Machine-Learning-Based Combined Path Loss and Shadowing Model in LoRaWAN for Energy Efficiency Enhancement
Bibliographic record
Abstract
Many practical Internet of Things (IoT) applications require deploying end nodes (ENs) in hard-to-access places where replacing batteries is difficult or impossible. As a result, the ENs demand high-energy efficiency. Long-range wide area network (LoRaWAN) is an IoT protocol that aims to achieve low-energy consumption. However, the energy consumption in LoRaWAN is related to transmission power, which can be set mainly based on path loss and shadow fading modeling and link budget analysis. Hence, appropriately setting this transmission power parameter saves energy and guarantees reliable communication links. Traditional path loss and shadow fading modeling and transmission power setting do not consider the variations caused by different environmental effects. In this work, we show via real-life data analysis that path loss and shadow fading depend on environmental variables. We propose machine learning models to calculate the empirical path loss and shadow fading, which is used to set the transmission power to save ENs’ energy. Our models include the effects of distance, frequency, temperature, relative humidity, barometric pressure, particulate matter, and signal-to-noise ratio. Specifically, the models are based on multiple linear regression, support vector regression, random forests, and artificial neural networks, exhibiting a root mean square error (RMSE) up to 1.566 dB and$R^{2}$up to 0.94. For energy saving, the developed models serve to set the transmission power and spreading factor based on the adaptive data rate (ADR) algorithm principles, which reduces the link margin saving energy up to 43% compared with the traditional ADR protocol.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".