SpectrumFM: A New Paradigm for Spectrum Cognition

2025-08-02

AI Summary

Existing spectrum cognition methods suffer from poor generalization and inadequate cross-scenario performance. This paper introduces SpectrumFM—the first foundation model for spectrum cognition—featuring a novel hybrid encoder that jointly integrates convolutional layers and multi-head self-attention to capture both local signal structures and global time-frequency dependencies. We design two self-supervised pretraining tasks: masked signal reconstruction and next-time-step prediction, and incorporate LoRA for parameter-efficient fine-tuning. Evaluated under challenging low-SNR conditions (−4 dB), SpectrumFM achieves a 30% improvement in spectrum sensing detection probability, a 10.2% gain in anomaly detection AUC, and a 9.6% increase in wireless technology classification accuracy—significantly outperforming state-of-the-art approaches. SpectrumFM establishes a transferable, robust, and unified modeling framework for intelligent spectrum cognition.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

The enhancement of spectrum efficiency and the realization of secure spectrum utilization are critically dependent on spectrum cognition. However, existing spectrum cognition methods often exhibit limited generalization and suboptimal accuracy when deployed across diverse spectrum environments and tasks. To overcome these challenges, we propose a spectrum foundation model, termed SpectrumFM, which provides a new paradigm for spectrum cognition. An innovative spectrum encoder that exploits the convolutional neural networks and the multi-head self attention mechanisms is proposed to effectively capture both fine-grained local signal structures and high-level global dependencies in the spectrum data. To enhance its adaptability, two novel self-supervised learning tasks, namely masked reconstruction and next-slot signal prediction, are developed for pre-training SpectrumFM, enabling the model to learn rich and transferable representations. Furthermore, low-rank adaptation (LoRA) parameter-efficient fine-tuning is exploited to enable SpectrumFM to seamlessly adapt to various downstream spectrum cognition tasks, including spectrum sensing (SS), anomaly detection (AD), and wireless technology classification (WTC). Extensive experiments demonstrate the superiority of SpectrumFM over state-of-the-art methods. Specifically, it improves detection probability in the SS task by 30% at -4 dB signal-to-noise ratio (SNR), boosts the area under the curve (AUC) in the AD task by over 10%, and enhances WTC accuracy by 9.6%.

Problem

• Improves spectrum cognition generalization and accuracy
• Enhances adaptability via self-supervised learning tasks
• Optimizes performance for diverse downstream spectrum tasks

Innovation

• SpectrumFM model for spectrum cognition
• CNN and self-attention spectrum encoder
• Self-supervised learning with LoRA fine-tuning