Related Experiment Video
Updated: Dec 12, 2025
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
1.9K
Adaptive Residual Weighted K-Nearest Neighbor Fingerprint Positioning Algorithm Based on Visible Light Communication
Shiwu Xu1,2, Chih-Cheng Chen3,4, Yi Wu1
1Key Laboratory of OptoElectronic Science and Technology for Medicine of Ministry of Education, College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350007, China.
Sensors (Basel, Switzerland)
|August 14, 2020
Summary
The adaptive residual weighted K-nearest neighbor (ARWKNN) algorithm improves positioning accuracy in visible light communication by dynamically adjusting K based on signal strength. This method significantly reduces average positioning errors compared to existing algorithms.
Area of Science:
- * Wireless communication and localization technologies.
- * Signal processing and algorithm development.
Background:
- * Weighted K-nearest neighbor (WKNN) is a common fingerprint positioning method.
- * Optimizing the K value is crucial for minimizing positioning error in WKNN.
- * Visible light communication (VLC) presents unique challenges for accurate localization.
Purpose of the Study:
- * To propose an adaptive residual weighted K-nearest neighbor (ARWKNN) algorithm for enhanced fingerprint positioning in VLC systems.
- * To dynamically optimize the K value in WKNN based on received signal strength indication (RSSI) residuals.
- * To evaluate the performance of ARWKNN against various existing positioning algorithms.
Main Methods:
- * Fingerprint matching using RSSI vectors.
- * Dynamic adjustment of the K value based on matched RSSI residuals.
- * Performance evaluation using simulation with a signal-to-noise ratio of 20 dB and Manhattan distance in a 2-D space.
Main Results:
- * ARWKNN demonstrated reduced average positioning errors compared to Random Forest, Extreme Learning Machine, Artificial Neural Network, Grid-Independent Least Square, Self-Adaptive WKNN, WKNN, and KNN.
- * ARWKNN achieved the minimum average positioning error in 2-D using Clark distance and in 3-D using minimum maximum distance metrics.
- * The proposed algorithm significantly reduces average positioning error while maintaining similar algorithm complexity compared to SAWKNN, WKNN, and KNN.
Conclusions:
- * The ARWKNN algorithm offers superior positioning accuracy in VLC environments.
- * Dynamic K value optimization based on RSSI residuals is effective for improving localization performance.
- * ARWKNN provides a significant advancement in fingerprint positioning accuracy and efficiency.
