Explore MATLAB source code curated for "观测数据" with clean implementations, documentation, and examples.
This program simulates the estimation of AR parameters in ARMA models from observational data using least squares methods, covering scenarios with both known and unknown parameters. The implementation incorporates numerical algorithms for parameter optimization and statistical analysis.
Implementation of a one-dimensional Wiener filter design approach. The methodology involves generating three sets of observation data, estimating their AR model parameters using methods like the Yule-Walker equations or Burg's algorithm, and analyzing the impact of signal length and model order on experimental results through performance metrics such as mean square error.
MATLAB simulation implementation of Least Squares (LS) and Total Least Squares (TLS) estimation methods. The observation data is generated with zero-mean, unit-variance Gaussian white noise for n=1,2,...,128 samples. The simulation tests TLS with AR order 4 to estimate AR parameters and sinusoidal frequency, then uses SVD-TLS for the same estimation. Key implementation aspects include: (1) Testing AR orders 4 and 6, (2) Executing SVD-TCS with unknown AR order, (3) Running simulations minimum 20 times for statistical reliability.
A program for demonstrating two-station cross localization errors that generates observation data under the assumption of a target moving uniformly from point A (dx1, dy1) to point B (dx2, dy2) in kilometers. Using velocity input V (km/h), dx2 = V * t where t is observation duration (hours), with dy1 = dy2 = R representing target distance. The sampling time interval is assumed to be 1 hour. The program implements coordinate transformation algorithms and error calculation methods to simulate realistic localization scenarios.
A comprehensive parser for reading and processing RINEX 2.1 observation data files with robust error handling and multi-system support