LMS (least mean square) method for updating coefficient of self-adaptive equalizer without frequency difference and phase difference feedback

Abstract

The invention discloses an LMS (least mean square) method for updating a coefficient of a self-adaptive equalizer without frequency difference and phase difference feedback and relates to the technical field of coherent light communication. The LMS method comprises steps as follows: an X polarization receipt signal and a Y polarization receipt signal of the self-adaptive equalizer are subjected to time domain equalization; the error formula of time domain equalization is defined; the coefficient of the self-adaptive equalizer is updated with a gradient algorithm according to the error definition to guarantee that the self-adaptive equalizer constantly tracks changes of the polarization state and frequency difference and phase difference feedback is not required. With the adoption of the LMS method, updating of the coefficient of the self-adaptive equalizer does not depend on frequency difference and phase difference feedback any more, a coherent light communication system can correctly track the changes of the polarization state or even extreme changes of the polarization state, and the robustness of the system is guaranteed.

Description

technical field [0001] The present invention relates to the technical field of coherent optical communication, and specifically relates to an LMS (Least mean square, least mean square) method for updating coefficients of an adaptive equalizer without frequency difference and phase difference feedback. Background technique [0002] The theory and experiment of coherent optical communication began in the 1980s. Coherent optical communication systems are recognized to have the characteristics of high sensitivity, good frequency selectivity, and high spectral efficiency, which are beneficial to the improvement of communication capacity. In addition, in coherent optical communication, the damage caused by dispersion and polarization mode dispersion to the signal is linear damage. At this time, the communication channel can be modeled as a linear time-invariant system, so through moderately complex DSP (Digital Signal Processing, digital signal processing ) technology can compens...

AI Technical Summary

Problems solved by technology

[0014] However, the problem of the existing LMS method is: in the actual signal processing process, both the frequency difference estimation and the phase difference estimation depend on the correct output of the adaptive equalizer

In the actual deployed system, although the polarization state changes slowly in most cases, extremely fast polarization changes may occur instantaneously. At this time, the correct output of the adaptive equalizer cannot be guaranteed, and there is no correct frequency difference and phase difference feedback. The existing The LMS method will not work correctly, so using the existing LMS method does not guarantee the survival of the system under extremely fast polarization changes

Images