DecoyModel

Fibre channel model feeding the decoy bounds (Ma-Qi-Zhao-Lo Eqs. 5-11)

5/5 passed in 1ms

Test	What it does	Result
test_dark_count_floor	With the channel dead the gain is Y0 alone and the QBER saturates at 1/2.	✅ pass
test_gain_double_count	The gain sums Eq. (7) exactly, below Eq. (10) by its double-count Y0*(1 - exp(-eta*mu)) that puts Y1 above its ceiling, and stays in [0, 1].	✅ pass
test_gain_poisson_sum	The closed-form gain Y0 + (1 - Y0)(1 - exp(-etamu)) equals the Poisson sum over Y_i = Y0 + eta_i - Y0eta_i, with eta_i = 1 - (1-eta)^i.	✅ pass
test_ideal_noiseless	Without background the infinite-decoy Y1 is eta and e1 the misalignment.	✅ pass
test_qber_poisson_sum	The same for the QBER: E_muQ_mu = e0Y0 + e_det*(1 - exp(-etamu)) equals the Poisson sum of e_iY_i with e_i = (e0Y0 + e_deteta_i)/Y_i.	✅ pass