Semantic V2X Communications for Image Transmission in 6G Systems
Year: 2024
Type: article
Abstract: The success of future Intelligent Transportation Systems (ITS) heavily relies on the availability of wireless communications technologies that are able to fulfil the strict requirements imposed by applications such as autonomous driving. However, due to the scarce spectrum allocated to ITS, current vehicle-to-everything (V2X) networks experience a severe capacity bottleneck. In this paper, we consider a typical ITS application where images are periodically captured and transmitted by the vehicles. Using an open dataset of real video frames captured by vehicles, we show that the current bandwidth allocated to ITS is insufficient, requiring the vehicles to transmit the video at only a small portion of the frame rate at which it can be captured by the sensors. Furthermore, we show that semantic communications (SemCom) using semantic segmentation are essential to solve this problem, since the receivers are not interested in receiving high definition images but a semantic interpretation that assesses the importance of the elements in it. Our results show that, although the benefits of SemCom in terms of energy consumption are limited to algorithms with a low computational complexity, SemCom are able to greatly improve the performance in terms of maximum supported load and latency, making SemCom essential to make future ITS over 6G viable.
Source: IEEE Network
Institutions: Universitat Politècnica de València, Aalborg University
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,995
Awards: G1175115578, G1246816390, G2262748287, G24120701, G2601397997, G2867640486, G294895050, G300979063, G3480869486, G3681454997, G3845001103, G4042783231, G425668148, G5080475149, G5195188938, G586857820, 6G Goal-Oriented AI-enabled Learning and Semantic Communication Networks, G7084143925, G8875184233
OpenAlex ID: W4400071982
