A method for automatic detection and decoding of morse signals based on machine learning

Abstract

The invention discloses a Morse signal automatic detection and decoding method based on machine learning. The method extracts and compares the representative features in the time-frequency diagram of the signal, and realizes the automatic detection and decoding of the Morse signal in a channel containing multiple types of signals. At the same time, by introducing the clustering of the three types of character lengths in the code text, the decoding accuracy of the manually sent code text with low accuracy of the traditional Morse decoding method is improved. Through the actual test of different channel environments, the Morse automatic detection accuracy rate of the method of the present invention is kept above 95%, the automatic decoding accuracy rate is kept above 80%, the average processing delay of a single Morse signal is stable within 0.25 seconds, and the entire Morse detection The decoding method has high timeliness.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a method for automatic detection, discrimination and decoding of Morse signals based on machine learning. Background technique [0002] As a key technical field for computers to realize automatic pattern recognition, machine learning aims to use computers to imitate the human decision-making process based on existing knowledge and then analyze and judge new data. Among them, classifier is an important research direction of machine learning technology. The concept of classification refers to obtaining a classification function or constructing a classification model (that is, a classifier) ​​on the basis of existing data. This function or model can map the data records in the database to one of the given categories, so that it can be applied to data prediction. In short, a classifier is a general term for methods for classifying samples in data mining, including algorith...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is to provide a classifier Morse detection and decoding method based on machine learning to ensure that the traditional Morse signal detection and decoding methods have high labor costs and the accuracy is greatly affected by human factors. On the premise of detection and decoding accuracy, it can effectively reduce labor costs and improve the efficiency of Morse detection and decoding

Images