Research On Intelligent Control Technology Of Air Network In Space-air-ground Integrated IoT Network Based On Age Of Information

As the Internet of Things(Io T)develops and the concept of interconnection of everything emerges,people’s lives have become closely intertwined with the network.Io T technology has infiltrated various fields from smart home to smart cities and from smart manufacturing to intelligent transportation.However,the traditional ground-based cellular networks suffer from limited coverage,weak signals,and susceptibility to interference,which poses even greater challenges for ground communication,especially in remote mountainous areas and other regions.Therefore,building an efficient and flexible space-airground integrated network(SAGIN)has become a research focus.The primary goal of a SAGIN is to achieve efficient coordination among satellite networks,airborne relay networks,and ground mobile networks,thereby meeting people’s communication security requirements.This article utilizes age of information(Ao I)to measure the freshness of data in a SAGIN and employs deep reinforcement learning(DRL)methods to reduce system delays and configuration costs,thus meeting communication service requirements in remote and wide coverage areas.This dissertation focuses on enhancing the overall system timeliness,optimizing unmanned aerial vehicles(UAV)trajectory,and minimizing network configuration cost in SAGIN.To achieve these objectives,we investigate the following areas primarily:Firstly,this article introduces the concept of Ao I as a metric for measuring data freshness in SAGIN,and proposes Ao I models for different network scenarios.While the existing Ao I calculation methods are suitable for conventional SAGIN scenarios,they often fail to accurately measure data freshness in more complex and larger-scale scenarios.To address this issue,this dissertation proposes a power-penalty mechanism that incorporates power penalty into Ao I calculation to reduce the impact of low freshness data on overall system latency and improve the learning efficiency of UAV.Simulation results demonstrate that the power-penalty Ao I effectively avoids the issue of excessively high Ao I and outperforms other benchmark algorithms.The second contribution of this dissertation is a novel dynamic trajectory optimization strategy for multiple UAVs.For complex and dynamic scenarios in SAGIN,this work investigates how to ensure data timeliness,improve UAV energy efficiency,and enhance the cooperation among UAVs.The optimization problem is formulated as a Markov decision process and solved using DRL algorithms.To reduce the complexity of the algorithm,the matching theory is introduced,and a stable matching algorithm is used to pair UAVs with Io T sensing devices.Finally,the proposed algorithm is validated through simulation results,which demonstrates that it reduces data transmission latency and improves UAV energy efficiency.Thirdly,a cost optimization strategy for configuration is proposed based on the SAGIN.The SAGIN often requires a large number of network devices to ensure connection stability,resulting in extremely high system configuration costs.To address this issue,a configuration cost optimization strategy based on the soft actor-critic algorithm and the constrained Markov decision process is proposed.In the first step,multiple UAV are clustered,and trajectory planning is carried out using DRL.Each cluster of UAVs is managed by a highaltitude access platform,and then the height and number of UAVs and high-altitude access platforms are optimized to expand the coverage area while maintaining transmission stability and unchanged timeliness Finally,the optimal cost per unit area is derived.Simulation results confirm the convergence and effectiveness of the algorithm.The above research proposes an Ao I measurement method for data freshness in the SAGIN.It also designs optimization schemes based on DRL to address the issues of UAV trajectory and network configuration cost,which improves the timeliness of the SAGIN,reduces network configuration costs,and provids a solution for the engineering application of the SAGIN.