Research On Modulation Formats And Symbol Rate Identification For High Spectral Efficiency Signal

Optical performance monitoring technology can efficiently obtain the physical parameters for achieving channel identification,bandwidth resuource allocation and signal quality evaluation in optical network transmission.In order to achieve these functions,the monitoring system is required to recognize the modulation formats and symbol rate of the signal.In direct detection optical fiber network,the existing optical performance monitoring technology is based on the symbol rate and modulation format which is already known or acquired from the upper level protocol.However,due to the limitation of the intermediate network node processing ability,the intermediate node with the cross layer communication is impossible.Consequently,the symbol rate and modulation format identification technology in middle node is particularly important.By means of the study of existing recognition methods,a new recognition method of modulation formats is improved and proposed.Improved modulation formats and symbol rate identification methods are more flexible,and the recognition accuracy and stability of high frequency signal are verifed through simulation analysis.The paper analyses the history,recent progress,application prospects,and trend of modulation format and symbol rate identification technology.The working principle of transport system,including launch transmission,receiving module are briefly introduced.Besides,a system simulation model is established,and operating principle of the modulation format and symbol rate identification in direct detection are presented.The identification of signal in different optical signal to noise ratio and dispersion is discussed in the research of symbol identification with the aid of autocorrelation algorithm.Then the symbol rate of different signal can be accurately identified within certain scope of chromatic dispersion and optical signal noise ratio.Later,it is demonstrated that roll-offs factors significantly affect the identification accuracy of Nyquist signals.Variance analysis and principal component analysis algorithm are presented to deal with asynchronous delay sampling signal for modulation format identification.The variance algorithm is a simple method to study signal variance range for different formats.The simulation analysis indicates the identification accuracy is greatly influenced by the dispersion: the decrease of accuracy accompanied by the increase of dispersion.Within the lower dispersion,variance range is not overlapped,and the average recognition accuracy can reach up to 90%.In order to improve the flexibility and the accuracy,the principal component analysis algorithm is combined with autocorrelation algorithm to investigate the effect of delay time.Then the relationship between delay time and identification accuracy is investigated.It is clear from the simulation results: the recognition accuracy exceeds 90% with delay time of 3/10~3/4 minimum symbol period and 0~500 ps/nm dispersion range.The results show that roll-offs factors have limited influence on the modulation format recognition accuracy.