Font Size: a A A

Research On Spectrum Scheduling Method Of Space TT&C Network Based On SRNN Algorithm

Posted on:2022-02-19Degree:MasterType:Thesis
Country:ChinaCandidate:Y XiaoFull Text:PDF
GTID:2492306341951929Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
As we all know,the spectrum resources in nature are limited.Besides,with the rapid development of aerospace industry,the spectrum resources are more and more tense.In the transmission from satellite to ground,the signal of satellite can be received by several ground stations and receiving ships on the ground.The environment of these receiving equipment is different,and the order of receiving signal of equipment will affect the resource utilization of this frequency band.Therefore,how to reasonably and efficiently schedule the spectrum resources is an important problem in the field of Aerospace TT&C.This thesis studies the management and scheduling of wireless resources in the single-channel and multi-channel scenarios of space TT&C system.This thesis proposes a spectrum recognition strategy based on sliced recurrent neural networks.Firstly,the channel transmission process of space communication model Loo model and Corazza model is simulated.The relevant data is collected and encoded by One-Hot.Then the data are input into the sliced recurrent neural networks so that the sliced recurrent neural networks learn the data features and identify the main performance parameters of each band signal.Finally,the simulation of RNN,SRNN(4,2)and SRNN(4,3)is carried out.Experimental results show that SRNN(4,3)is the fastest processing speed.This thesis proposes a spectrum scheduling algorithm for sliced recurrent neural networks-chaotic Random-Key discrete differential artificial bee colony.Firstly,the S-band signal is extracted by spectrum identification technology.Chaotic distribution is used to generate the position of ground station or receiving ship.Each iteration time differential evolution algorithm works in parallel with artificial bee colony to obtain the local optimal solution.Differential evolution algorithm is used to design an adjustable selection operator which can calculate the probability of honey source being selected.The Random-Key discretization method is used to adjust the neighborhood position of differential evolution algorithm and bee colony algorithm.The spectrum scheduling algorithm solves the optimal scheduling problem of space TT&C system under single-channel and multi-channel.Experimental results show that the algorithm can improve Jain fairness index and throughput in single-channel and multi-channel scenarios.The spectrum scheduling algorithm proposed in this thesis can achieve the best scheduling scheme of single-channel and multi-channel under the influence of various environmental factors,which provides a good idea for the future research of spectrum resource management and scheduling in the downlink band of space TT&C network.
Keywords/Search Tags:sliced recurrent neural networks, discrete artificial bee colony algorithm, spectrum identification, spectrum scheduling
PDF Full Text Request
Related items