Research On Information Age Minimization Strategy In Backscatter-assisted Wireless Power Supply Communication Networ

With the 5th generation mobile communication technology(5G)entering the stage of large-scale commercial,emerging real-time applications such as the Internet of Vehicles,virtual reality,health monitoring and industrial Internet are emerging.Such emerging applications usually need to rely on a large number of wireless devices’ real-time perception and monitoring of the surrounding physical environment information to output intelligent control decisions,and the accuracy of decisions highly depends on the freshness of the received information.Therefore,they have extremely strict requirements for the timeliness and reliability of information.However,the lack of traditional performance indicators(such as throughput and delay)in measuring the timeliness of information and the inherent resource constraints of wireless devices(such as limited battery capacity and low computing power)seriously hinder the further development of emerging applications.Therefore,how to capture the freshness of system information and improve the timeliness of system delivery information under the scenario of limited equipment energy and computing resources is an urgent problem to be solved in the development of emerging applications.Recently,Age of Information(AoI),as a new indicator that can accurately capture the freshness of state information,has been widely concerned by scholars at home and abroad.Based on the wireless power supply communication network architecture assisted by backscatter,the thesis takes AoI as a measure of information freshness,and studies the problem of timely acquisition of state information in real-time perception of the Internet of Things.Firstly,considering the time-varying of wireless fading channel and the randomness of equipment energy collection,a joint sample and hybrid backscatter communication update strategy is studied in the single-source real-time sensing Internet of Things.Secondly,considering the impact of data processing and information transmission on the timeliness of information,a joint binary offload-hybrid backscatter communication update strategy is studied in a multi-source real-time sensing Io T system with time-varying channels.The main contributions and innovations of the thesis are as follows:(1)For the energy-constrained single-source real-time sensing Io T scenario,a strategy of joint sampling and hybrid backscatter communication update is designed to study the problem of minimizing the long-term average AoI at the destination.Specifically,firstly,considering the time-varying channel conditions and the randomness of equipment energy collection,the problem of minimizing the long-term average AoI at the destination is modeled as a stochastic optimization problem.Secondly,in order to solve the problem,the system optimization problem is transformed into an average cost Markov decision process(MDP)problem.Then,when the dynamic information of the environment is known,the optimal strategy is obtained through the iterative algorithm of correlation value.In the absence of environmental dynamic information,the Q learning algorithm and exploration and utilization method are used to learn the optimal strategy through trial-and-error interaction with the environment.Finally,the simulation results show that compared with the two reference strategies,the proposed strategy significantly improves the AoI performance of the system.At the same time,it is found that the AoI performance of the system increases with the reduction of the size of the update package or the increase of the battery capacity.(2)Aiming at the multi-source and multi-user real-time perception of the Internet of Things scene with limited energy and computing resources,aiming at minimizing the longterm average on-demand AoI per user of the system,a joint binary offload and hybrid backscatter communication update strategy is designed by considering the impact of data processing and information transmission process on information timeliness.Firstly,the minimum system’s long-term average on-demand AoI per user is modeled as a stochastic optimization problem by considering the equipment energy consumption constraints,the unloading and updating mode constraints,and the mode operation parameter constraints.Secondly,in order to solve the problem,it is transformed into a two-level discrete-time MDP problem.Then,the optimal unloading and updating strategy is found by using the two-level deep reinforcement learning algorithm,in which the unloading network runs based on the time frame,while the updating network runs based on the time slot.Finally,the simulation results show that the proposed joint binary offload-hybrid backscatter communication update strategy is significantly better than the benchmark strategy.