Authors

C. C. Ezeh

Department of Computer Science, Rivers State University, Port Harcourt, Nigeria

V. I. E. Anireh

Department of Computer Science, Rivers State University, Port Harcourt, Nigeria

D. Matthias

Department of Computer Science, Rivers State University, Port Harcourt, Nigeria

Abstract

Broadband connectivity remains constrained by high deployment cost, sparse population distribution, long inter-node distances, difficult terrain, and limited adaptability in existing path-selection mechanisms. Existing integrated access and backhaul architectures provide useful coverage extension, but they do not adequately address cost-effective hybrid infrastructure planning or adaptive routing across heterogeneous backhaul links. This study developed a model for improved reliability in broadband connectivity using intelligent path selection. The model combined a hybrid fiber-microwave broadband architecture with Python NetworkX-based graph simulation and Q-learning-based path selection. The model is applicable to heterogeneous broadband environments, whiles this study implements and evaluates it in a rural deployment scenario because rural networks present stronger cost, distance, terrain, and reliability constraints. Synthetic rural topology data were generated to represent rural nodes, gateway connectivity, household distribution, traffic demand, fiber links, microwave links, and failure scenarios. The Q-learning algorithm selected paths using latency, cost, reliability, and load-balance criteria, while the proposed system was evaluated against an existing Integrated Access and Backhaul (IAB) baseline and alternative deployment baselines. Results showed that the proposed model recorded average latency of 7.71 ms compared with 66.11 ms for the existing baseline, throughput of 1,011.20 Mbps compared with 654.65 Mbps, packet loss of 0.00% compared with 36.29%, availability of 92.922% compared with 48.408%, failover time of 2.72 seconds compared with 6.59 seconds, and load balance of 0.710 compared with 0.257. The model also recorded five-year total cost of ownership of USD 1,684,302.07 compared with USD 2,892,917.44 and cost per household of USD 463.36 compared with USD 795.85, producing a 41.78% cost saving relative to the existing IAB baseline. Scalability testing up to 150 rural nodes and 260 proposed hybrid links maintained low latency, about 92% reliability/availability, balanced load distribution, and measurable cost-per-household output. The study contributes a validated cost-resilience evaluation model, a tested multi-criteria path-selection approach, and a transferable simulation-to-decision-support framework for broadband planning.

Keywords

Broadband connectivity intelligent path selection Q-learning hybrid fibre-microwave backhaul NetworkX simulation reliability scalability.

Citation of this Article

C. C. Ezeh, V. I. E. Anireh, & D. Matthias. (2026). A Model for Improved Reliability in Broadband Connectivity Using Intelligent Path Selection. Journal of Artificial Intelligence and Emerging Technologies (JAIET). 3(8), 9-18. Article DOI: https://doi.org/10.47001/JAIET/2026.308002  

Licence Copyright (c) 2026 Journal of Artificial Intelligence and Emerging Technologies. This work is licensed under a Creative Commons Attribution Non Commercial 4.0 International Licence.

