IoT environment traffic prediction-enabled network slicing and cooperative offloading method

By employing a collaborative optimization architecture with both long and short time scales, a lightweight Transformer model, and a multi-agent reinforcement learning algorithm in the IoT edge computing system, the problem of network slice resource configuration mismatch caused by the spatiotemporal dynamic fluctuations of user business needs is solved. This enables proactive pre-scheduling of resources and distributed collaborative offloading decisions, thereby improving the system's resource utilization efficiency and load balancing capabilities.

CN122421022APending Publication Date: 2026-07-17FUZHOU UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
FUZHOU UNIV
Filing Date
2026-05-11
Publication Date
2026-07-17

Smart Images

  • Figure CN122421022A_ABST
    Figure CN122421022A_ABST
Patent Text Reader

Abstract

本发明提供一种面向IoT环境流量预测使能的网络切片与协作卸载方法，将系统运行时间划分为长时隙和短时隙，每个长时隙包含预设数量的短时隙；在每个长时隙的起始阶段，各边缘服务器基于历史卸载请求数据，通过预训练的时序流量预测模型预测当前长时隙的卸载请求数量，并根据预测结果向中央服务器提交网络切片资源请求；中央服务器通过网络虚拟化技术为各边缘服务器分配逻辑隔离的网络切片，网络切片包含计算资源与通信资源；在每个短时隙内，各边缘服务器接收其覆盖范围内物联网设备包含计算需求与最大容忍时延的计算卸载任务请求，在当前长时隙已分配的网络切片资源约束下，采用多智能体强化学习算法执行分布式协作卸载决策。
Need to check novelty before this filing date? Find Prior Art