光信号多维度特征提取及灵活识别数据集

面向灵活光网络的信号识别需求，为了实现对多种调制格式信号的识别，基于OptiSystem和Matlab搭建了计算模拟平台，产生多维度信号经光纤传输并进行特征提取及灵活识别。本数据集主要包括光信号多维度特征提取及灵活识别的实验仿真数据。根据具体的研究内容和技术方案，进一步把数据集划分为“基于PCA和SVD的MFI方法数据集”、“基于星座图和SVM的MFI方案数据集”、“基于LSTM的相干光通信系统的调制格式识别数据集”、“基于多特征输入和混合训练神经网络的信号识别方案数据集”，数据量大约为250Mb。

To address the signal recognition requirements in flexible optical networks and achieve recognition of signals with various modulation formats, a computational simulation platform was built based on OptiSystem and Matlab. Multi-dimensional optical signals were generated, transmitted through optical fibers, and then subjected to feature extraction and flexible recognition. This dataset mainly includes experimental simulation data for multi-dimensional feature extraction and flexible recognition of optical signals. According to specific research contents and technical solutions, the dataset is further divided into the following subsets: "Dataset for MFI method based on PCA and SVD", "Dataset for MFI scheme based on constellation diagram and SVM", "Dataset for modulation format recognition in coherent optical communication systems based on LSTM", and "Dataset for signal recognition scheme based on multi-feature input and hybrid-trained neural network". The total data volume of the dataset is approximately 250 Mb.

搜集汇总

背景与挑战

背景概述

该数据集面向灵活光网络的信号识别需求，基于计算模拟平台产生多维度光信号传输特征，用于调制格式识别研究。数据集包含基于PCA、SVM、LSTM等多种方法的仿真数据，总数据量约为250MB。