基于深度强化学习的物联网传输数据实时调度方法  

Real Time Scheduling Method of Internet of Things Transmission Data Based on Deep Reinforcement Learning

作　　者：张继飞[1] 赵玉超[1] 张春红 ZHANG Jifei;ZHAO Yuchao;ZHANG Chunhong(Xiamen Ocean Vocational College,Xiamen 361000,China;Xiamen Institute of Technology,Xiamen 361000,China)

机构地区：[1]厦门海洋职业技术学院,福建厦门361000 [2]厦门工学院,福建厦门361000

出　　处：《通信电源技术》2021年第3期111-113,共3页Telecom Power Technology

基　　金：厦门市海洋与渔业发展专项资金项目“水产养殖自动化装备核心技术攻关与产业化示范”(19CZB015HJ02);福建省海洋经济发展补助资金项目(ZHHY-2019-4);福建省中青年教师教育科研项目(科技)“GPRS远程测控系统研发”(JA15749)。

摘　　要：针对传统物联网传输数据实时调度方法的调度时间长,存在数据缺失的现象,提出了基于深度强化学习的物联网传输数据实时调度方法。采用深度强化学习采集物联网传输数据的特征序列样本,通过计算物联网传输数据的自适应转发特征量,建立了数据调度的频域均衡控制模型。根据模型判定物联网传输数据安全性,采用深度强化学习判别了物联网传输数据安全性。将最小丢包率作为物联网传输数据的调度目标,引入深度强化学习的贡献思想,构建了物联网传输数据实时调度模型,实现了物联网传输数据的实时调度。实验结果表明,基于深度强化学习的物联网传输数据实时调度方法在调度时间和数据完整性方面具有更好的性能。For the traditional real-time scheduling method of Internet of Things transmission data scheduling time is long,there is a lack of data phenomenon.A real-time scheduling method based on deep reinforcement learning is proposed.Deep reinforcement learning is used to collect the characteristic sequence samples of the data transmitted by the Internet of Things.By calculating the adaptive forwarding characteristic of data transmitted in the Internet of Things,the frequency domain equalization control model of data scheduling is established.According to the model to determine the security of data transmission in the Internet of Things.Deep reinforcement learning is used to identify the security of data transmission in the Internet of Things.The minimum packet loss rate is taken as the scheduling target of data transmission in the Internet of Things.The contribution idea of deep reinforcement learning is introduced,and the realtime scheduling model of Internet of Things transmission data is constructed.Real time scheduling of data transmission in Internet of Things is realized.The experimental results show that the real-time scheduling method based on deep reinforcement learning has better performance in scheduling time and data integrity.

关 键 词：深度强化学习 物联网 传输数据 实时调度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TN929.5[自动化与计算机技术—控制科学与工程] TP391.44[电子电信—通信与信息系统]