Learned-key sensitivity, brute-force and real-token stages; legend order

exp_learned.py completes the learned side of the result stages, so every
figure can carry both realizations of keyed masking. real() holds the
structured artifacts aside and restores them, since exp_real_sec writes
fixed file names.

replot_security.py ranks legend handles from one declared order at all
three ax.legend call sites, so entries no longer follow plot-call order
and drift between figures.
This commit is contained in:
KiHoLee
2026-08-28 20:31:50 +09:00
parent c00e8ab666
commit 669737e83d
16 changed files with 210 additions and 26 deletions
+9
View File
@@ -0,0 +1,9 @@
snr_db,ter_legit,ter_eve,ter_insider,ter_oma
0,0.4516948856,0.9998824906,0.9962722018,0.5232856128
4,0.223975418,0.9998662393,0.9947945836,0.2740881771
8,0.09790408233,0.9997899832,0.9940332727,0.123424874
12,0.04099952996,0.9997912333,0.9936357409,0.05201666133
16,0.01653382271,0.9997299784,0.9935107309,0.02118419474
20,0.006725538043,0.999737479,0.9934594768,0.008630690455
24,0.002707716617,0.9997112269,0.9934182235,0.003452776222
28,0.001037583007,0.9997274782,0.9934144732,0.001385110809
1 snr_db ter_legit ter_eve ter_insider ter_oma
2 0 0.4516948856 0.9998824906 0.9962722018 0.5232856128
3 4 0.223975418 0.9998662393 0.9947945836 0.2740881771
4 8 0.09790408233 0.9997899832 0.9940332727 0.123424874
5 12 0.04099952996 0.9997912333 0.9936357409 0.05201666133
6 16 0.01653382271 0.9997299784 0.9935107309 0.02118419474
7 20 0.006725538043 0.999737479 0.9934594768 0.008630690455
8 24 0.002707716617 0.9997112269 0.9934182235 0.003452776222
9 28 0.001037583007 0.9997274782 0.9934144732 0.001385110809