We use Edgeworth expansions to show the higher-order accuracy of the bootst
rapped matched filter (MF). Specifically, the bootstrapped MF has a probabi
lity of false alarm of order n(-1) distant from the preset level, where n i
s the sample size. We also show that the order is n(-3/2) for sinusoidal si
gnals. Simulation results compare the bootstrapped MF with the classical ME
We briefly discuss the bootstrapping approach when the noise variance is u
nknown and when signal parameters are unknown.