Related Experiment Video
Updated: Oct 3, 2025
08:05
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
10.9K
LT-FS-ID: Log-Transformed Feature Learning and Feature-Scaling-Based Machine Learning Algorithms to Predict the
Abhilash Singh1, J Amutha2, Jaiprakash Nagar3
1Fluvial Geomorphology and Remote Sensing Laboratory, Indian Institute of Science Education and Research Bhopal, Bhopal 462066, India.
Sensors (Basel, Switzerland)
|February 15, 2022
Summary
This study introduces a new method for predicting intrusion detection barriers in Wireless Sensor Networks (WSNs). The novel approach accurately forecasts barrier numbers, enhancing border security with explainable AI.
Area of Science:
- Computer Science
- Network Security
- Machine Learning
Background:
- Advancements in computational power and explainable AI facilitate rapid intrusion detection.
- Wireless Sensor Networks (WSNs) are crucial for real-time monitoring and security applications.
Purpose of the Study:
- To develop a novel approach for accurately predicting the number of barriers needed for intrusion detection and prevention in WSNs.
- To enhance the efficiency and accuracy of border security systems using WSNs and machine learning.
Main Methods:
- Monte Carlo simulation was used to extract key features: Region of Interest (RoI) area, sensor sensing range, sensor transmission range, and sensor count.
- Feature importance and sensitivity analyses were performed to assess feature relevance and risk.
- Log transformation and feature scaling were applied before training a tuned Support Vector Regression (SVR) model, termed the LT-FS-SVR model.
Main Results:
- The LT-FS-SVR model demonstrated high accuracy in predicting the number of barriers, achieving a correlation coefficient (R) of 0.98.
- The model yielded a Root Mean Square Error (RMSE) of 6.47 and a bias of 12.35.
- Comparative analysis showed the proposed LT-FS-SVR model outperformed benchmark algorithms including Gaussian Process Regression (GPR), Generalised Regression Neural Network (GRNN), Artificial Neural Network (ANN), and Random Forest (RF).
Conclusions:
- The proposed LT-FS-SVR model offers a highly accurate and effective solution for predicting barrier requirements in WSN-based intrusion detection systems.
- This research contributes to improving border security by leveraging explainable AI and advanced machine learning techniques for faster and more precise threat prevention.
- The study validates the superiority of the developed model over existing methods, paving the way for practical implementation in real-world security scenarios.
