Joint Configuration Optimization for Edge-Cloud Collaborative Inference in Visual IoT Systems
Keywords:
Visual Internet of Things, Edge Computing, Collaborative Inference, Deep Reinforcement Learning, Resource OptimizationAbstract
The rapid proliferation of visual Internet of Things applications, such as autonomous navigation, smart surveillance, and augmented reality, has driven the demand for real-time and highly accurate computer vision systems. However, executing complex deep neural networks exclusively on resource-constrained end devices is impractical due to severe computational and energy limitations. Relying solely on cloud computing introduces unacceptable communication latency and significant bandwidth consumption, which are detrimental to time-sensitive applications. To address these challenges, edge-cloud collaborative inference has emerged as a promising paradigm. In this architecture, deep learning models are dynamically partitioned between end devices, edge servers, and cloud servers. This paper presents a comprehensive framework for the joint configuration optimization of edge-cloud collaborative inference in visual Internet of Things systems. The proposed framework simultaneously optimizes multiple critical parameters, including camera frame rate, image capture resolution, neural network partition points, and intermediate feature map quantization levels. By formulating the configuration selection as a multidimensional optimization problem, we aim to minimize end-to-end inference latency and overall system energy consumption while strictly satisfying predefined baseline accuracy requirements. Due to the highly dynamic nature of network conditions and the non-convexity of the problem, we develop a deep reinforcement learning approach that adapts to time-varying channel states and server workloads. Extensive simulation results demonstrate that the proposed joint optimization strategy significantly outperforms conventional edge-only, cloud-only, and static partitioning baselines across various network environments. The findings provide critical insights into the design and deployment of efficient, scalable, and robust visual sensing networks for next-generation smart city infrastructure.
Downloads
References
1. Yang, Z., Liang, B., & Ji, W. (2021). An intelligent end–edge–cloud architecture for visual IoT-assisted healthcare systems. IEEE internet of things journal, 8(23), 16779-16786.
2. Yang, W., Qin, Y., & Yang, Z. (2022). A reinforcement learning based data storage and traffic management in information-centric data center networks. Mobile Networks and Applications, 27(1), 266-275.
3. Wang, S., Zhang, X., Wang, P., & Li, X. (2026). Design of Magnetic Sensor Array-Based PUFs for IoT Security. IEEE Transactions on Instrumentation and Measurement, 75, 1-9.
4. Yao, Y., Lu, Z., Chen, G., Huang, C., Feng, C., & Quek, T. Q. (2026). Interference Management in ISAC-SAGINs Based on Transformer-Enabled Mean-Field Reinforcement Learning Method. IEEE Transactions on Wireless Communications.
5. Chen, W., Li, X., Liu, X., & Yin, L. (2015). Study of self-calibrating MEMS accelerometers.AIP Advances,5(4).
6. Zhang, W., Li, M., & Chen, X. (2026). Distributed Edge-Fog-Cloud Architecture for Intelligent Fault Diagnosis of Permanent Magnet Synchronous Motors Using Temporal Fusion Transformers and Federated Learning. Fundamental Scientific Reports in Multidisciplinary Areas, 2 (02), 121-131.
7. Li, X., Chen, W., Liu, X., & Ren, M. (2015). A high-order sigma-delta accelerometer interface circuit.IEICE Electronics Express,12(2), 20141035-20141035.
8. Xia, L., Yang, M., Wang, J., Yan, Z., Ren, Y., Yu, G., & Lei, K. (2025, September). Mamba4Net: Distilled Hybrid Mamba Large Language Models For Networking. In2025 IEEE 33rd International Conference on Network Protocols (ICNP)(pp. 1-10). IEEE.
9. Yang, W., Qin, Y., Jiang, Z., & Chu, X. (2021, June). Traffic Management for Distributed Machine Learning in RDMA-enabled Data Center Networks. In ICC 2021-IEEE International Conference on Communications (pp. 1-6). IEEE.
10. Yang, Z., Ji, W., & Wang, Z. (2024). Adaptive joint configuration optimization for collaborative inference in edge-cloud systems. Science China Information Sciences, 67(4), 149103.
