SEQCUP MATLAB Script
Description
- SEQCUP for Trilateration for Field Application: seqcup_trilateration.m - SEQCUP - Simulation for Levelling: seqcup_level_simulation_example.m - SEQCUP: seqcup_level_simulation_example2 % This script performs a reproducible Monte Carlo experiment to analyse the % stepwise behaviour of the SEQCUP procedure on a synthetic levelling network. % Using a fixed random seed (rng(42)), it generates two-epoch observations with % Gaussian noise, randomly selects displaced points and displacement magnitudes, % and applies the SEQCUP sequential testing scheme expansion tests up to q_max). %The script classifies each run into correct % identification, over-identification, under-identification, overlap, or no-solution % and computes the conditional stepwise false-alarm rate when the current null % model coincides with the true displaced configuration. - seqcup_GNSS_baseline_simulation_example.m - Monte Carlo calibration of the critical value for the initial SEQCUP stage q=1 % under the pure null model, using Q_delta in local ENU with 3D correlation per baseline. % - Post-selection comparisons use a SINGLE ΔSSE statistic: ΔSSE = SSE0 - SSE_A % between the current null hypothesis and a higher-dimensional alternative. % - Everything is "by POINT": each candidate adds/removes a full 3D block (E,N,U) of a station. % - Full covariance propagation in the local ENU system (ECEF→ENU for both coordinates and covariance matrices). % - For each displacement magnitude in M_list, the script runs mc Monte Carlo realizations % and stores empirical identification probabilities and the conditional stepwise false-alarm rate. % - Random seed is fixed with rng(42) for reproducibility of all Monte Carlo experiments. - seqcup_structured_full.m: pre-screening tool designed to evaluate the geometric separability and structural admissibility of candidate displacement hypotheses in geodetic networks. - Dataset for trilateration: Rofatto, Vinicius; Matsuoka, Marcelo; Klein, Ivandro (2025), “Dataset Field Trilateration: SEQCUP performance”, Mendeley Data, V2, doi: 10.17632/msg783rh2y.2
Files
Institutions
- Universidade Federal de Uberlandia