High-Order Statistics Based TDOA Estimation Method

This article introduces a Time Difference of Arrival (TDOA) estimation method based on high-order statistics. This approach enables deeper understanding and practical implementation of TDOA estimation techniques. High-order statistics serve as mathematical tools for analyzing and processing complex data patterns beyond second-order moments. Through higher-order cumulants or polyspectra analysis, we can achieve more robust TDOA estimation against Gaussian noise interference, particularly effective in non-Gaussian environments. The implementation typically involves calculating fourth-order cumulants using functions like cumest() in MATLAB, which mitigates phase ambiguity and improves time-delay resolution. Additionally, we explore practical application scenarios including underwater acoustic localization and radar systems, demonstrating how this method enhances estimation reliability through spectral correlation analysis. We believe this content provides valuable technical insights and implementation guidance to achieve superior results in TDOA-based localization systems.