Code and derived data for: Random-Truncation Maximum Likelihood for Weibull Clutter Parameters from CFAR Exceedances with Unrecorded Thresholds

Code, CA-CFAR convolution tables, per-run records and result files supporting the manuscript "Random-Truncation Maximum Likelihood for Weibull Clutter Parameters from CFAR Exceedances with Unrecorded Thresholds" by Ming-Lung Kung, Sheng-Tao Chen and Chien-Hung Chen (submitted to Signal Processing). The package contains the Monte Carlo simulation and estimation code for OS-CFAR and CA-CFAR detectors in Weibull clutter: random-truncation maximum likelihood when the threshold of each retained sample is not recorded, with the shape parameter alone or jointly with the scale. It includes the numerical-convolution CA-CFAR tables (cleaned to a running maximum when loaded) and a check of these tables over the range of shapes the estimators evaluate, every result file behind the reported numbers, the data files of Figs. 1-8 and S1 (every plotted value with the file and field it comes from) with the scripts that draw the printed figures from them, Tables 2 and S1, and the studies behind Supplementary Tables S2-S6: the Fisher information on the shape with the scale unknown, a shape grid (k = 0.4, 0.8, 1.2; 28 cells, 300 runs each), the coverage of 95% intervals for the joint estimator (1000 runs per setting), and the analysis of two public IPIX sea-clutter records (#17 and #26). A number ledger maps each new number of the text to its result file and JSON key, and a script recomputes the numbers added in the last revision and checks them against the printed values. The IPIX records themselves are not redistributed; README.md gives their download addresses and full SHA-256 checksums. All runs used Python 3.12.10, NumPy 2.4.6 and SciPy 1.17.1. README.md describes the directory structure, the path convention (every script finds the package from its own location), the drawing script, data file and command of each figure, the generator and command of each table, the script, result file and seed vector of every check of the Supplementary material, the master seeds, what is not included, and a re-run check of the package. DEPOSIT_CHANGES.md lists the edits made to the files for this deposit; ERRATA.md corrects text fields inside unchanged result files. The package is uploaded as five zip files, each under 10 MB, that together form the folder SP_repository_package_v3/ with 605 files: extract all of them into the same folder and check the result with SHA256SUMS.txt (see README_PARTS.txt).