Digital Signal Processing for Front-end Non-idealities in Coherent Optical OFDM system

In this thesis, we propose digital signal processing algorithms for combatting the front-end non-idealities in CO-OFDM systems, including carrier frequency offset, linear phase noise and IQ mismatch. Firstly, we propose a novel frequency offset compensation method, which is successfully demonstrated through experiment. Our algorithm achieves the widest estimation range with one training symbol. Secondly, we propose a novel decision-aided phase estimation method and its combination with pilot-aided and decision feedback methods to compensate for common phase error induced by linear phase noise. Furthermore, a modified time-domain blind algorithm is proposed for mitigating the intercarrier interference induced by linear phase noise in CO-OFDM system with non-constant amplitude modulation formats. Thirdly, we propose a decision-aided joint compensation method for Tx IQ mismatch and channel distortion. Lastly, we analytically derive new LLR with linear phase noise term for LDPC coded OFDM system with different modulation formats: differential MPSK, pilot-aided MPSK and pilot-aided M-QAM.

Author dc:creator

• CAO SHENGJIAO

• coherent optical communication, OFDM, carrier frequency offset, linear phase noise, IQ mismatch, LDPC