Transformer-EKF for UWB Positioning: A Benchmark Against CNN and BiLSTM Models
Bibliographic record
Abstract
Ultra-Wideband (UWB) indoor positioning systems suffer significant accuracy degradation in Non-Line-of-Sight (NLoS) conditions, where multipath distortions in the Channel Impulse Response (CIR) lead to biased range estimates. While deep learning (DL) models have shown potential in predicting such errors from CIR data, most prior studies focus on isolated architectures and static pipelines, without evaluating their broader integration into full positioning systems. This work presents a comparative study of three DL models—Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory networks (BiLSTM), and Transformer—for CIR-based range error prediction, integrated into both Weighted Least Squares (WLS) and Extended Kalman Filter (EKF) frameworks. In WLS, predicted errors are used as adaptive weights to suppress unreliable measurements; in EKF, they dynamically scale the measurement noise covariance for more accurate filtering. We evaluate all models across four realistic indoor environments using a public UWB dataset. Our results show that Transformer-EKF achieves the best performance, reducing mean positioning error to 0.59m in Environment 2 (severe NLoS/multipath) while maintaining real-time inference at 0.059 milliseconds per position. These findings establish a comprehensive benchmark for learning-based UWB positioning and demonstrate the value of fusing data-driven range correction with model-based tracking.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".