Reduced Complexity MLSE Algorithm Based On BDFE In Optical Fiber Communication System

In high-speed optical fiber communication system, dispersion becomes the main constraining factor of transmission capacity, which will lead to optical pulse broadening. The pulse broadening can cause Inter-symbol interference (ISI) and increase the system bit error rate, both of which will reduce the performance of optical fiber communication system. However, the traditional dispersion compensation in optical domain has many disadvantages such as high cost and less adaptability, so a new technology for fiber dispersion compensation is necessary. With the development of signal processing technology and the cost reduction of integrated circuits, Electronic Dispersion Compensation (EDC) is gradually being applied in practical projects. Among the several dispersion compensation equalizer structures, Maximum Likelihood Sequence Estimate (MLSE) equalizer has the best equalization performance. However, limited by the complexity of the algorithm, MLSE equalizer has been unable to be used widely.The thesis focuses on improving the equalization performance and reducing the complexity of the algorithm. First, the structure of optical fiber communication system and the principle of dispersion are described, then simulate and analyze the effect of dispersion. After that, the general structure and principle of EDC and equalizer are introduced. And then, the typical equalizers consisting of FFE (forward feed equalizer), DFE (decision feedback equalizer) and MLSE (maximum likelihood sequence estimate) are introduced. Through simulation, the equalization performance and the algorithm complexity are compared. Then, the structure of DFE with time reverse and bi-directional DFE (BDFE) are described. At last, a new equalization structure consisting of BDFE and MLSE is designed. The new structure will ensure the equalization performance and reduce the complexity of algorithm, which is examined by simulation.