Using IOTA Tangle and Machine Learning for a Defensive Model-Based Approach Against Replication Attacks on Wireless Sensor Networks
Bibliographic record
Abstract
Wireless Sensor Networks (WSNs) are essential for data collection across various domains but face growing risks from replication attacks, which introduce new vulnerabilities and security challenges. To address this issue, we propose a novel hybrid approach that integrates Distributed Ledger Technology (DLT) with adaptive Machine Learning (ML) methods, aiming to bolster both security and trustworthiness within WSNs. Specifically, our approach utilizes DLT to secure voting records and manage rewards, while adaptive ML models detect replica nodes by analyzing network parameters, including location, signal strength, and transmission rate. We present and evaluate three ML-based models for detecting replication attacks: 1) Random Forest Model (RFM), 2) Adaptive Weighted Random Forest Model based on Predicted Replica Nodes (AWRFM-PRN), and 3) Adaptive Weighted Random Forest Model based on Predicted Good and Replica Nodes (AWRFM-PGRN). The AWRFM-PRN and AWRFM-PGRN models enhance detection accuracy through iterative weight adjustments based on previous predictions. Our simulations show that the hybrid approach significantly improves detection performance compared to traditional methods. We evaluated our models by increasing the dataset size with varying proportions of replica nodes across ten subsets. We found that the AWRFM-PGRN model achieved around 71% accuracy when replica nodes comprised 50% or more of the network. Meanwhile, the AWRFM-PRN model demonstrated high effectiveness with accuracy ranging from 80% to 99% for replica nodes constituting 15% to 40% of the network. Furthermore, all models delivered nearly 99.9% accuracy when the proportion of replica nodes was between 5% and 10%. This innovative integration of DLT with adaptive ML modeling establishes a benchmark for robust and tamper-proof security in WSNs, offering significant enhancements over traditional ML techniques such as RFM, particularly in scenarios with high replica node counts.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".