Research On Random Access Technology For 5G-Intergrated Satellite Communication Systems

Thanks to its irreplaceable superiority in providing seamless coverage,satellite communication has been regarded as an important part of the next-generation wireless networks.In recent years,both academic and industry have been vigorously pushing forward the integration of satellite and terrestrial networks,aiming at building a global network architecture based on the 5G ecosystem.However,there are great differences in channel characteristics and system architecture between the two systems,which makes the construction of the integrated system face great difficulties and challenges,especially some key technical problems of air interface still need to be solved,such as random access technology.Considering the channel characteristics of satellite communication with large frequency offset,large delay and large loss,this paper focuses on the design of random access preamble and detection algorithm suitable for 5G satellite mobile communication system based on different air port waveforms.The main contents and contributions are as follows:First,on the basis of the orthogonal frequency division multiplexing(OFDM)waveform that commonly used in the 5G system,this paper adopts the frequency-domain sequence repetition scheme to ensure the beam coverage performance of satellites.However,the cubic metric(CM)of the frequency-domain sequence repeating scheme is higher.In order to reduce the value of CM,we propose an enhanced phase shift-based preamble,and a genetic algorithm is utilized to search for the optimal phase set when the CM value of the proposed preamble is lowest.Further,a preamble based on a small-scale inverse discrete Fourier transform(IDFT)is proposed,which can reduce the CM value by reducing the scale of IDFT,and has low detection complexity.Moreover,we presented corresponding preamble detection schemes for the two preambles that can perform joint timing advance(TA)and carrier frequency offset(CFO)estimation.The results show that with the increase of the number of replicates,the proposed two schemes have better CM performance than the existing schemes,and have significantly improved detection and estimation performance.Second,in order to overcome the shortcomings of OFDM waveform,such as high outof-band radiation and sensitivity to synchronization error,we make use of the generalized frequency division multiplexing(GFDM)waveform as the air interface waveform of integrated satellite-terrestrial communication network.On this basis,the applicability of two classical preamble detection algorithms in GFDM systems is first studied,and we revealed the performance defects of classical algorithms in large frequency offset and low signal-tonoise ratio(SNR)environment.Further,we propose a preamble detection algorithm based on length-controllable dual ZC differential correlation by utilizing two ZC sequences and all relevant products generated at the local and receiver ends.The proposed method can not only eliminate the influence of frequency offset,but also improve the detection performance at low SNR by flexibly adjusting the correlation length.The results show that the proposed preamble detection algorithm is robust to large frequency offset,and its detection performance will be significantly improved with the increase of correlation length.