Semantic-Aware Spectrum Sharing in Internet of Vehicles Based on Deep Reinforcement Learning

2024-06-11

IEEE Internet of Things Journal

AI Summary

In high-mobility Internet of Vehicles (IoV) scenarios, vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications suffer from spectrum scarcity and inefficient semantic information transmission. To address these challenges, this paper proposes a semantic-aware spectrum sharing framework. We first introduce two novel metrics—High-speed Semantic Spectrum Efficiency (HSSE) and High-speed Semantic Rate (HSR)—to quantify semantic communication performance under dynamic mobility. Leveraging the Soft Actor-Critic (SAC) deep reinforcement learning algorithm, we jointly optimize semantic feature extraction, transmit power allocation, and symbol duration to enable semantic-driven dynamic resource management. Experimental results demonstrate that our approach improves HSSE by 15% and effective semantic transmission success rate (SRS) by 7% over baseline methods. This work establishes a scalable, deep reinforcement learning–based solution for semantic communication resource orchestration in high-speed, highly dynamic IoV environments.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Agent Communication

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

This article investigates semantic communication in high-speed mobile Internet of Vehicles (IoV), focusing on spectrum sharing between vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications. We propose a semantic-aware spectrum-sharing (SSS) algorithm using deep reinforcement learning (DRL) with a soft actor-critic (SAC) approach. We start with semantic information extraction, redefining metrics for V2V and V2I spectrum sharing in IoV environments, introducing high-speed semantic spectrum efficiency (HSSE) and semantic transmission rate (HSR). We then apply the SAC algorithm to optimize decisions V2V and V2I spectrum-sharing decisions on semantic information. This optimization aims to maximize HSSE and enhance the success rate of effective semantic information transmission (SRS), including determining the optimal V2V and V2I sharing strategies, transmission power, and the length of transmitted semantic symbols. Experimental results show that the SSS algorithm outperforms other baseline algorithms, including other traditional-communication-based spectrum-sharing algorithms and spectrum-sharing algorithm using other reinforcement learning approaches. The SSS algorithm exhibits a 15% increase in HSSE and approximately a 7% increase in SRS.

Problem

• Addressing spectrum scarcity in IoV communications
• Optimizing V2V and V2I spectrum sharing via DRL
• Enhancing semantic transmission efficiency in high-speed IoV

Innovation

• Deep reinforcement learning for spectrum sharing
• Semantic-aware metrics HSSE and HSR
• SAC algorithm optimizes V2V and V2I