A Modulation Intelligent Recognition Method Driven by Data and Knowledge

Abstract

The invention discloses a modulation intelligent recognition method driven by data and knowledge, which mainly solves the problem of low classification accuracy of the existing modulation recognition method under low signal-to-noise ratio, dependence on a large number of training samples, and the difficulty of high-order modulation modes in the recognition process. confusion etc. The implementation steps are: spectral data collection; construction of corresponding attribute vector labels according to different modulation methods; construction and pre-training of attribute learning models according to attribute labels of different modulation methods; construction and pre-training of modulation recognition visual models; construction of feature space conversion Model, combined with the visual model and attribute learning model to build a data-knowledge dual-driven modulation mode intelligent recognition framework; transfer the parameters of the pre-trained visual model and pre-trained attribute learning model, retrain the conversion model; judge whether the network training is over, and output the classification results. The invention significantly improves the identification accuracy under low signal-to-noise ratio and reduces confusion among high-order modulation modes.

Description

technical field [0001] The invention belongs to the technical field of communication, and relates to a modulation intelligent identification method driven by data and knowledge. Background technique [0002] Intelligent identification of modulation mode is a crucial technology in wireless intelligent communication. It distinguishes different types of modulated signals by learning the unique characteristics of received signals. With the gradual popularization of 5G, research on 6G wireless communication networks has emerged. As a key component of 6G communication network, intelligent communication technology urgently needs in-depth research, among which the intelligent identification of modulation mode is one of the key technologies. Therefore, it is very important to carry out research on the intelligent identification of modulation modes. At the same time, modulation identification has been widely used in military and civilian fields. In military applications, modulation...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the disadvantages of the existing modulation recognition technology in complex and dynamic actual scenes, such as poor performance, dependence on a large number of training samples, and serious confusion between high-order modulation modes, the present invention proposes a modulation mode intelligence driven by data and knowledge. The recognition method introduces semantic attributes as knowledge, and embeds attribute features into the visual feature space through the conversion model, which greatly improves the accuracy of modulation recognition under low signal-to-noise ratio, reduces the confusion between high-order modulation methods in the recognition process, and Ability to reduce dependence on the amount of training data

Images