Research On Dynamic Spectrum Allocation Of Aviation L-Band ATG System

The International Civil Aviation Organization has expanded the civil aviation communication band to the L-band and identified it as the future civil aviation broadband communication band in response to the fact that the existing very high frequency bands cannot meet the growing demand for civil aviation communications.China also protected the L-band(960-1164MHz)as a key frequency band of future civil aviation broadband communication by law,and a civil aviation ATG(Air-to-ground Communication System)will be deployed in this band.Currently,avionics systems such as DME(Distance Measuring Equipment)and secondary radar are deployed in the aeronautical L-band,and the ATG system can only be deployed on the spectral gap of these systems.If the discrete idle spectrum of these systems can be used reasonably,the capacity of the ATG system can be greatly improved.In aviation communication scenario,the rapid movement of aircraft makes the relative spatial position between aircraft and aircraft / base stations change rapidly,meanwhile,the spectrum occupation of the DME system is highly random,these two reasons cause the idle spectrum resources to be time-varied.This thesis unites the project "Development and test of air-to-ground broadband data link communication system",and aims at dynamic spectrum allocation of the L-band ATG system in the case where the available idle spectrum is discrete and time-varying.The main work and contributions of this thesis are as follows:(1)The estimation model of idle spectrum holes size is established for the coexistence scenario of L-band future ATG system L-DACS1 and DME system.According to the frequency points occupied by DME system are discrete and random,this thesis derive a probability distribution function by discrete order statistics to describe the size of DME system’s idle spectrum holes,and the back propagation neural network is used to establish the estimation model.The simulation results show that the distribution function derived can make a good estimation of the size of the idle spectrum hole.(2)Based on the estimation of idle spectrum holes,a stochastic chance constrained multi-knapsack model is established for dynamic spectrum allocation.The model is difficult to solve because it contains multiple optimization goals,multiple constraints and random variables.To deal with this problem,the model is changed into two sub-models,and it is solved by using a non-dominated genetic algorithm with elite strategy to obtain the suboptimal solution.According to the simulation results,we can know that the dynamic spectrum allocation algorithm proposed in this thesis can effectively improve the spectrum utilization and increase the access number of aircraft as well as system throughput in the scenario where the available spectrum resources has fast time-varying and discrete characteristics.