Adaptive Resource Scheduling for IoT Big Data Stream Processing Based on Deep Reinforcement Learning
Main Article Content
Keywords
deep reinforcement learning, Internet of Things, big data stream processing, adaptive resource scheduling, intelligent decision-making
Abstract
With the rapid development of Internet of Things (IoT) technology and the explosive growth of data scale, traditional static resource scheduling methods can no longer meet the dynamic and heterogeneous requirements of IoT big data stream processing. This paper proposes an adaptive resource scheduling method based on deep reinforcement learning, which achieves intelligent resource scheduling decisions in IoT environments by constructing multi-level state space models, designing intelligent action spaces, and multi-objective reward functions. Experiments were conducted based on three typical application scenarios: smart cities, industrial IoT, and smart grids, establishing large-scale testing environments containing over 28,000 devices. Results demonstrate that compared to traditional scheduling methods, the proposed method achieves improvements of 42.3%-84.0% in response time, 56.7%-70.2% enhancement in system throughput, 34.1%-50.1% improvement in resource utilization, and 50.2%-60.3% enhancement in energy efficiency. During 18 months of actual deployment, cumulative operational cost savings reached 101.8 million yuan, with a payback period of only 1.8 years. Long-term stability testing shows that the algorithm processed 1.52 billion scheduling decisions during 30 days of continuous operation, with performance fluctuations controlled within 6.4%, demonstrating excellent convergence and robustness. The research findings provide theoretical foundations and technical support for intelligent resource management in IoT big data stream processing, holding significant importance for promoting the industrial application of IoT technology.
References
- Alqerm, I., & Pan, J. (2021). DeepEdge: A new QoE-based resource allocation framework using deep reinforcement learning for future heterogeneous edge-IoT applications. IEEE Transactions on Network and Service Management, 18(4), 3942-3954. https://doi.org/10.1109/TNSM.2021.3123959
- Atieh, M., Mohammad, O., Sabra, A., & Rmayti, N. (2022). Iot, deep learning and cybersecurity in smart homes: A survey. In Cybersecurity in smart homes: Architectures, solutions and technologies (pp. 203-244). wiley. https://doi.org/10.1002/9781119987451.CH6
- Du, W., Du, X., Ma, M., Huang, S., Sun, X., & Xiong, L. (2022). Polymer electrode materials for lithium-ion batteries. Advanced Functional Materials, 32(21), Article 2110871. https://doi.org/10.1002/ADFM.202110871
- Farag, H., Gidlund, M., & Stefanovic, C. (2021). A deep reinforcement learning approach for improving age of information in mission-critical IoT [Paper presentation]. 2021 IEEE Global Conference on Artificial Intelligence and Internet of Things, GCAIoT 2021, Dubai, United Arab Emirates.
- He, Y., Wang, Y., Qiu, C., Lin, Q., Li, J., & Ming, Z. (2021). Blockchain-based edge computing resource allocation in IoT: A deep reinforcement learning approach. IEEE Internet of Things Journal, 8(4), 2226-2237. https://doi.org/10.1109/JIOT.2020.3035437
- Idrissi, I., Azizi, M., & Moussaoui, O. (2022, 3-4 March 2022). A stratified IoT deep learning based intrusion detection system [Paper presentation]. 2022 2nd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), Meknes, Morocco.
- Khelifi, H., Luo, S., Nour, B., Sellami, A., Moungla, H., Ahmed, S. H., & Guizani, M. (2019). Bringing deep learning at the edge of information-centric internet of things. IEEE Communications Letters, 23(1), 52-55. https://doi.org/10.1109/LCOMM.2018.2875978
- Khodaparast, S. S., Lu, X., Wang, P., & Nguyen, U. T. (2021). Deep reinforcement learning based energy efficient multi-UAV data collection for IoT networks. IEEE Open Journal of Vehicular Technology, 2, 249-260. https://doi.org/10.1109/OJVT.2021.3085421
- Malik, S., Tyagi, A. K., & Mahajan, S. (2022). Architecture, generative model, and deep reinforcement learning for IoT applications: Deep learning perspective. In S. Pal, D. De, & R. Buyya (Eds.), Internet of things (pp. 243-265). Springer. https://doi.org/10.1007/978-3-030-87059-1_9
- Raman, R., Kumar, V., Saini, D., Rabadiya, D., Rastogi, S., & Kumbhojkar, N. (2024). Energy-efficient deep Q-network for reinforcement learning in wireless IoT routing protocols [Paper presentation]. 2024 Asian Conference on Intelligent Technologies, ACOIT 2024, Kolar, India.
- Román-Ramírez, L. A., & Marco, J. (2022). Design of experiments applied to lithium-ion batteries: A literature review. Applied Energy, 320, Article 119305. https://doi.org/10.1016/J.APENERGY.2022.119305
- Selvi, K. T., Mangai, K. A., Lett, J. A., Fatimah, I., & Sagadevan, S. (2024). Exploring the electrode materials for high-performance lithium-ion batteries for energy storage application. Journal of Energy Storage, 92, Article 112208. https://doi.org/10.1016/J.EST.2024.112208
- Xu, G., Jiang, M., Li, J., Xuan, X., Li, J., Lu, T., & Pan, L. (2024). Machine learning-accelerated discovery and design of electrode materials and electrolytes for lithium ion batteries. Energy Storage Materials, 72, Article 103710. https://doi.org/10.1016/J.ENSM.2024.103710
- Zhu, P., Slater, P. R., & Kendrick, E. (2022). Insights into architecture, design and manufacture of electrodes for lithium-ion batteries. Materials and Design, 223, Article 111208. https://doi.org/10.1016/J.MATDES.2022.111208
- Zhu, X., Zhang, T., Zhang, J., Zhao, B., Zhang, S., & Wu, C. (2023). Deep reinforcement learning-based edge computing offloading algorithm for software-defined IoT. Computer Networks, 235, Article 110006. https://doi.org/10.1016/J.COMNET.2023.110006
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.