Indoor positioning in wireless local area networks with online path-loss parameter estimation

Area of Science:

• Computer Science
• Electrical Engineering
• Signal Processing

Background:

• Global Positioning System (GPS) accuracy is limited in indoor and urban canyon environments.
• Received Signal Strength (RSS) in Wireless Local Area Networks (WLANs) is a common alternative for location-based services.
• Existing RSS-based systems face performance limitations due to the complex and changing nature of radio propagation.

Purpose of the Study:

• To develop a robust localization algorithm for non-stationary environments.
• To enhance the accuracy of RSS-based positioning systems.
• To address the challenges posed by dynamic radio propagation characteristics.

Main Methods:

• A Sequential Monte Carlo (SMC) approach implementing optimal Bayesian estimation.
• Incorporation of Rao-Blackwellization to improve the efficiency of the estimation scheme.
• Analysis and implementation of two key statistical models for RSS characterization.

Main Results:

• The proposed algorithm effectively estimates and tracks key parameters of RSS propagation.
• Demonstrated improved positioning accuracy under various simulated and real-world conditions.
• Validation of the algorithm's performance in a real-world scenario.

Conclusions:

• The developed localization algorithm offers enhanced accuracy for indoor and urban environments.
• The method effectively handles the non-stationarity of RSS propagation.
• This approach provides a more reliable solution for location-based services where GPS fails.