CRDSA Based Random Access And Control Scheme In Satellite Network

Satellite is one of the key solutions to realize remote communications.In the future,Satellite will play an important role in the space information network.Moreover,the Internet of things has been increasingly improving our life,therefore,a more secure,reliable and scalable Internet architecture is required to ensure the devices can contact with each other at anytime and anywhere.First of all,the problems of throughput deteriorates,collision increases,and access delay prolongs caused by massive Machine type communication(MTC,also known as the Machine to Machine,M2M)terminals random accessing in the satellite communication system need to be solved.In recent years,the random access technologies of the satellite have been increasingly improved.Although the current researches have carried out a multi-angle discussion on the problems of random access,such as improving peak throughput and reducing peak-toaverage power ratio.To the best of our knowledge,there are few literatures fully consider the characteristics of large capacity,non-linearity and large delay in the background of the Internet of things.Thus,this thesis employs the CRDSA(Contention Resolution Diversity Slotted ALOHA)protocol which has been adopted in the DVB-RCS2 standard as reference.and systematically solves the problems of network overload and high collision caused by massive MTC terminals random accessing.This thesis mainly focuses on two aspects,that is,increasing peak throughput and implementing congestion control.The main innovation points and research works of this thesis are as follows:1)Multi-slot and Multi-user Joint Detection(MMJD)algorithm is proposed to improve the peak throughput of CRDSA protocol.Firstly,the defects of CRDSA is fully analyzed,and we point out the disadvantage of loop problem in the protocol,which limits the performance of throughput and packet loss(PLR).Then the existing conflict resolution schemes are introduced to solve CRDSA loop problem,and the performance analysis is given by simulation results.The MMJD demodulation algorithm is proposed to solve the mentioned problems.The algorithm analyzes the signal model of the conflict packets on the loop,and gives out the concrete process flow.Finally,simulation results prove that the proposed MMJD algorithm effectively solves the existing loop problem of CRDSA,and improves the peak throughput and PLR performance.2)The Least Square Support Vector Machine(LS-SVM)algorithm is proposed to improve the overload performance of CRDSA protocol.First,we analyze the defects when applying CRDSA protocol to M2 M communication.and point out that there are serious overload and large access delay problems in the protocol,which results in deterioration of system throughput and infinite of access delay.The traditional back-off algorithm lacks the precognition and always chooses to back off only after the packet collisions happens,thus,introduces large access delay and is unable to satisfy the requirement of the applications in M2 M.Then,some existing prediction algorithms with the ability to predict are introduced,and the drawbacks of the algorithm is fully analyzed.The Least Square Support Vector Machine(LS-SVM)algorithm which employs the SVM regression theory is proposed to solve the existing problems,and We analyze the model of CRDSA random access control scheme and the specific prediction process.Finally,the system-level simulation is performed,that is to say,we combine the proposed MMJD demodulation algorithm and LS-SVM prediction algorithm.The performance of throughput,packet loss rate(PLR),access delay and normalized mean square error(NMSE)is simulated,with the consideration of large delay,heavy traffic load,and nonlinear channel in satellite M2 M communication scenarios,the simulation results prove the superiority of the proposed algorithm.