Efficient Machine Learning for Wi-Fi CSI-based Human Activity Recognition Using Fast Monte Carlo based Feature Extraction
Bibliographic record
Abstract
High-dimensional doppler data extracted from Wi-Fi channel state information (CSI) offers distinctive velocity and time patterns that are useful for human activity recognition (HAR), but its scale poses significant challenges for real-time inference and deployment on resource-constrained devices. This work proposes an efficient, fast monte carlo (MC) feature selection framework based on the frieze-kannanvempala (FKV) algorithm and coefficient estimation to address this bottleneck. The CSI is preprocessed, and doppler traces are computed to encode the velocity and direction of distinct activities. Afterwards, we perform FKV to decompose the doppler data, and the coefficient of the resulting singular vectors is estimated. Using rejection sampling, the topmost features are selected on the basis of their weights, thereby reducing the size of our features. The method identifies a compact set of velocity-time features that preserve critical motion information while significantly reducing computational overhead. The experimental evaluations demonstrated that the decision tree classifier achieved the highest precision at 99.8%, followed by convolutional neural networks (CNN) 96%, the hybrid CNN-long-short-term memory (CNN-LSTM) achieved 87%, while the LSTM model lagged at 53%. These results demonstrated that the integration of fast MC-based feature selection significantly reduced computational overhead without sacrificing classification performance, making it suitable for scalable and real-time HAR applications.
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 | 0.000 | 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 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".