首页|业务驱动的低轨卫星物联网终端模态切换方法

业务驱动的低轨卫星物联网终端模态切换方法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
针对低轨卫星物联网场景下基于窄带物联网(narrow band Internet of Things，NB-IoT)体制的物联终端大尺度地理范围内多场景应用业务时延和终端功耗需求动态变化问题，提出一种利用马尔可夫链模型评估NB-IoT 终端在扩展不连续接收(extended discontinuous reception，eDRX)和节能模式(power saving mode，PSM)下的时延功耗的方法，建立了以下行业务延迟和终端功耗为优化目标的多目标优化问题。在信关站利用终端历史业务数据信息离线训练基于支持向量机(support vector machine，SVM)的时延功耗的回归预测模型，以回归预测模型作为非支配排序遗传算法(non-dominated sorting genetic algorithms-Ⅱ，NSGA-Ⅱ)的目标函数，得到多目标优化问题的Pareto前沿解集，进一步从Pareto前沿解集中选择满足当前应用时延功耗需求的工作状态定时器参数值，在线配置终端。仿真结果表明，相比于传统的地面物联网终端固定式定时器参数配置方法，所提出的业务驱动的定时器参数配置方法在终端动态多场景应用下能够更好地满足业务时延和终端功耗需求。
Business-driven terminal mode switching method for low Earth orbit satellite-based Internet of Things
To address the dynamic variation issue in business latency and terminal power consumption requirement for narrow band Internet of Things(NB-IoT)regime in large-scale geographic and diverse scenario applications within the context of low-Earth-orbit satellite Internet of Things,a method that utilizes a Markov chain model to assess the latency and power consumption of NB-IoT terminals during extended discontinuous reception(eDRX)and power saving mode(PSM)is proposed.The approach establishes a multi-objective optimization problem with downlink business latency and terminal power consumption as optimization targets.Offline training of a regression prediction model based on support vector machine(SVM)is conducted using terminal historical business data information at the gateway.This regression model is then employed as the objective function for the non-dominated sorting genetic algorithm-Ⅱ(NSGA-Ⅱ)to obtain the Pareto frontier solution set for the multi-objective optimization problem.Subsequently,selecting values for the work state timer parameters from the Pareto frontier solution set that meets the current application's latency and power consumption requirements,online configuration of the terminal is performed.Simulation results demonstrate that compared to traditional ground-based IoT terminal fixed timer parameter configuration methods,the proposed business-driven timer parameter configuration method can better fulfill business latency and terminal power consumption requirements in dynamic multi-scenario applications for terminals.

low-Earth-orbit satellite Internet of Thingsnarrow band Internet-of-Things(NB-IoT)Markov chainlatency and power consumptionmulti-objective optimization

洪涛、王凡、李治、钟志伟、丁晓进、刘子威、张更新

展开 >

南京邮电大学通信与信息工程学院,江苏南京 210003

低轨卫星物联网 窄带物联网 马尔可夫链 时延功耗 多目标优化

江苏省前沿引领技术基础研究专项国家自然科学基金国家自然科学基金国家自然科学基金科技部项目

BK20192002U21A2045062171234619714402022YFB2902600

2024

10.12305/j.issn.1001-506X.2024.08.34

系统工程与电子技术
中国航天科工防御技术研究院 中国宇航学会 中国系统工程学会

系统工程与电子技术

CSTPCD北大核心
影响因子：0.847
ISSN：1001-506X
年,卷(期)：2024.46(8)