UAV Individual Identification via Distilled RF Fingerprints-Based LLM in ISAC Networks

2025-08-17

AI Summary

To address security-critical UAV identity authentication in low-altitude Integrated Sensing and Communication (ISAC) networks, this paper proposes a Radio Frequency Fingerprinting (RFF)-based Large Language Model (LLM) framework. We innovatively integrate a lightweight GPT-2 architecture (RFF-LLM) with Lite-HRNet and introduce, for the first time, a dynamic knowledge distillation mechanism: a Proximal Policy Optimization (PPO)-driven adaptive temperature scheduler overcomes local optima inherent in static distillation. The method operates directly on raw I/Q signals and achieves 98.38% identification accuracy on our proprietary DRFF-R1 dataset, with only 0.15M parameters and 2.74 ms inference latency—significantly outperforming state-of-the-art approaches. Key contributions are: (i) the first application of LLMs to RFF-based UAV identification; (ii) a reinforcement learning–guided dynamic distillation paradigm; and (iii) end-device deployability with high accuracy, ultra-low model size, and minimal latency.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Unmanned aerial vehicle (UAV) individual (ID) identification is a critical security surveillance strategy in low-altitude integrated sensing and communication (ISAC) networks. In this paper, we propose a novel dynamic knowledge distillation (KD)-enabled wireless radio frequency fingerprint large language model (RFF-LLM) framework for UAV ID identification. First, we propose an RFF-LLM framework based on the modified GPT-2 model to improve the identification accuracy in complex outdoor environments. Then, considering the parameter overhead of the RFF-LLM, we design a dynamic KD strategy to compress the model. Specifically, the proximal policy optimization (PPO) algorithm is employed to dynamically adjust the distillation temperature, overcoming the local optimum dilemma inherent in static KD. As a next step, the knowledge of the RFF-LLM is adequately transferred to the lightweight Lite-HRNet model. Finally, our experiments are conducted based on the self-built drone RFF dataset of Release one, namely DRFF-R1, by collecting the I/Q signals of 20 commercial UAVs in channel 149. The experiment results show that the proposed framework achieves 98.38% ID identification accuracy with merely 0.15 million parameters and 2.74 ms response time, which outperforms the benchmarks.

Problem

• Identify UAVs using RF fingerprints in ISAC networks
• Improve accuracy with distilled LLM in complex environments
• Compress model via dynamic knowledge distillation strategy

Innovation

• Modified GPT-2 model for UAV ID identification
• Dynamic knowledge distillation with PPO algorithm
• Lightweight Lite-HRNet for efficient knowledge transfer