Robust Time-of-Arrival Location Estimation Algorithms for Wildlife Tracking

Area of Science:

• Ecology and Wildlife Management
• Signal Processing and Estimation Theory
• Biologging and Telemetry

Background:

• Time-of-arrival (TOA) transmitter localization systems are crucial for wildlife tracking.
• Outlier measurements in TOA data can significantly degrade the accuracy of estimated locations.
• Existing algorithms may struggle with data corruption from various sources like interference or clock issues.

Purpose of the Study:

• To develop and present a novel suite of location estimation algorithms for TOA systems.
• To enhance the robustness and accuracy of wildlife tracking by addressing outlier data.
• To improve the computational efficiency and reliability of localization algorithms.

Main Methods:

• Development of algorithms to detect and discard outlier time-of-arrival observations.
• Implementation of methods to resolve ambiguities when multiple locations are equally consistent with measurements.
• Integration of digital elevation map data to infer altitude and refine location estimates near sensors.
• Novel approximation of the covariance matrix for more reliable error estimation.

Main Results:

• Extensive testing on real-world wildlife tracking data validated the efficacy of the new algorithms.
• The algorithms successfully detected and discarded outlier TOA measurements, improving location accuracy.
• Performance tests confirmed the algorithms are computationally efficient and suitable for high-throughput, real-time applications.

Conclusions:

• The new suite of algorithms offers a significant advancement in time-of-arrival transmitter localization for wildlife tracking.
• These algorithms provide more accurate, robust, and efficient location estimates compared to baseline methods.
• The developed system is well-suited for real-time, high-volume data processing in ecological monitoring.