A Node Localization Algorithm Based on Multi-Granularity Regional Division and the Lagrange Multiplier Method in Wireless Sensor Networks | JoVE Visualize

A Node Localization Algorithm Based on Multi-Granularity Regional Division and the Lagrange Multiplier Method in Wireless Sensor Networks

Fengjun Shang¹, Yi Jiang², Anping Xiong³

¹College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China. shangfj@cqupt.edu.cn.

Sensors (Basel, Switzerland)|November 22, 2016

View abstract on PubMed

Summary

This study introduces an improved wireless sensor network localization algorithm using multi-granularity region partitioning and Received Signal Strength indicator (RSSI) for accurate node positioning. The method enhances accuracy by combining Voronoi diagrams and Lagrange multipliers for precise coordinate calculation.

Area of Science:

Computer Science
Electrical Engineering
Network Engineering

Background:

The Internet of Things (IoT) relies on Wireless Sensor Networks (WSNs) for data acquisition and processing.
Accurate localization of sensor nodes is crucial for WSN functionality within IoT environments.
Existing localization methods face challenges due to environmental factors affecting signal strength.

Purpose of the Study:

To develop an advanced localization algorithm for sensor nodes in WSNs.
To improve the accuracy and robustness of node positioning in diverse network conditions.
To address limitations in current range-based localization techniques.

Main Methods:

Proposed a multi-granularity region partition strategy for location area division.
Utilized Received Signal Strength indicator (RSSI) with Gaussian fitting for distance estimation.
Employed Voronoi diagrams for regional division and Lagrange multipliers for coordinate calculation.
Implemented vector similarity and centroid algorithms for nodes within positioning units.
Applied Taylor expansion for coordinate correction of nodes outside positioning units.

Main Results:

The proposed algorithm effectively divides regions using Voronoi diagrams and anchor nodes.
Accurate distance estimation is achieved through optimal RSSI value computation.
The combined approach of multi-granularity division and Lagrange multipliers refines coordinate calculation.
Validation in a real environment confirmed the localization method's effectiveness.

Conclusions:

The developed localization algorithm offers enhanced accuracy for sensor nodes in WSNs.
The integration of multiple techniques addresses practical challenges in sensor node positioning.
This research contributes to more reliable data acquisition in IoT applications.