Um Novo Módulo Arquitetural de Controle Centralizado com DRL para Redes IoT Densas sem Bateria Baseadas em Retroespalhamento Ambiente
Resumo
Redes IoT densas compostas por dispositivos sem bateria, baseadas em AmBC, enfrentam desafios de interferência, controle e estabilidade. Este trabalho apresenta o GlobalAmBC-DRL Control Module, uma arquitetura de controle centralizada que utiliza aprendizado por reforço profundo sobre um estado global agregado da rede. A abordagem permite reduzir a interferência entre dispositivos e aumentar a estabilidade da comunicação e a eficiência energética. O módulo é implementado como extensão reutilizável em OMNeT++ e avaliado sob diferentes densidades de rede. Os resultados indicam menor variabilidade na taxa de sucesso e melhor equilíbrio energético em comparação com estratégias estáticas, fornecendo uma base arquitetural reprodutível para IoT sustentável.Referências
Huynh, N. V. et al. (2018a). Optimal and low-complexity dynamic spectrum access for rf-powered ambient backscatter system with online reinforcement learning. arXiv pre-print. Early work applying RL to backscatter communications.
Huynh, N. V., Thai, H. D., Lu, X., Niyato, D., and et al. (2018b). Ambient backscatter communications: A contemporary survey. IEEE Communications Surveys & Tutorials. Overview of ambient backscatter communications and challenges.
Imran, M., Ahmad, I., and Chen, Y. (2025). Simulation-based optimization of energy harvesting systems for 6g networks using deep reinforcement learning. Journal of Communications and Networks. DRL-based optimization of energy harvesting systems targeting future 6G networks.
Jamshed, M. A. et al. (2025). Artificial intelligence, ambient backscatter communication. arXiv preprint. Survey on AI techniques applied to AmBC.
Khan, M. A., Rehman, A. U., Khan, Z., and Niyato, D. (2023). Deep reinforcement learning for backscatter communications: Augmenting intelligence in future internet of things. IEEE Internet of Things Journal. Survey and analysis of DRL techniques applied to backscatter communication systems and IoT.
Liu, Q. et al. (2021). Ambient backscatter communication-based smart 5g iot network. EURASIP Journal on Wireless Communications and Networking. Example of AmBC in smart IoT networks.
Liu, V., Parks, A., Talla, V., Gollakota, S., Wetherall, D., and Smith, J. R. (2013). Ambient backscatter: Wireless communication out of thin air. ACM SIGCOMM Computer Communication Review, 43(4):39–50.
Memon, M. L. et al. (2019). Backscatter communications: Inception of the battery-free era. Electronics. Review on backscatter communications and battery-free systems.
Perez-Ramirez, D. F. et al. (2021). Deepgantt: A scalable deep learning scheduler for backscatter networks. arXiv preprint. Deep learning approaches for scalable backscatter communications.
Shoeibi, A., Gholami, A., and Mosavi, A. (2024). A novel drl-based scheme for risassisted energy harvesting in batteryless iot networks. Future Internet, 16(12):460. DRL-based optimization for energy harvesting and performance in batteryless IoT networks.
Sooriarachchi, V. P. et al. (2025). Ambient backscatter-and simultaneous wireless information and power transfer. Applied Sciences. Hybrid mechanisms combining power transfer with backscatter communications.
Xie, L., Xu, Y., and Wang, R. (2021). Resource allocation in ambient backscatter communication networks using deep reinforcement learning. IEEE Internet of Things Journal, 8(6):4494–4506.
Zargari, S. H. and Larsson, E. G. (2024). Refined deep reinforcement learning for mimo bistatic backscatter resource allocation. arXiv preprint arXiv:2405.14046. DRL-based resource allocation in MIMO bistatic backscatter communication systems.
Zhao, Y., Li, B., and Wang, Y. (2022). Deep reinforcement learning for resource optimization in iot networks. IEEE Transactions on Network and Service Management, 19(2):1234–1246.
Zheng, K. et al. (2024). Ambient iot towards 6g: Standardization, potentials, and challenges. arXiv preprint. Recent directions in Ambient IoT and backscatter communications.
Huynh, N. V., Thai, H. D., Lu, X., Niyato, D., and et al. (2018b). Ambient backscatter communications: A contemporary survey. IEEE Communications Surveys & Tutorials. Overview of ambient backscatter communications and challenges.
Imran, M., Ahmad, I., and Chen, Y. (2025). Simulation-based optimization of energy harvesting systems for 6g networks using deep reinforcement learning. Journal of Communications and Networks. DRL-based optimization of energy harvesting systems targeting future 6G networks.
Jamshed, M. A. et al. (2025). Artificial intelligence, ambient backscatter communication. arXiv preprint. Survey on AI techniques applied to AmBC.
Khan, M. A., Rehman, A. U., Khan, Z., and Niyato, D. (2023). Deep reinforcement learning for backscatter communications: Augmenting intelligence in future internet of things. IEEE Internet of Things Journal. Survey and analysis of DRL techniques applied to backscatter communication systems and IoT.
Liu, Q. et al. (2021). Ambient backscatter communication-based smart 5g iot network. EURASIP Journal on Wireless Communications and Networking. Example of AmBC in smart IoT networks.
Liu, V., Parks, A., Talla, V., Gollakota, S., Wetherall, D., and Smith, J. R. (2013). Ambient backscatter: Wireless communication out of thin air. ACM SIGCOMM Computer Communication Review, 43(4):39–50.
Memon, M. L. et al. (2019). Backscatter communications: Inception of the battery-free era. Electronics. Review on backscatter communications and battery-free systems.
Perez-Ramirez, D. F. et al. (2021). Deepgantt: A scalable deep learning scheduler for backscatter networks. arXiv preprint. Deep learning approaches for scalable backscatter communications.
Shoeibi, A., Gholami, A., and Mosavi, A. (2024). A novel drl-based scheme for risassisted energy harvesting in batteryless iot networks. Future Internet, 16(12):460. DRL-based optimization for energy harvesting and performance in batteryless IoT networks.
Sooriarachchi, V. P. et al. (2025). Ambient backscatter-and simultaneous wireless information and power transfer. Applied Sciences. Hybrid mechanisms combining power transfer with backscatter communications.
Xie, L., Xu, Y., and Wang, R. (2021). Resource allocation in ambient backscatter communication networks using deep reinforcement learning. IEEE Internet of Things Journal, 8(6):4494–4506.
Zargari, S. H. and Larsson, E. G. (2024). Refined deep reinforcement learning for mimo bistatic backscatter resource allocation. arXiv preprint arXiv:2405.14046. DRL-based resource allocation in MIMO bistatic backscatter communication systems.
Zhao, Y., Li, B., and Wang, Y. (2022). Deep reinforcement learning for resource optimization in iot networks. IEEE Transactions on Network and Service Management, 19(2):1234–1246.
Zheng, K. et al. (2024). Ambient iot towards 6g: Standardization, potentials, and challenges. arXiv preprint. Recent directions in Ambient IoT and backscatter communications.
Publicado
25/05/2026
Como Citar
GALHARDO, Edwardes A.; WESTPHALL, Carlos B.; OLIVEIRA JR, Antonio.
Um Novo Módulo Arquitetural de Controle Centralizado com DRL para Redes IoT Densas sem Bateria Baseadas em Retroespalhamento Ambiente. In: SIMPÓSIO BRASILEIRO DE REDES DE COMPUTADORES E SISTEMAS DISTRIBUÍDOS (SBRC), 44. , 2026, Praia do Forte/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1387-1400.
ISSN 2177-9384.
DOI: https://doi.org/10.5753/sbrc.2026.19911.
