Optimizing Image Transmission In 6G-Enabled V2X Network Using Semantic Communication

The future success of Intelligent Transportation Systems (ITS), especially for applications like autonomous driving, depends heavily on wireless communication technologies that meet very strict performance requirements. However, the spectrum currently allocated for ITS is limited, causing serious capacity bottlenecks in existing Vehicle-to-Everything (V2X) networks. In this work, we focus on a typical ITS scenario where vehicles periodi- cally capture and send images. Using real video data from open datasets, we demonstrate that the current bandwidth assigned to ITS is insufficient, forcing vehicles to transmit video at only a small fraction of the sensors’ original capture rate. To overcome this chal- lenge, semantic communication using semantic segmentation plays a crucial role. Instead of sending high-definition images, receivers get a simplified, meaningful interpretation of the scene that highlights important elements. Our results show that while the energy efficiency benefits of semantic communication depend on keeping algorithm complexity low, it significantly improves how much data the network can handle and reduces delays. This makes semantic communication essential for enabling practical and efficient ITS services over future 6G V2X networks.