A semantic communication model for the task of high quality image transmission to edge-end devices
Year: 2024
Type: article
Abstract: With the development of Industrial Internet of Things (IIOT) equipment to intelligentization and digitalization, a large number of intelligent mobile devices and Automated Guided Vehicles (AGV) are widely used in modern intelligent warehouse management, which generates a large number of machine vision tasks as well as massive demand for image and video data transmission. Guaranteeing the quality of image data transmission while saving the amount of transmitted data by edge-end devices has become one of the key issues in modern IIOT. As an efficient new communication method, semantic communication technology can significantly improve the transmission efficiency by focusing on the intrinsic meaning of the transmitted data. In this paper, a semantic communication model is proposed for the modern intelligent warehouse environment, using the ray tracing method to obtain its channel characteristic parameters, according to the obtained channel characteristic parameters to establish an end-to-end semantic communication scheme applicable to the warehouse environment, to achieve the optimization of the communication process transmission. Simulation analysis shows that compared with the traditional communication scheme, the proposed scheme can still achieve a structural similarity index higher than 0.8 when the transmission signal-to-noise ratio (SNR) is less than 6 dB, which effectively improves the image reconstruction quality compared with the traditional compression scheme. • Proposing a semantic communication model for the Industrial Internet of Things. • The model is deployed on mobile edge devices to conserve data resources. • Obtaining scenario-specific channel parameters through channel modeling. • Include channel parameters in the model to improve communication quality.
Source: Internet of Things
Authors: Zhaoxu Wen, Jiandong Fang, Xiuling Wang
Institution: Inner Mongolia University of Technology
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,400
OpenAlex ID: W4402945708