Font Size: a A A

Research On Channel Estimation And Equalization Technology In PDM-O-OFDM System

Posted on:2013-08-13Degree:MasterType:Thesis
Country:ChinaCandidate:J LiFull Text:PDF
GTID:2268330392465604Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
In recent years, with the considerable progress of high-speed analog to digital conversion(ADC), digital signal processing (DSP) technology, coherent detection can not only improvethe receiver sensitivity, but also maintain the linearity of chromatic dispersion and polarizationmode dispersion effects, and is suitable for the detection scheme of long-haul fiber opticcommunication systems. Polarization division multiplexing (PDM) does not change the fiberlink, making the system transmission capacity doubled via increasing the complexity of thetransmitter and receiver. Orthogonal frequency division multiplexing (OFDM) techniqueswidely used in optical fiber communication systems because of its anti-dispersion ability, and itcan be reduced the requirements of the electrical domain equalizer rate of the receiver. Thecombination of PDM, coherent detection and OFDM increases the flexibility of the system, andhas been recognized as the best choice for the next generation100Gb/s optical transmissiontechnology. Among the various DSP technologies in PDM-CO-OFDM systems, channelestimation is an important one, which is the basis of depolarization. The Least Square (LS)estimation is widely used because of its simplicity. However, it has a low tolerance of noise.The principles of the PDM-CO-OFDM system were mainly analyzed, and the crosstalk anddamage caused by the fiber channel effect on signal were researched. The c hannel estimationtechnique was detailed introduced, and specific to the defects of LS channel estimation, scatteredpilot-based channel estimation with joint noise reduction program was proposed. The maincontents are as follows:(1) First, optical OFDM system model was built and its principles were elaborated, afteron this basis, PDM-CO-OFDM system and its channel model were focused researched, and thegroup speed dispersion, polarization mode dispersion, polarization dependent loss effect onPDM-CO-OFDM system signal when high-speed transmission were analyzed;(2) Second introduced channel estimation technique. The performance of the severalestimation criteria and interpolation algorithms commonly used were simulated withMATLAB, the simulation results show that the performance of LS algorithm and linearinterpolation algorithm are not optimal, but simple to implement; In order to make the system overhead minimum, meanwhile spectrum utilization achieve maximum, In this paper, anoptimal discrete pilot pattern was selected by several simulation experiments:(3) Then40Gb/s CO-OFDM system was constructed by using the VPI simulation tools,and estimated based on the optimized scattered pilot channel, different channel estimationperformances were simulated jointed with MATLAB. The simulation results show that even ifthe fiber channel is rapidly varying, the method can also be greatly reduced the systemoverhead and the bit-error rate; About0.9dB improvement in OSNR sensitivity is achieved overconventional scattered pilot and1.8dB improvement over block pilot while maintaining thesame system overhead;(4) Finally,80Gb/s PDM-CO-OFDM systems was constructed by using the VPIsimulation tools, specific to the defects of LS channel estimation, on the basis of optimizedscattered pilot, an joint denoising method based on multiwavelet denoising and Intra-symbolfrequency-domain averaging (ISFA) denoising was proposed, and confirmed by the VPI andMATLAB co-simulation. The simulation results show that this proposed method has improved0.96dB in OSNR sensitivity at the BER of1e-3(threshold of FEC) compared to the common LSmethod, and can be sufficient for applications with forward error correction (FEC) when OSNRis about14.4dB.
Keywords/Search Tags:orthogonal frequency division multiplexing (OFDM), coherent detection, polarization division multiplexing (PDM), scattered pilot, channel estimation, equalization, joint denoising
PDF Full Text Request
Related items