Enwar 3.0: An Agentic Multi-Modal LLM Orchestrator for Situation-Aware Beamforming, Blockage Prediction, and Handover Management

2026-05-04

AI Summary

This work addresses the robustness challenges of vehicular millimeter-wave communication under dynamic environments, sensor degradation, and link fluctuations by proposing a joint optimization framework that integrates multimodal perception with a large language model (LLM). The approach incorporates a classifier-based sensor health assessment mechanism, an LLM coordination architecture driven by structured task prompts, and a chain-of-thought (CoT)-guided reinforcement learning strategy enhanced with human feedback. A synthetic degradation training pipeline is designed to enable real-time detection of multisource sensor impairments. Through context-aware adaptive model selection, the system achieves over 88% beam selection accuracy across 15 sensor configurations, an F1 score exceeding 98% in occlusion prediction, and 87% correctness in complex decision-making reasoning, demonstrating high accuracy, strong robustness, and favorable interpretability.

Abstract

Maintaining robust millimeter-wave (mmWave) connectivity in vehicular networks requires real-time adaptation to environmental dynamics, sensor degradation, and link variability. This paper presents Enwar 3.0, an environment-aware reasoning framework that unifies multi-modal sensing, agentic large language models (LLMs), and context-driven model selection for predictive beamforming, blockage detection, and handover management. Building upon prior iterations of Enwar, the proposed architecture integrates a classifier-driven assessment of sensor health with a primed LLM that orchestrates multiple specialized agents through structured, task-aware prompting. A novel synthetic degradation pipeline enables the training of a sensor degradation classifier that detects real-time impairments across camera, radar, LiDAR, and GPS inputs, achieving over 99% accuracy. The LLM, trained via chain-of-thought (CoT) priming and human-in-the-loop feedback, coordinates agent calls for beam selection, blockage forecasting, and environment perception while dynamically loading sensor-specific models based on environmental context. Extensive evaluations across 15 sensor combinations demonstrate that Enwar 3.0 delivers state-of-the-art performance in both predictive accuracy and interpretability, with beam selection accuracy exceeding 88%, blockage F1-scores surpassing 98%, and reasoning correctness reaching 87% on complex decision prompts. This work establishes a scalable foundation for LLM-integrated wireless systems that reason, perceive, and adapt in real-time.

Problem

• mmWave connectivity
• beamforming
• blockage prediction
• handover management
• sensor degradation

Innovation

• Agentic LLM
• Multi-Modal Sensing
• Sensor Degradation Classification
• Context-Aware Beamforming
• Chain-of-Thought Reasoning