SpectrumLLM: Large Language Models for Next-Generation Spectrum Prediction
Bibliographic record
Abstract
With the emergence of dynamic spectrum sharing in 5G networks, spectrum prediction (SP) has become essential for proactive spectrum management, reducing interference and optimizing dynamic spectrum access (DSA). However, existing SP methods often struggle to achieve high accuracy and generalization due to the complexity and rapidly evolving nature of modern wireless environments. Large Language Models (LLMs), as a significant breakthrough in the field of artificial intelligence, have demonstrated exceptional capabilities in handling time-series data and multi-modal tasks. This article presents the first exploration of integrating LLMs with SP, introducing a SpectrumLLM architecture that aligns radio spectrum state (RSS) series with the text space of LLMs for more efficient SP by applying tokenization, embedding and prompt engineering. Experimental evaluations demonstrate that LLMs significantly outperform traditional SP methods such as DLinear, Autoformer, and LSTM, offering superior adaptability and predictive accuracy. The findings highlight the potential of LLMs in revolutionizing SP for next-generation wireless systems, paving the way for intelligent spectrum management in future 5G-advanced and 6G networks.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".