Datasets
Standard Dataset
Deep-Depth Probability-Maintained Notch Sequences for PAM
- Citation Author(s):
- Submitted by:
- Jingnan Li
- Last updated:
- Sat, 12/14/2024 - 04:24
- DOI:
- 10.21227/j64h-wa80
- License:
- Categories:
- Keywords:
Abstract
Nonlinear distortion is critical for optical communication systems with a high baud rate and a high-order modulation format. Thus, a simple and accurate method to measure the nonlinear distortion is highly desired. Although simple notch, which directly removes the certain frequency components of nonlinear system input and then measures the re-growth components of nonlinear system output, is straightforward to measure nonlinear distortion, it is only applicable to the Gaussian signal. However, pulse amplitude modulation (PAM) and quadrature amplitude modulation (QAM), which are widely used in communication systems, are not Gaussian signals. In this study, we propose a deep-depth probability-maintained notch generation method for the PAM to accurately characterize the nonlinear distortion based on the signal spectrum measurement. Employing the proposed time order perturbation, the notch depth for PAM2/4/8 achieves 20/24/28 dB, which allows for a large dynamic range nonlinear distortion measurement for various modulation formats. The experiment demonstrates that the nonlinear noise power ratios are measured within a root mean square error of 0.6 dB for the uniform PAM2/4/8 as well as the probabilistic shaped PAM4/8. The proposed method is also applicable to square QAM formats because a square QAM is composed of two independent PAMs.
We disclosed the generated deep-depth probability-maintained notch sequences for PAM2/4/8. There are 6 sequences with different notch center frequencies for each modulation format. The sequence length is 65536, the notch width is 2.5% of the baud rate, and the 6 notch center frequencies are uniformly distributed within half of the baud rate.