Automatic modulation identification method based on attention mechanism and multi-feature fusion

Abstract

The invention discloses an automatic modulation recognition method based on an attention mechanism and multi-feature fusion, and the method comprises the steps: inputting three streams of signals into a network, extracting spatial features and time sequence features through a CNN and an LSTM, carrying out the fine adjustment of a feature vector through the attention mechanism and Gaussian noise, and finally achieving the modulation classification through a full connection layer. The method comprises the steps of performing signal preprocessing on data of a public data set; constructing an automatic modulation recognition network based on an attention mechanism and multi-feature fusion; iQ data is collected in real time and sent to a network identification modulation mode after data preprocessing. According to the method, complementarity between different networks and various features reflected by different signal representation are fully considered, related features of signals in space and time are extracted by using a combined structure of the CNN and the LSTM, the modulation identification precision is improved, and the method can be well applied to a non-cooperative communication system.

Description

technical field [0001] The invention belongs to the fields of communication signal processing and artificial intelligence, and in particular relates to an automatic modulation recognition method based on attention mechanism and multi-feature fusion. Background technique [0002] Modulation recognition technology is a very key technology in the field of wireless communication. Identifying the modulation mode of wireless communication signals is a basic technology in the fields of non-cooperative communication, electronic countermeasures, smart antennas, software radios, and wireless spectrum management. Both military field and civilian field have very extensive application and very important significance. In the military field, electronic information warfare is the new battlefield. After intercepting the enemy's communication signal on the battlefield, the modulation and demodulation technology can be used to directly obtain the communication content of the opponent, and the...

AI Technical Summary

Problems solved by technology

In the field of communication, in order to solve the problems of poor adaptability of modulation recognition classifier design and strong reliance on expert prior knowledge, it is proposed that the use of deep learning can solve this problem to a great extent.

Images