Research On Signal Equalization Algorithm For Mode Division Multiplexing System Based On Few Mode Fiber

With the sophisticated communications unceasing spring up such as big data,5G and cloud computing,the business of communications is fast-growing,and the flow of Internet increase fleetly.As the main transmission medium for various and complex information in today's information society,the nonlinear impact of single-mode fiber(SMF)makes its transmittability closer to the Shannon limit.The mode division multiplexing(MDM)system based on few mode fiber(FMF)uses mutually independent and orthogonal modes in the FMF as signal transmission channels,thereby doubling the capacity of the communication system.However,when the FMF is transmitting signals,it will encounter mode coupling(MC),differential mode group delay(DMGD)and mode dependent loss(MDL).These impairments deteriorate the quality of signal transmission and increase the difficulty of digital signal processing(DSP)at the receiving end.Especially with the increase of transmission distance,DMGD and MDL ceaselessly accumulate,conventional signal equalization algorithms cannot recover the source signal to normal.In order to compensate for DMGD,many people use OFDM modulation at the transmitting end.However,the conventional OFDM technology has some shortcomings,such as the use of cyclic prefix(CP),greatly reducing the transmission efficiency of the system;serious out of band leakage,sensitive to frequency offset awfully,has inadaptate the requirements of current communication.To solve the above problems,after studying the mechanism of the above multiple impairments,we build a 6×6 MDM simulation system based on FMF.And two methods are used to compensate the DMGD and MDL respectively.The specific work of this paper is as follows:(1)Analyze and derive the mode characteristics of the FMF,and testify its mode orthogonality;study the three damage mechanisms of DMGD,MC and MDL in FMF.On this basis,taking advantage of the matrix transmission model,a 6×6 MDM simulation system based on FMF is set up on the VPI platform,and the function of the embedded DSP algorithm is realized in conjunction with Matlab.(2)An improved FD-LMS demultiplexing algorithm is put forward to solve the problem that the conventional FD-LMS algorithm will lose efficacy with the increase of DMGD when the algorithm is used for demultiplexing of the 6×6 MDM system.Firstly,the tenets of time domain equalization algorithm and frequency domain equalization algorithm are analyzed.Then,TD-LMS and FD-LMS are applied to the established MDM simulation system as their representative algorithms.The demultiplexing performance and computation complexity of the two algorithms are compared.Finally,the modified FD-LMS algorithm is put forward.When the transmission distance is 80km,the coupling coefficient is 0.1,the G_offset is 6d B,and the DMGD is 50ps/km,the modified FD-LMS algorithm saves 1.6d B of OSNR cost compared with the conventional FD-LMS algorithm on the premise of almost no increase in computational complexity.(3)In view of the problems of low system transmission efficiency,sensitivity to frequency offset,and serious out-of-band leakage in conventional OFDM modulation technology,it is proposed to use FBMC-OQAM modulation technology to apply to MDM systems.In this system,CLLL-MMSE algorithm is used to solve the impairment compensation of MDM system with MDL.Firstly,through the selection of prototype filter and the design of pilot structure to realize channel estimation,the MDM system based on FBMC-OQAM modulation is successfully established.Then,the orthogonal optimization of channel matrix is realized by CLLL algorithm.After that,the CLLL algorithm and MMSE algorithm are combined.Under strong coupling conditions,when the transmission distance is 1200km,the DMGD is 9ps/km,and the G_offset is 4d B,the MDL value after the lattice reduction is reduced by about 12d B,the CLLL-MMSE algorithm saves about 3d B of OSNR cost compared with the conventional MMSE algorithm,and the computational complexity is not high.