Research On Resource Allocation And Interference Cancellation Of Traffic-Driven Beam-Hopping Satellite System

Satellite communication systems play an increasingly crucial role in contemporary society due to their tremendous capacity,high speed and the ability to cover areas that cannot be covered by terrestrial communication equipment such as oceans,deserts and mountains.With the continuous and rapid growth of high-quality satellite multimedia services and the extension of human activities,the public has put forward higher requirements on the transmission efficiency,data transmission quality and coverage area of satellite communication systems,so it is an urgent need to improve and enhance the resource utilization efficiency and throughput of satellite communication system.Thanks to the application of multi-point beam and frequency-reuse technologies,high-throughput satellites can provide tens of times higher system capacity than traditional satellites.However,the "fragmented" resource allocation method of traditional high-throughput satellites leads to frequent "busy and uneven" beams,which causes a tremendous waste of resources and cannot give full play to the advantages of huge capacity of high-throughput satellites.Beam hopping technology has been widely concerned by domestic and foreign scholars due to its flexible resource allocation.However,the existing beam-hopping technology has some issues,such as the lack of traffic-driven model,the uneven traffic demand in different clusters caused by clustering method,and the potential cofrequency interference between beams.To address these issues,this thesis establishes the GEO satellite traffic demand model.Based on the traffic model,the dynamic clustering,resource-allocation and interference avoidance in the beam-hopping satellite communication system are studied in depth.At the same time,the beam-hopping technology demonstration system platform is designed to verify the research content.The main research contents and results are as follows:Spatio-temporal traffic model: Due to the flexibility of beam-hopping technology in resourceallocation,many domestic and foreign scholars focus on different types of beam-hopping resource allocation algorithms.These studies use the method of generating random numbers to represent the traffic demand,and do not establish the traffic model corresponding to the resource allocation algorithm.To counter this problem,this thesis proposes a modeling method of spatio-temporal model for beam-hopping satellite communication system,which describes the traffic characteristics from space and time dimensions.On the basis of the two main factors of joint economic development degree and population density to establish the traffic intensity demand model,the spatial traffic model is established by combining the grid division approach to quantitatively analyze the traffic in the spatial dimension.On the basis of human activity pattern,combined with time zone division,a normalized time weighting factor is defined for each time interval with the peak traffic demand as a reference,and a time traffic demand is established.The simulation results show that there is obvious variability in the spatial dimension of the traffic demand,which varies periodically with time,and the establishment of the traffic demand model has more practical guidance for the hopping beam technology in resource-allocation.Research on hopping beam resource allocation: The traditional beam-hopping satellite communication system allocates resource on the basis of uniform clustering.The uniform clustering ensures that the number of beams in each cluster is basically the same.However,due to the obvious spatial difference and time variability of the traffic demand of each beam,the traffic demand in each cluster is greatly different.This clustering method not only leads to the beam in different clusters in the state of "busy and idle uneven",but also needs to establish the objective function of power allocation between clusters.In addition,some of the existing intra-cluster resource allocation algorithms are complicated to solve,which will directly or indirectly affect the complexity of satellite systems.To address these problems,this thesis adopts the dynamic clustering method instead of the traditional uniform clustering method.Under the constraints of the satellite communication system payload,capacity and the maximum number of beams illuminated per time slot,the dynamic clustering method is combined with the beam location information.The clustering method makes the traffic demand in each cluster basically consistent,so the satellite power is evenly distributed to each cluster to complete the power distribution between clusters.On the basis of dynamic clustering,a beam hopping slot allocation algorithm based on fairness objective function is established,and the objective function is solved by the CVX toolbox.The traffic demand in the cluster after dynamic clustering and traditional clustering is simulated,and the time slot allocation results of the CVX toolbox and lagrangian function are compared.The simulation results show that the beam-hopping resource allocation based on dynamic clustering improves the utilization efficiency of system resources,reduces the computational complexity,and enables the beam-hopping satellite communication system to meet the requirements of on-demand coverage.Interference cancellation strategy for hopping beam system: Beam-hopping satellite communication system adopts full frequency-reuse to further improve the frequency utilization efficiency,when using the same frequency beams illuminated at the same time and their center distance is less than the same frequency-reuse distance,the problem of co-channel interference will arise.To address this problem,this thesis combines the beam-hopping pattern design and the precoding approach for interference cancellation research.Based on the beam-hopping time slot solution,the beam-hopping pattern design is carried out by "spatial isolation + time-sharing",considering all clusters jointly and combining the number of time slots of each beam.However,the beam-hopping pattern design cannot completely avoid the interference problem between beams.The beam interference search algorithm is used to search for beams with co-channel interference,and these beams are precoded for interference cancellation.The simulation results show that the beamhopping pattern design avoids the interference problem of some beams fundamentally by adjusting the order of the illuminated beams.For the other beams with interference,the precoding algorithm can effectively avoid interference and improve the system capacity.Simulation platform based on Orbit calculation tool +MATLAB: Based on the above contents,this thesis designs the software verification platform of beam-hopping system,which is a visual platform based on Orbit calculation tool +MATLAB,and integrates the modules of scenario establishment,resource-allocation,beam-hopping demonstration and performance analysis of beamhopping satellite communication system.The software platform can be used to simulate the process of beam-hopping system,and the modules in it can be added and/or deleted to provide better visualization according to different scenario requirements.