Research On Task Offloading And Resource Scheduling Technology Of Space-Air Network

With the advantages of wide coverage,high destruction resistance,flexible configuration and multiple communication link options,the space-air network can provide high-quality services for users with lowlatency and high-stability.Hence,ground users can accomplish more complex tasks by offloading data to the space-air network.Space-air network-assisted communication services have many advantages,but there are also problems such as energy limitations,data leakage from wireless transmissions,and mismatch between supply and demand of communication resources.Therefore,it is important to study the use of task offloading and resource scheduling technology to improve the quality of communication services in the space-air network scenario.In this paper,the following researches are conducted for task offloading and data backhaul of ground users in the space-air network.First,in order to solve the energy consumption problem when ground users offload data to the unmanned aerial vehicle(UAV)airbased edge computing network,a secure and energy-efficient task offloading strategy based on air-ground collaborative multi-dimensional resource allocation is proposed.Under the premise of secure transmission,the limitations of system delay,power,secrecy interruption probability,computational capability,and UAV transmission capability during data offloading are fully considered,and a pairwise stability theory improved heuristic search algorithm is used to optimize the matching scheme between users and UAVs,while the successive convex approximation(SCA)algorithm jointly optimizes the air-ground communication and computational resource allocation.The two algorithms are combined for iterative optimization,and finally a user matching and resource allocation scheme that minimizes the total energy consumption of the UAV system is obtained.Finally,it can obtain the user matching and resource allocation scheme that minimizes system energy consumption.The simulation results show that this strategy effectively reduces the total energy consumption of the UAV air-based system while ensuring data security.Second,in order to solve the communication resource supplydemand mismatch problem when UAV nodes carry out ground user data backhaul through space-based networks,a beam-level resource adaptation strategy based on service demand prediction is proposed in this paper.First,the hidden markov model(HMM)is used to predict the service demand of each beam at the next moment.Based on the predicted values,the bandwidth,carrier frequency and power is dynamically and jointly pre-allocated to each beam by deep reinforcement learning(DRL)algorithm to achieve resource adaptation to the actual demand.The simulation results show that this strategy effectively reduces the waste of resources,while reduces the cofrequency interference between beams and improves the system overall level of signal to interference plus noise ratio(SINR).