Abstract:The high mobility of automobiles causes channel estimation uncertainties in vehicle-to-vehicle (V2V) communications. Moreover, scarce spectrum resources further bottlenecked the Quality of Service (QoS) in vehicular communication networks. Nonorthogonal multiple access (NOMA) technique is introduced to solve these problems, which reuses resource blocks (RBs) to improve the network's throughput and reduce latency with exploiting the power domain. In this article, we devise a multiagent reinforcement learning (MARL)-based resource allocation method for roadside units (RSUs) in vehicular communication networks. A joint sub-band scheduling and transmit power allocation problem is investigated, aiming to find rational and reasonable solutions at each RSU, under QoS constraints and power limits, from a global perspective. Due to the complicated structure of this high-dimensional optimization, it is extremely challenging to accomplish in polynomial time. The proposed method adopts MARL technique in RSUs to actualize collaboration and self-learning. RSUs act as agents, collectively interacting with the environment to maximize global energy efficiency, the ratio of the sum rate received at vehicles to the total power consumption of relevant RSUs, by trading-off between transmission rate and link interference. With distinctive observations and relevant feedback rewards, each agent learns to improve spectrum and power allocation by updating Q-networks using the gained experiences. The proposed method has much less complexity compared to the centralized implemented method, but still provides approximate performances. Simulation results demonstrate that the proposed method outperforms two existing baselines and a state-of-the-art power allocation scheme in terms of both average energy efficiency and probability of failure.

Enabling Adaptive Optimization of Energy Efficiency and Quality of Service in NR-V2X Communications Via Multi-agent Deep Reinforcement Learning

Multi-Agent Reinforcement Learning-Based Resource Management for V2X Communication.

Matching Combined Heterogeneous Multi-Agent Reinforcement Learning for Resource Allocation in NOMA-V2X Networks

Energy-Efficient Resource Allocation for V2X Communications

Multi-agent Deep Reinforcement Learning Based Resource Management in Heterogeneous V2X Networks

Deep Reinforcement Learning Based Wireless Resource Allocation for V2X Communications

Joint Optimization of Age of Information and Energy Consumption in NR-V2X System based on Deep Reinforcement Learning

Multi-agent Reinforcement Learning Based Resource Allocation for Vehicular Networks.

Deep Reinforcement Learning Enabled Energy-Efficient Resource Allocation in Energy Harvesting Aided V2X Communication.

Optimization of Resource Allocation for V2X Security Communication based on Multi-Agent Reinforcement Learning

Deep Reinforcement Learning Based Resource Allocation with Heterogeneous QoS for Cellular V2X.

Spectrum-Energy-Efficient Mode Selection and Resource Allocation for Heterogeneous V2X Networks: A Federated Multi-Agent Deep Reinforcement Learning Approach

Multiagent Deep-Reinforcement-Learning-Based Resource Allocation for Heterogeneous QoS Guarantees for Vehicular Networks

Resource Allocation in V2X Communications Based on Multi-Agent Reinforcement Learning with Attention Mechanism

Meta-Reinforcement Learning Based Resource Allocation for Dynamic V2X Communications

Decentralized Multi-Agent DQN-Based Resource Allocation for Heterogeneous Traffic in V2X Communications

Multi-Agent Reinforcement Learning-Based Decentralized Spectrum Access in Vehicular Networks with Emergent Communication

Multi-Agent Reinforcement Learning Based Adaptive Parameter Optimization for Semi-Persistent Scheduling in C- V2X Mode 4

Multi-Agent Reinforcement Learning for Long-Term Network Resource Allocation through Auction: a V2X Application

Multi-Agent RL Enables Decentralized Spectrum Access in Vehicular Networks

Multi-Agent-Deep-Reinforcement-Learning-Enabled Offloading Scheme for Energy Minimization in Vehicle-to-Everything Communication Systems