| Recent years have witnessed a runaway rise of Internet of Things(IoT),which puts forward higher requirements for efficient and flexible wireless networking.Nevertheless,it is not practical to deal with the access needs of remote areas and processing tasks of sudden traffic due to the fixed deployment location and limited coverage of traditional communication infrastructures.Owing to the flexible deployment and low cost,Unmanned aerial vehicles(UAVs)emerge as a promising technology to provide wireless services,as well as a key component of the future IoT systems.In the emerging IoT paradigm,the freshness of sensory information plays a crucial role in real-time data-analyzing and accurate application-level decision-making.As the performance metric of information freshness,the age of information(AoI)depends on the efficiency of the status update data transmission and calculation processing.Therefore,it is critical to guarantee the stable power supply and real-time efficient computing performance of IoT devices.However,in practical IoT systems,for most devices with limited battery capacity and computation resources,it’s difficult to meet the needs of delay-sensitive and computing-intensive tasks.Based on the above resource bottlenecks encountered by IoT devices in AoI optimization,the main research work and innovations of this paper are summarized as follows:(1)Considering that the limited energy supply of low-power devices makes it difficult to ensure the continuous and stable data transmission,this thesis proposes a UAVs-driven wireless power transmission(WPT)mechanism,where UAVs are deployed to wirelessly charge IoT devices,which can prolong the battery life of devices and enhance the sustainable performance of the IoT systems.In addition,to explore the influence of dynamic time-varying channels on the efficiency of energy transmission and data acquisition,a practical line-of-sight(LoS)/NLoS channel model is established to precisely capture the dynamic channel characteristics.Furthermore,a novel deep reinforcement learning-based proactive UAV trajectory planning algorithm is proposed to adapt to the dynamic channel to achieve the optimal system-level AoI.Extensive simulation results demonstrate that the proposed algorithm can significantly reduce the average AoI by approximately 20%to 65%compared to three other existing trajectory planning algorithms.(2)Considering that the limited computation resources of devices make it difficult to achieve efficient and timely data processing,this thesis proposes a UAVs-driven mobile edge computing(MEC)scheme,where IoT devices are allowed to offload their computation-intensive tasks to the edge servers carried by UAVs.which can relieve the computational pressure of IoT devices and improve the computing performance of the IoT system.Additionally,considering the energy limitation of UAVs,a calculation cost model in the MEC system is established to jointly optimize the calculation energy consumption of the UAV and the AoI performance of IoT devices,so as to realize the energy efficient calculation assistance.Then,by adjusting the weight coefficients of energy consumption cost and AoI cost,diversified computing service demand can be fulfilled.Simulation results are given to show the effectiveness of the proposed mechanism in the computing assistance and verify the correctness of AoI theoretical analysis in the MEC system. |
