Abstract
Currently, edge AI is used in critical infrastructure for anomaly detection, state estimation, predictive maintenance, and emergency response. However, if the scheduler treats network attacks and computational failures as irrelevant events, it can lead to situations where minimizing latency results in unreliable outputs, delayed timing, and unusability in real-world operations. Therefore, this paper proposes a joint security-aware and fault-aware scheduling framework and develops a phased evaluation scheme. This framework combines policy-based eligibility screening, risk-sensitive ranking, selective and diversity-aware replication, bounded migration, result verification, and authorized model degradation. This allows it to improve continuity without permanently achieving full redundancy.
References
[1] Bi, S., Yuan, X., Hu, S., Li, K., Ni, W., Hossain, E., & Wang, X. (2024). Failure analysis in next-generation critical cellular communication infrastructures. arXiv preprint arXiv:2402.04448.
[2] Hao, Z. (2026). Low-Overhead Scheduling for Real-Time AI Workloads on Multi-Core Edge Chips. International Journal of Advance in Applied Science Research, 5(3), 15-25.
[3] Zhukabayeva, T., Zholshiyeva, L., Karabayev, N., Khan, S., & Alnazzawi, N. (2025). Cybersecurity solutions for industrial internet of things–edge computing integration: Challenges, threats, and future directions. Sensors, 25(1), 213.
[4] Curtis, M., & Price, C. (2026). Edge AI for Real-Time Blackout Prevention in Robot Power Networks. Available at SSRN 6953398.
[5] Hao, Z. (2026). Dynamic Task Prioritization for Edge AI in Smart Cities: Balancing Latency and Energy Efficiency. Journal of Intelligence and Engineering Technology, 1(1), 60-69.
[6] Zhang, H., Guo, J., Li, K., Zhang, Y., & Zhao, Y. (2024). Offline Signature Verification Based on Feature Disentangling Aided Variational Autoencoder. arXiv E-Prints. arXiv preprint arXiv:2409.19754.
[7] Ren, L. (2025). Real-time threat identification systems for financial api attacks under federated learning framework. Academic Journal of Business & Management, 7(10), 65-71.
[8] Ren, L. (2025). Reinforcement learning for prioritizing anti-money laundering case reviews based on dynamic risk assessment. Journal of Economic Theory and Business Management, 2(5), 1-6.
[9] Hao, Z. (2025). Task Affinity-Aware Scheduling for Multi-Core Edge Devices in Autonomous Vehicles. Engineering Frontiers, 1(2).
[10] Hao, Z. (2026). Structure-Aware Deep Reinforcement Learning for Latency-Minimal Scheduling of Edge AI Inference on Heterogeneous Cores. Journal of Intelligence and Engineering Technology, 1(1), 50-59.
[11] Liu, Y. (2026). Heterogeneous Resource Slot Optimization in Multi-Dimensional Recommendation Landscapes: A Submodular Constrained Framework with Cross-Space Spillover Effects. Journal of Progress in Engineering and Physical Science, 5(1), 26-31.
[12] Shengtao, L. (2025). Machine Learning-Based Logistics Network Optimization Algorithm. Academic Journal of Computing & Inform0ation Science, 8(5), 46-54.
[13] Huang, S. (2025). Real-time adaptive dispatch algorithm for dynamic vehicle routing with time-varying demand. Academic Journal of Computing & Information Science, 8(9), 108-118.
[14] Liu, Y. (2026). Cascading Resilience Through Predictive Multi-Dimensional Safeguards: System Stability Architecture for Billion-Scale Concurrent Platforms. Innovation in Science and Technology, 5(1), 35-45.
[15] Liu, X. (2025). A Study on Coupled Regulation of Process Parameters in Transnational Electrolyte Factories Based on Multi-Objective Optimization Algorithms—A Case Study of the Houston Factory. Journal of Progress in Engineering and Physical Science, 4(6), 5-14.
[16] Xinshun, L. (2026). Research on Technical Specification for Collaborative Optimization of Energy Efficiency and VOC in Continuous Chemical Production. Journal of Academic Research and Advances, 2(1), 41-51.
[17] Liu, X. (2025). Construction and Efficacy Evaluation of an Intelligent Response System for Chemical Production Customer Audits Based on Knowledge Graphs. Innovation in Science and Technology, 4(10), 22-28.
[18] Meng, S. (2026). Bridging Research and Market Adoption in Artificial Intelligence: an Investment-Driven Framework for Commercializing AI Security Technologies. Academic Journal of Sociology and Management, 4(3), 12-19.
[19] Meng, S. (2026). Accelerating Commercial Space Technology Commercialization Through Government-Guided Industrial Investment Funds: a Case Study of CAS Space and China's Emerging Aerospace Ecosystem. Journal of Economic Theory and Business Management, 3(2), 1-5.
[20] Hao, Z. (2025). Fault-Tolerant Real-Time Scheduling for Edge AI in US Critical Infrastructure. Engineering Frontiers, 1(4).
[21] Koteswaramma, R., & Shaik, M. A. (2026). A novel Artificial Intelligence method for Secure and Fault-Resilient Adaptive Scheduling based on Industrial Internet of Healthcare Things. Advances in Data Science and Adaptive Analysis.
[22] Joseph, A. (2026). Optimizing Resource Allocation and Fault Tolerance in Cloud-Based Artificial Intelligence Workloads.
[23] Hao, Z., Yin, M., Xu, J., Liu, Z., & Chen, Y. (2026, March). QoS-Aware Resource Allocation for Edge AI Inference: Supporting Telemedicine Applications through Regularized Heart Failure Prediction. In 2026 IEEE 8th International Conference on Communications, Information System and Computer Engineering (CISCE) (pp. 477-480). IEEE.
[24] Hao, Z. (2026). Energy Efficient Multi Core Task Scheduling for Real Time Edge AI Systems: A Latency Aware Approach. International Journal of Advance in Applied Science Research, 5(3), 1-14.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2026 International Journal of Frontiers of Modern Synthesis