Localization algorithm with on-line path loss estimation and node selection

Area of Science:

• Wireless Sensor Networks
• Localization Algorithms
• Signal Propagation

Background:

• Received Signal Strength (RSS)-based localization offers low complexity compared to Time of Arrival (TOA) or Angle of Arrival (AOA) techniques.
• Accurate RSS localization relies on suitable propagation models, which are challenging to establish in indoor environments.
• Energy efficiency is a critical constraint for wireless sensor networks.

Purpose of the Study:

• To develop a cooperative localization algorithm for wireless sensor networks that dynamically estimates the path loss exponent using RSS measurements.
• To enhance positioning accuracy and energy efficiency in sensor networks.
• To propose a node selection mechanism for optimizing neighbor usage and discarding unreliable links.

Main Methods:

• Dynamic estimation of the path loss exponent based on RSS measurements.
• Implementation of a node selection mechanism to limit the number of neighbors for positioning.
• Validation through computer simulations and real-world experiments using Crossbow MICA2 motes.

Main Results:

• The proposed algorithm dynamically estimates the path loss exponent, improving localization accuracy.
• The node selection mechanism effectively limits computational load and discards poor quality links.
• Demonstrated a favorable trade-off between position accuracy and energy efficiency.

Conclusions:

• The cooperative localization algorithm offers a practical solution for sensor networks, balancing accuracy and energy consumption.
• The dynamic path loss exponent estimation and node selection mechanism are key to the system's performance.
• The approach is suitable for resource-constrained sensor network applications.