11. Li, X., Yin, L., Chen, W., Gao, Z., & Liu, X. (2017). A high-resolution tunneling magneto-resistance sensor interface circuit.Modern Physics Letters B,31(04), 1750030.
12. Zhao, Q., & Huang, Y. (2026). DAREA: A Dynamic AI-Driven Architecture for Real Estate Asset Tokenization on Blockchain.Fundamental Scientific Reports in Multidisciplinary Areas,2(02), 182-192.
13. Yang, Z., Ji, W., Guo, Q., & Wang, Z. (2023, October). Javp: Joint-aware video processing with edge-cloud collaboration for dnn inference. In Proceedings of the 31st ACM International Conference on Multimedia (pp. 9152-9160).
14. Yang, Y., Ma, X., Li, C., Zheng, Z., Zhang, Q., Huang, G., ... & Zhao, Q. (2021). Believe what you see: Implicit constraint approach for offline multi-agent reinforcement learning. Advances in Neural Information Processing Systems, 34, 10299-10312.
15. Li, X., Wang, P., Li, G., Ni, L., & Zhang, Y. (2023). Design of interface circuits and lightweight PUF for TMR sensors. IEEE Sensors Journal, 23(11), 11754-11761.
16. Yao, Y., Zhu, Z., Miao, P., Cheng, X., Shu, F., & Wang, J. (2025). Optimizing hybrid RIS-aided ISAC systems in V2X networks: A deep reinforcement learning method for anti-eavesdropping techniques. IEEE Transactions on Vehicular Technology, 74 (6), 9224-9239.
17. Zaharia, M., Chowdhury, M., Das, T., Dave, A., Ma, J., McCauley, M., et al. (2010). Spark: Cluster computing with working sets. In Proceedings of HotCloud.
18. Yao, Y., Sun, J., Miao, P., Chen, G., & Tafazolli, R. (2026). Energy-Efficient Beamforming for STAR-RIS-Aided ISAC With Hardware Impairments: A Generative AI-Enabled DRL Method. IEEE Transactions on Wireless Communications.
19. Chen, L., Ge, L., Jiang, X., Jiang, Z., & Li, H. (2026). Tensor-Based Target Sensing for Resource-Irregular ISAC Systems. IEEE Transactions on Signal Processing.
20. Wan, H., Zhang, Y., Wang, J., Wu, D., Li, M., Chen, X., ... & Ji, X. (2025). Toward universal embodied planning in scalable heterogeneous field robots collaboration and control. Journal of Field Robotics, 42(5), 2318-2336.
21. Xu, C., Wang, C., & Li, Z. (2026). H2-LBM: A Hierarchical Hybrid Deep Reinforcement Learning Framework for L7 Load Balancing and Global Traffic Scheduling in Multi-Cloud LLM Serving. Fundamental Scientific Reports in Multidisciplinary Areas, 2 (02), 109-120.
22. Hu, Y., Yang, Z., Zhao, C., & Ji, W. (2026, May). Adaptive guidance semantically enhanced via multimodal llm for edge-cloud object detection. In ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 12962-12966). IEEE.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Margaret Wai-Kit Choi, Linda Woo
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This article is licensed under the Creative Commons Attribution 4.0 International License. This license permits use, sharing, and adaptation of the article for non-commercial purposes, provided the original authors and source are properly credited.
Alternatively, you may choose other Creative Commons licenses based on your preferences:
-
CC Attribution-NonCommercial-NoDerivatives 4.0: This license allows others to download the works and share them with others as long as they credit the author, but they can’t change them in any way or use them commercially.
-
CC Attribution-NonCommercial 4.0: This license allows others to remix, tweak, and build upon the work non-commercially, as long as they credit the author, but they cannot use the material for commercial purposes.
-
CC Attribution-NonCommercial-ShareAlike 4.0: This allows others to remix, tweak, and build upon the work non-commercially, as long as they credit the author and license their new creations under identical terms.
These terms ensure that your article is freely accessible and usable, while respecting the rights of the original authors.
