DT)通过对物理实体的同步模拟,可有效提升配电网实时决策能力,但其对数据与算力的需求很高,边缘计算可下沉模型训练任务至边缘侧,减小DT模型训练的通信、计算负担。针对边缘侧资源与训练需求不适配的问题,提出一种面向数字孪生配电网的低时延业务决策方法。首先,构建配电网低时延业务决策框架,引入边缘协作提升配电网DT模型训练精度与效率,建立最小化模型训练时延与损失加权和的优化问题。其次,提出基于双向贪婪演进的边缘协作决策算法,利用Q学习实现全局信息未知情况下模型训练时效性和精度的联合优化,并基于贪婪策略解决协作服务器选择冲突问题。仿真结果表明,与基于强化学习的业务决策(reinforcement learning-based service decision
Lu Yunlong,Maharjan Sabita,Zhang Yan.Adaptive Edge Association for Wireless Digital Twin Networks in 6G[J].IEEE INTERNET OF THINGS JOURNAL,2021.
Lu Yunlong,Huang Xiaohong,Zhang Ke,Maharjan Sabita,Zhang Yan.Low-Latency Federated Learning and Blockchain for Edge Association in Digital Twin Empowered 6G Networks[J].IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,2021.
Chuan Zhou,Peng Zhang 0001,Wenyu Zang,Li Guo 0001.On the Upper Bounds of Spread for Greedy Algorithms in Social Network Influence Maximization.[J].IEEE Trans. Knowl. Data Eng.,2015.