Study On Phase Noise Compensation Algorithm In Coherent Optical C-mQAM System

With the rapid development of Internet of things & big data and artificial intelligence information processing technology,the demand for information transmission capacity is increasing day by day.The coherent detection technology improves receiver sensitivity and can combine with high order modulation format and digital signal processing(DSP)to improve the channel capacity and the transmission distance of system.However,due to the randomness of phase noise of each data point in the system,how to effectively compensate phase noise with low-complexity algorithm is the key to the future development of coherent optical communication.In addition,a lot of new technologies have been introduced in coherent optical communication system to enhance the transmission capacity.For example,probabilistic shaping technology can be combined with coherent optical communication system to enhance the transmission capacity close to Shannon's limit.Therefore,phase noise mitigation plays an important role in such system.Aiming at the influence of phase noise on transmission signal in coherent optical communication system,this paper mainly studies the phase noise compensation algorithms and applies them to the probability shaping system with high transmission capacity.The specifics are as follows:1.This paper studies two typical phase noise compensation algorithms,that is n-PSK partitioning and blind phase search(BPS)algorithms.Compared with the traditional viterbi-viterbi algorithm,n-PSK segmentation algorithm can make use of phase information of all constellation points and effectively improve the system's tolerance to phase noise.However,the algorithm eliminates the modulation phase by N/2 power operation.The higher the modulation order,the higher the computational complexity.The BPS algorithm has a good performance,but it needs to process a block length symbol in parallel,and each symbol needs to make use of multiple test phases to obtain its phase noise estimation.Its computational complexity is too high.2.Based on C-mQAM coherent optical communication system,a lowcomplexity blind phase noise compensation algorithm(BPR)is proposed in this paper.Firstly,some special cost functions are constructed based on the phase distribution of constellation points in c-mQAM.A rough compensarion of the phase noise can be realized by minimizing the cost function.Considering the computational complexity,the cost function can be approximated by a simple cosine function.Then,the maximum likelihood phase noise estimation algorithm is used to realize residual phase noise compensation.Finally,the BPR algorithm,the n-PSK segmentation algorithm and the BPS algorithm are simulated in a 28 GBaud back-to-back coherent QAM communication system and their computational complexities are compared.Simulation results show that,when C-16 QAM modulation or C-64 QAM modulation is used,compared with the n-PSK segmentation algorithm and the BPS algorithm,the BPR algorithm can achieve similar performance while reducing the computational complexity.3.Because the optimal probability distribution of the transmitted signal in the additive white Gaussian noise(AWGN)channel is Gaussian distribution,PS technology is introduced into the 28 GBaud back-to-back coherent optical system,and phase noise compensation algorithm is adopted at the receiver.Based on C-mQAM modulation,this paper proposes a probability shaping scheme.That is,half of input bits are used for probability matching and used to realize QAM amplitude mapping,and the other half of input bits are directly used to realize phase mapping by differential encoding.According to the simulation results,PS technology can effectively improve the performance of the phase noise compensation algorithms.