Agentic-V2X: Small Language Model Agents for Deadline-Aware V2X Scheduling in 5G/6G Networks

2026-07-05

AI Summary

This work addresses the challenges posed by large language models—such as high latency, hallucination, and insufficient control guarantees—in meeting the near-real-time scheduling demands of 5G/6G vehicular networks (V2X). To overcome these limitations, the authors propose Agentic-V2X, a novel architecture that employs a compact local language model as a non-real-time rApp policy generator, working in tandem with a lightweight xApp-style controller to periodically produce and execute verified, deadline-aware scheduling policies. The framework incorporates policy verification and repair mechanisms, along with telemetry-driven structured policy generation and adaptive scheduling weight adjustment. Evaluated on the ns-3/ns3-ai platform, Agentic-V2X demonstrates superior reliability over proportional fairness in scheduling critical services under high-density scenarios, effectively balancing policy flexibility with execution safety and exhibiting strong practical applicability.

Abstract

Large Language Models (LLMs) are proposed as control interfaces for next-generation networks, but their latency, hallucinations, and lack of control guarantees make them unsuitable for near-real-time packet schedulers, especially in dynamic V2X environments. This paper introduces Agentic-V2X, an architecture where a small, locally deployed language model acts as a periodic non-real-time rApp-inspired policy creator, while a lightweight xApp-like controller executes validated policies at intervals suitable for scheduling. The framework targets deadline-aware 5G NR V2X scheduling with heterogeneous services (teleoperated driving, cooperative awareness, HD map sharing, and sensor sharing). Given a scenario summary, service objective, and telemetry, the LLM generates a structured policy containing service priorities, weight bounds, and safety constraints. A validator checks and repairs the policy before the controller enforces it via scheduler-weight adaptation in ns-3/ns3-ai. The evaluation compares proportional fair scheduling, static expert policies, a heuristic xApp, static LLM policies, and adaptive LLM-rApp policies over 126 completed runs. Metrics include deadline-constrained packet reception ratio, tail latency, deadline violations, throughput, fairness, policy validity, and safety interventions. Results show that the adaptive LLM-rApp/xApp design generates valid and executable policies and remains competitive at several operating points, including improved mean critical reliability over PF at the highest density. However, paired statistical analysis shows that the adaptive method is not the best aggregate method and remains below the strongest static policies overall. These results support Agentic-V2X as a safe, executable small-LLM policy-generation architecture rather than a universally dominant scheduler.

Problem

• V2X scheduling
• deadline-aware
• 5G/6G networks
• heterogeneous services
• real-time control

Innovation

• Small Language Model
• V2X Scheduling
• Deadline-Aware
• Policy Validation
• 5G/6G Networks