Search papers, labs, and topics across Lattice.
This paper introduces a Variational Autoencoder (VAE)-based multi-task semantic communication framework designed for satellite-assisted connected autonomous vehicles (CAVs), addressing the inefficiencies of conventional communication systems that transmit raw data. By focusing on task-relevant information, the framework optimizes the transmission of essential semantic features, enabling reliable and low-latency communication crucial for safety-critical applications such as traffic sign recognition. The results demonstrate a remarkable bandwidth reduction of 87.23% to 98.17% while ensuring stable performance across different signal-to-noise ratios, showcasing the effectiveness of probabilistic latent representations in enhancing communication efficiency.
Achieving up to 98% bandwidth reduction in satellite communication for autonomous vehicles without sacrificing performance is a game-changer for smart transportation systems.
The development of smart transportation systems and the introduction of 6G wireless communication technologies have significantly changed vehicle network topologies. Future connected autonomous vehicle (CAV) networks require bandwidth-efficient, reliable, and low-latency communication for safety-critical applications such as traffic sign recognition and decision-making. Conventional communication systems transmit raw data regardless of task relevance, which is inefficient in resource-constrained satellite channels where uplink bandwidth is scarce and propagation losses are large. Semantic communication addresses this limitation by transmitting task-relevant information instead of full signal representations. It extracts and conveys essential semantic features and leverages deep learning to optimize task performance at the receiver. Therefore, we present a Variational Autoencoder (VAE)-based multi-task semantic communication framework for satellite-assisted autonomous driving. Unlike deterministic autoencoder-based methods, the proposed model uses probabilistic latent representations for more robust and efficient encoding. The learned features are transmitted over noisy wireless channels to perform traffic sign reconstruction and classification. The framework is trained end-to-end to jointly optimize both tasks. Results show that the proposed approach achieves significant bandwidth reduction of up to 87.23\% to 98.17\% while maintaining stable performance across varying signal-to-noise ratio conditions.
