Extended Kalman Filter for 3D TDOA-AOA Localization

This documentation focuses on the main simulation program for Extended Kalman Filter (EKF) implementation in 3D TDOA-AOA localization. 3D TDOA-AOA (Time Difference of Arrival - Angle of Arrival) localization is a widely used positioning technique that utilizes both time difference and angular arrival information to accurately determine target positions in three-dimensional space. The core algorithm implements an Extended Kalman Filter to handle sensor measurement errors and enhance positioning accuracy. Unlike standard Kalman Filters, the EKF extension specifically addresses nonlinear system dynamics through Jacobian matrix linearization of the measurement and state transition models. The main program structure typically includes: - State vector initialization (position, velocity, and possibly acceleration components) - Measurement model implementation combining TDOA hyperbolic equations and AOA trigonometric relationships - Process and measurement noise covariance matrix configuration - Recursive prediction-correction cycle with linearized state transitions Key implementation aspects involve: 1. TDOA measurement processing using time difference calculations between multiple sensor pairs 2. AOA data incorporation through elevation and azimuth angle measurements 3. Jacobian matrix computation for linearizing nonlinear measurement functions 4. Adaptive noise covariance tuning for optimal filter performance During simulation, users can modify parameters such as sensor positions, noise characteristics, and initial state estimates to observe their impact on localization accuracy. This allows for algorithm optimization through parameter adjustment and performance analysis. The program serves as an effective tool for understanding and applying 3D TDOA-AOA localization techniques while demonstrating practical EKF implementation for nonlinear estimation problems.