Research On Efficient Radio Resource Control And Distributed Acceleration Simulation For Satellite Internet

With the continuous development of 5G technology in the direction of B5G(Beyond 5G)and 6G,users’ demand for ubiquitous and anytime connection services is also becoming stronger and stronger.Facing the problem of insufficient coverage of the ground mobile communication system at the present stage,the non-terrestrial network NTN is considered to be an effective solution.3GPP has started to promote the 5G NR standard to the NTN scenario,but since most of the original protocol standards are designed based on the terrestrial network scenario,it needs to be re-optimized and revised under the satellite.As the core control layer of the 5G access network,the radio resource control RRC controls all layers of the entire access network protocol stack and determines the performance of the access network,which needs to be studied intensively.At the same time,in view of the limited resources of the low-orbit satellite communication system,the beam hopping system is an effective means to improve the efficiency of resource allocation due to its beam scheduling flexibility,but how to efficiently allocate multiple resources such as time,space and frequency,which also brings new challenges to the radio resource control technology.To sum up,this paper conducts related research on the radio resource control and radio resource scheduling schemes under the beam-hopping communication system,and models the stability of the radio resource control connection and the radio resource scheduling method with the separation of the control plane and the user plane,analyzed and completed the following tasks:First of all,aiming at the state scheduling problem of radio resource control in the beam-hopping communication system,a radio resource control method based on the improved network control system is proposed,and the RRC state scheduling problem of large-scale instantaneous access of users in the beam-hopping scenario is proposed.This model is a communication network resource scheduling problem in resourceconstrained networked control systems.On the basis of the traditional model,cross-layer perception and tolerable upper and lower bounds are introduced,and the optimal scheduling method of network resources under the network control system is deduced to characterize the optimal RRC scheduling strategy in the satellite communication scenario.Finally,the optimal scheduling strategy of this model is compared with the traditional network control system model and the improved model method under twostate and three-state satellites respectively,and it is proved that the connection stability under this model has reached the optimum.And use the model in this paper to analyze the adaptability of satellite users using two-state and three-state scenarios.Secondly,under the network architecture where the NR user plane and the control plane are separated,the beam-hopping communication system encounters the secondary resource allocation strategy of beams and service beams.This paper aims at the actual needs of each communication scenario,and establishes a traffic arrival model and design The optimization goal is set,and the control plane goal is modeled as the minimum weighted delay,and the user plane goal is modeled as the throughput maximization.And based on the mathematical analysis of the above problems,an improved salp algorithm is proposed to solve it.Finally,the simulation proves that the algorithm proposed in this paper has excellent convergence speed and robustness in the strategy of beam-hopping secondary resource scheduling.With the increase of traffic arrival,the throughput can be guaranteed under the premise of controlling the control plane delay.The improvement of this paper is better than the traditional algorithm on the problem of this paper.Finally,in order to verify the wireless resource scheduling strategy proposed in this paper,a system-level simulation platform based on OPNET and STK was built.The satellite constellation was designed in STK,and the 5G communication protocol stack and resource scheduling algorithm for NTN were designed in OPNET.The design is carried out,and through the combination of the two,the network performance simulation under the large-scale constellation topology is jointly realized.However,the above-mentioned large-scale satellite-ground joint network simulation scenario has many nodes and a high rate of change in computing load,resulting in low simulation efficiency.In response to this problem,this paper proposes a container-based parallel acceleration simulation framework,which divides the simulation scenarios and loads them into different simulation node containers,and improves the simulation efficiency in a parallel manner.Aiming at the problem of unbalanced load caused by high load change rate of computing resources in highly dynamic simulation scenarios,an elastic scaling container scheduling scheme is proposed to match node service capabilities with computing load and improve simulation efficiency.