation.referencesZhang, X. et al.: , 2021; Cgdnet: Efficient hybrid deep learning model for robust automatic modulation recognition; IEEE; URL https://ieeexplore.ieee. org/abstract/document/9349627
dc.relation.referencesZhang, Y. et al.: , 2022; Real-time radio technology and modulation classification via an lstm auto-encoder; IEEE; URL https://ieeexplore.ieee.org/ abstract/document/9487492
dc.relation.referencesZhang, Z.; Zhao, J.; Wang, Z.; Wang, S. & Wang, J.: , 2021b; Ami-cnn: an attention-based multi-scale interpretable cnn for automatic modulation identification; en 2021 IEEE International Conference on Communications (ICC); IEEE; págs. 1–6
dc.relation.referencesZheng, Q.; Tian, X.; Yu, Z. & Saponara, S.: , 2025a; Recent advances in automatic modulation classification technology: Methods, results, and prospects; International Journal of Intelligent Systems; 2025: 4067323
dc.relation.referencesZheng, Q.; Tian, X.; Yu, Z. et al.: , 2025b; Robust automatic modulation classification using asymmetric trilinear attention net with noisy activation function; Engineering Applications of Artificial Intelligence ; 141: 109861; doi:10.1016/j. engappai.2024.109861
dc.relation.referencesZhou, H.; Bai, J.; Wang, Y.; Ren, J.; Yang, X. & Jiao, L.: , 2024a; Deep radio signal clustering with interpretability analysis based on saliency map; Digital Communications and Networks; 10: 1448–1458; doi:10.1016/j.dcan.2023.01.010
dc.relation.referencesZhou, L.; Ahmed, S. & Zhao, L.: , 2024b; Recent advances in automatic modulation classification for cognitive radio and 6g: A survey; IEEE Access
dc.relation.referencesZhu, X.; Cheng, Y.; He, J. & Guo, J.: , 2024; Adaptive maskbased interpretable convolutional neural network (ami-cnn) for modulation format identification; Applied Sciences; 14 (14): 6302; doi:10.3390/app14146302
dc.rights.accessrightsinfo:eu-repo/semantics/openAccess
dc.rights.licenseReconocimiento 4.0 Internacional
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subject.ddc000 - Ciencias de la computación, información y obras generales
dc.subject.proposalClasificación automática de modulaciónspa
dc.subject.proposalAprendizaje profundospa
dc.subject.proposalModelos ligerosspa
dc.subject.proposalInterpretabilidad de IAspa
dc.subject.proposalRadio definido por softwarespa
dc.subject.proposalComputación en el dordespa
dc.subject.proposalRedes neuronales convolucionalesspa
dc.subject.proposalRobustezspa
dc.subject.proposalAutomatic modulation classificationeng
dc.subject.proposalDeep learningeng
dc.subject.proposalLightweight modelseng
dc.subject.proposalAI Interpretabilityeng
dc.subject.proposalSoftware-Defined radioeng
dc.subject.proposalEdge computingeng
dc.subject.proposalConvolutional neural networkseng
dc.subject.proposalRobustnesseng
dc.subject.unescoTelecomunicación
dc.subject.unescoTelecommunications
dc.subject.unescoProcesamiento de datos
dc.subject.unescoData processing
dc.subject.unescoTecnología electrónica
dc.subject.unescoElectronic engineering
dc.titleAutomatic modulation classification for software defined radios : a deep learning approacheng
dc.title.translatedClasificación automática de modulación para radios definidos por software : un enfoque de aprendizaje profundospa
dc.typeTrabajo de grado - Maestría
dc.type.coarhttp://purl.org/coar/resource_type/c_bdcc
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.contentText
dc.type.driverinfo:eu-repo/semantics/masterThesis
dc.type.versioninfo:eu-repo/semantics/acceptedVersion
dcterms.audience.professionaldevelopmentInvestigadores
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
Tesis de Maestría en Ingeniería - Automatización Industrial.pdf
Tamaño:
26.09 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ingeniería - Automatización Industrial

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
5.74 KB
Formato:
Item-specific license agreed upon to submission
Descripción: