Semantic communication which breaks the Shannon limit in the existing communication field is considered as a promising technology. In this study, we present a novel semantic communication system that utilizes a convolutional neural network(CNN) architecture. The system incorporates dual convolutional layers, dual pooling layers, and a fully connected layer to facilitate semantic encoding. To evaluate the performance of the proposed system, we conduct experiments using the MNIST dataset. The experiment results demonstrate that 99.1% accuracy in text information can be obtained.
Semantic Communication System Based on Convolutional Neural Networks
2023-10-11
2173616 byte
Conference paper
Electronic Resource
English
Semantic Segmentation of Low Earth Object Satellites using Convolutional Neural Networks
British Library Conference Proceedings | 2021
|