Research On Frequency Offset Correction Algorithm For Coherent Optical OFDM Communication System

As the growth of various network services and internet data centers, the influence of information network has been unstoppable in modern human society. The high data rate and large capacity of optical communication network systems will definitely become the main development direction in the future. Coherent optical orthogonal frequency division multiplexing system(CO-OFDM) is one of the promising candidates of long-haul high capacity optical transmission system, which presents great dispersion tolerance and high spectral efficiency. Additionally, its flexibility is an important feature thanks to a wider selection of modulation format and payload location in subcarriers.However, a non-negligible drawback of CO-OFDM system is the impact of frequency offset(FO) between the local oscillator(LO) and the carrier laser source. In order to satisfy the practical applications for long-haul coherent optical OFDM communication systems, this dissertation conduct the research on frequency offset correction algorithm.Firstly, the research status of the state-of-the-art frequency offset correction(FOC) algorithm in CO-OFDM system are reviewed. Meanwhile, we analyze the principle and system structure of CO-OFDM. The detailed analyses on mathematical model of some classical FOC algorithms are followed.Secondly, we conclude that there will be none significant impact on system performance as long as the normalized absolute value of FO is less than 0.005 by numerical simulation and reference to the results of other researches. Then, aiming at the shortcoming of most FOC algorithms, in this dissertation, we propose a novel training symbol based FOC algorithm. In chapter ?, the mathematical principle of the proposed FOC algorithm has been discussed in details.Finally, in order to further verify the performance of our proposed FOC algorithm, numerical simulation and a 16QAM-CO-OFDM based 100 km single mode optical fibers(SSMF) transmission system have been set up. The efficiency and robustness of the proposed FOC algorithm have been validated by outstanding and stable experimental results. In addition, we also achieve high capacity long-haul(1000km) transmission experiments with the proposed FOC algorithm by recirculating loop. The experimental results further validate the stability and effectiveness of the algorithm.Comparing with most FOC algorithms, our algorithm can successfully overcome the bandwidth limitation and high-frequency distortion by designing the training symbol in frequency domain and taking advantage of frequency offset bias element. Additionally, most FOC algorithm's robustness and validity depend heavily on a good optical signal-to-noise ratio(OSNR). However, in practical communication systems, the OSNR generally degrades significantly after long distance transmission and multi-span amplifiers, thus those FOC algorithms may be unstable or even invalid. To address this issue, we adopt a scheme based on two-step approximately iterative operation to guarantee the stability of the algorithm under poor OSNR situations. In the consequence, the tolerance to OSNR is significantly enhanced until 8dB and the algorithm works stably for a long time.In this dissertation, we demonstrate a novel training symbol based FOC algorithm with excellent compensation accuracy and robustness in CO-OFDM long-haul transmission system under poor OSNR conditions, satisfying almost all practical laboratory application requirements. With the ability to eliminate the effect of frequency offset, the proposed FOC algorithm has a significant practical value for other researches in CO-OFDM transmission system.