Citation :

Hemant Soni, 2025. "Generative AI for Network Optimization: Autonomous Systems in TMT Infrastructure" ESP International Journal of Advancements in Science & Technology (ESP-IJAST)  Volume 3, Issue 4: 5-12.

Abstract :

It has become more difficult to manage modern telecom infrastructure due to complex networks and unpredictable traffic patterns. We designed and tested a generative AI-based system that makes network optimization decisions automatically. To construct a multi-objective optimizer that can handle speed, response time, and running costs simultaneously, the system combines transformer-based large language models with reinforcement learning agents. Experiments in a cloud-based simulation environment with realistic network topologies and traffic traces yielded performance gains that were statistically significant: throughput was increased by 34%, latency was reduced by 28%, and costs were cut by 41% as compared to conventional techniques. The proposed framework addressed scenarios such as 5G network slicing, edge computing workload placement, and content delivery optimization. These outcomes are derived from experiments in a controlled setting rather than actual deployments, but they still indicate that generative AI could assist network management. We will discuss the promise of this approach as well as its current limitations.

References :

[1] J. G. Andrews et al., "What Will 5G Be," IEEE Journal on Selected Areas in Communications, vol. 32, no. 6, pp. 1065-1082, June 2014.

[2] N. Feamster and H. Balakrishnan, "Detecting BGP Configuration Faults with Static Analysis," in Proc. USENIX Symp. Networked Systems Design Implementation (NSDI), 2005, pp. 43-56.

[3] T. Brown et al., "Language Models are Few-Shot Learners," Advances in Neural Information Processing Systems, vol. 33, pp. 1877-1901, 2020.

[4] Anthropic, "Claude 3 Model Card," Technical Report, 2024.

[5] H. Mao et al., "Neural Adaptive Video Streaming with Pensive," in Proc. ACM SIGCOMM Conf., 2017, pp. 197-210.

[6] GSMA, "The Mobile Economy 2024," Technical Report, 2024.

[7] Cisco, "Cisco Annual Internet Report (2018-2023) White Paper," Technical Report, 2020.

[8] M. Peng et al., "Fog-Computing-Based Radio Access Networks: Issues and Challenges," IEEE Network, vol. 30, no. 4, pp. 46-53, July 2016.

[9] D. Oppenheimer et al., "Why Do Internet Services Fail, and What Can Be Done About It?" in Proc. USENIX Symp. Internet Technologies and Systems, 2003.

[10] IHS Markit, "The Cost of Network Downtime," Technical Report, 2021.

[11] X. Fouke’s et al., "Network Slicing in 5G: Survey and Challenges," IEEE Communications Magazine, vol. 55, no. 5, pp. 94-100, May 2017.

[12] M. Feldmann et al., "A Year in Lockdown: How the Waves of COVID-19 Impact Internet Traffic," Communications of the ACM, vol. 64, no. 10, pp. 101-108, 2021.

[13] Sand vine, "The Global Internet Phenomena Report COVID-19 Spotlight," Technical Report, 2020.

[14] C. Han et al., "Green Radio: Radio Techniques to Enable Energy-Efficient Wireless Networks," IEEE Communications Magazine, vol. 49, no. 6, pp. 46-54, June 2011.

[15] A. Forts and M. Thorup, "Internet Traffic Engineering by Optimizing OSPF Weights," in Proc. IEEE INFOCOM, 2000, pp. 519-528.

[16] "Towards AI/ML-Driven Network Traffic Engineering -https://dl.acm.org/Doi/10.1145/3703412.3703436

[17] GSMA Intelligence, "Understanding 5G: Perspectives on Future Technological Advancements in Mobile," White Paper, 2019.

[18] 5G is not just another G: A review of the 5G business model and ecosystem challenges," https://www.sciencedirect.com/science/article/pii/S0040162525001520.

[19] McKinsey & Company, "Cutting Through the 5G Hype: Survey Shows Telcos' Nuanced Views," Report, 2021.

[20] A. Beloglazov and R. Buya, "Energy Efficient Resource Management in Virtualized Cloud Data Centers," in Proc. IEEE/ACM Int. Conf. Cluster, Cloud and Grid Computing, 2010, pp. 826-831.

[21] World Economic Forum, "The Impact of 5G: Creating New Value across Industries and Society," White Paper, 2020.

[22] D. Adduce et al., "Overview and Principles of Internet Traffic Engineering," IETF RFC 3272, May 2002.

[23] B. Forts and M. Thorup, "Optimizing OSPF/IS-IS Weights in a Changing World," IEEE Journal on Selected Areas in Communications, vol. 20, no. 4, pp. 756-767, May 2002.

[24] Z. Xu et al., "Experience-Driven Networking: A Deep Reinforcement Learning Based Approach," in Proc. IEEE INFOCOM, 2018, pp. 1871-1879.

[25] R. Li et al., "Deep Reinforcement Learning for Resource Management in Network Slicing," IEEE Access, vol. 6, pp. 74429-74441, 2018.

[26] K. Rusek et al., "Route Net: Leveraging Graph Neural Networks for Network Modeling and Optimization in SDN," IEEE Journal on Selected Areas in Communications, vol. 38, no. 10, pp. 2260-2270, Oct. 2020.

[27] J. Min et al., "TR-IDS: Anomaly-Based Intrusion Detection through Text-Convolutional Neural Network and Random Forest," Security and Communication Networks, vol. 2018, Article ID 4943509, 2018.

[28] Z. Liu et al., "Generative Adversarial Active Learning for Unsupervised Outlier Detection," IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 8, pp. 1517-1528, Aug. 2020.

[29] James Won‐Ki Hong., " A Comprehensive Survey on LLM‐Based Network Management and Operations," https://dl.acm.org/doi/10.1002/nem.70029.

[30] P. Velichkova et al., "Graph Attention Networks," in Proc. Int. Conf. Learning Representations (ICLR), 2018.

[31] S. Bai et al., "An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling," arrive preprint arXiv:1803.01271, 2018.

[32] H. Touran et al., "Llama: Open and Efficient Foundation Language Models," arrive preprint arXiv:2302.13971, 2023.

[33] J. Schulman et al., "Proximal Policy Optimization Algorithms," arrive preprint arXiv:1707.06347, 2017.

[34] M. Castro and B. Liskov, "Practical Byzantine Fault Tolerance," in Proc. USENIX Symp. Operating Systems Design Implementation (OSDI), 1999, pp. 173-186.

[35] B. Lantz et al., "A Network in a Laptop: Rapid Prototyping for Software-Defined Networks," in Proc. ACM SIGCOMM Workshop Hot Topics in Networks, 2010, pp. 19:1-19:6.

[36] A. Fogel et al., "A General Approach to Network Configuration Analysis," in Proc. USENIX Symp. Networked Systems Design Implementation (NSDI), 2015, pp. 469-483.

[37] B. McMahan et al., "Communication-Efficient Learning of Deep Networks from Decentralized Data," in Proc. Int. Conf. Artificial Intelligence and Statistics (AISTATS), 2017, pp. 1273-1282.

Keywords :

Generative AI, Network Optimization, Autonomous Systems, TMT Infrastructure, Reinforcement Learning, Large Language Models, 5G Networks, Edge Computing.