Information transmission method applied to wireless communication system

Abstract

The invention discloses an information transmission method applied to a wireless communication system. The method comprises the following steps: a sending end compresses, quantizes and sends information to be transmitted through an encoder; the receiving end recovers the received compressed information through a decoder matched with the encoder; the encoder and the decoder are neural networks; the decoder comprises a residual network and comprises five information reconstruction modules and one addition module; each information reconstruction module comprises a first convolution sub-module, a full connection sub-module and a second convolution sub-module; the two convolution sub-modules both adopt a swsh function as an activation function; each full-connection sub-module comprises at least two adjacent full-connection layers; and the addition module is used for adding the input of the residual network and the output of the last information reconstruction module so as to reconstruct the information. According to the method, the information transmission performance in an outdoor high-compression-rate scene can be improved, and the mean square error between reconstructed information and original information is reduced.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to an information transmission method applied to a wireless communication system. Background technique [0002] In a wireless communication system, the transmission of a large amount of information needs to consume a large amount of wireless spectrum resources. However, a lot of information is information with data sparsity, including a large amount of 0 data. Direct transmission of this information is a waste of precious wireless spectrum resources. [0003] In order to reduce the occupation of wireless spectrum resources, a method based on machine learning to compress, transmit and reconstruct information with data sparsity has emerged in recent years; compared with traditional compressed sensing methods, this machine-based The learning method does not require high data sparsity, and has the advantages of good robustness and low complexity. [0004] H...

AI Technical Summary

Problems solved by technology

[0004] However, in the process of implementing the present invention, the inventors found that in outdoor high-compression rate scenarios, the existing machine learning-based methods showed poor performance in the information reconstruction link, and the average of the reconstructed information and the original information Minute error (MSE) is quite different

Images