References

  1. Lee, H., Jeong, S., & Lee, K. (2023). The South Korean case of deploying rural broadband via fiber networks by implementing universal service obligation and public-private partnership based project. Telecommunications Policy, 47(3), Article 102506. https://doi.org/10.1016/j.telpol.2023.102506.
  2. Zhang, J., Wang, Q., Mitchell, P., & Ahmadi, H. (2024). An integrated access and backhaul approach to sustainable dense small cell network planning. Information, 15(1), Article 19. https://doi.org/10.3390/info15010019.
  3. Prasanth, L. L., & Uma, E. (2024). A computationally intelligent framework for traffic engineering and congestion management in software-defined network (SDN). EURASIP Journal on Wireless Communications and Networking, 2024, Article 63. https://doi.org/10.1186/s13638-024-02392-2.
  4. Li, X., Li, J., Zhou, J., & Liu, J. (2025). Towards robust routing: Enabling long-range perception with the power of graph transformers and deep reinforcement learning in software-defined networks. Electronics, 14(3), Article 476. https://doi.org/10.3390/electronics14030476.
  5. Madapatha, C., Makki, B., Guo, H., & Svensson, T. (2023). Constrained deployment optimization in integrated access and backhaul networks. In Proceedings of the 2023 IEEE Wireless Communications and Networking Conference (WCNC). IEEE. https://doi.org/10.1109/WCNC55385.2023.10119093.
  6. Shang, W., &Friderikos, V. (2025). Energy efficient optimization of in-band integrated access and backhaul heterogeneous networks. IEEE Transactions on Vehicular Technology, 74(4), 6504-6517. https://doi.org/10.1109/TVT.2024.3514928.
  7. Zhang, Y., Kishk, M. A., &Alouini, M.-S. (2024). Deployment optimization of tethered drone-assisted integrated access and backhaul networks. IEEE Transactions on Wireless Communications, 23(1), 644-660. https://doi.org/10.1109/TWC.2023.3301880.
  8. Tezergil, B., & Onur, E. (2022). Wireless backhaul in 5G and beyond: Issues, challenges and opportunities. IEEE Communications Surveys & Tutorials, 24(4), 2579-2632. https://doi.org/10.1109/COMST.2022.3203578.
  9. Madapatha, C., Lechowicz, P., Natalino, C., Monti, P., & Svensson, T. (2025). Joint fiber and free space optical infrastructure planning for hybrid integrated access and backhaul networks. In Proceedings of the 2025 IEEE 36th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC). IEEE. https://doi.org/10.1109/PIMRC62392.2025.11275515.
  10. Gedel, I. A., &Nwulu, N. I. (2024). Low latency 5G IP transmission backhaul network architecture: A techno-economic analysis. Wireless Communications and Mobile Computing, 2024, Article 6388723. https://doi.org/10.1155/2024/6388723
  11. Filgueiras, H. R. D., Lima, E. S., Cunha, M. S. B., Lopes, C. H. S., Souza, L. C., & Borges, R. M. (2023). Wireless and optical convergent access technologies toward 6G. IEEE Access, 11, 9232-9259. https://doi.org/10.1109/ACCESS.2023.3239807.
  12. Koratagere Anantha Kumar, S., & Oughton, E. J. (2023). Infrastructure sharing strategies for wireless broadband. IEEE Communications Magazine, 61(7), 78-84. https://doi.org/10.1109/MCOM.005.2200698.
  13. Valentin-Sivico, J., Canfield, C., Low, S. A., & Gollnick, C. (2023). Evaluating the impact of broadband access and internet use in a small underserved rural community. Telecommunications Policy, 47(4), Article 102499. https://doi.org/10.1016/j.telpol.2023.102499.
  14. Islam, T. U., Boateng, J. O., Nadim, M., Zu, G., Shahid, M., & Li, X. (2025). Design and implementation of ARA wireless living lab for rural broadband and applications. Computer Networks, 268, Article 111188. https://doi.org/10.1016/j.comnet.2025.111188.
  15. Kim, G., Kim, Y., & Lim, H. (2022). Deep reinforcement learning-based routing on software-defined networks. IEEE Access, 10, 30227-30239. https://doi.org/10.1109/ACCESS.2022.3151081.
  16. Huang, R., Guan, W., Zhai, G., He, J., & Chu, X. (2022). Deep graph reinforcement learning based intelligent traffic routing control for software-defined wireless sensor networks. Applied Sciences, 12(4), Article 1951. https://doi.org/10.3390/app12041951.
  17. Pei, X., Sun, P., Hu, Y., Li, D., Chen, B., & Tian, L. (2024). Enabling efficient routing for traffic engineering in SDN with deep reinforcement learning. Computer Networks, 250, Article 110220. https://doi.org/10.1016/j.comnet.2024.110220.
  18. Ding, Q., Li, N., Ding, H., Wang, J., Li, T., & Chen, Y. (2025). Traffic prediction and load balancing routing algorithm based on deep Q-network for SD-IoT. Advanced Engineering Informatics, 65, Article 103596. https://doi.org/10.1016/j.aei.2025.103596.