Machine Learning-Based Underground Mine Path Loss Prediction Using mm-Wave Massive MIMO Measurements
Bibliographic record
Abstract
Accurate path loss (PL) modeling in complex environments, such as underground mines, is essential for the deployment of the next-generation wireless networks. This article presents a deep neural network (DNN)-based approach to predict path loss using massive multiple-input multiple-output (MIMO) propagation measurements at 28 GHz within an underground mine environment. Compared to traditional empirical models, including log-distance and multislope methods, the proposed DNN model demonstrates superior predictive accuracy, achieving a root mean squared error (RMSE) of 1.46 dB and a correlation of 99.17%, effectively capturing the intricate propagation characteristics of the environment. Building on the 28-GHz model, this study explores the generalization of the DNN model to other frequencies, such as 26 and 38 GHz, with minimal additional measurements. The generalization process begins by incorporating free-space (FS) path loss data at both 28 GHz and the target frequency, providing a foundational understanding of frequency-dependent behavior. Subsequently, the model is refined using a subset (5%) of target frequency measurements, resulting in improved predictive performance, with RMSE values of 2.23 dB for 26 GHz and 2.86 dB for 38 GHz. Finally, a conditional generative adversarial network (cGAN) is employed to generate synthetic data, enabling the model to learn from an augmented dataset and further enhance its accuracy without requiring extensive real measurements. This approach achieves RMSE values of 1.64 dB for 26 GHz and 2.02 dB for 38 GHz. This multistage framework demonstrates the capability of the DNN model to generalize across frequencies and adapt to challenging scenarios with limited resources, offering a scalable and practical solution for wireless communication challenges in complex environments.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".