Beardedbruce is correct- but as youths we had no fear.
I remember a bomb we made with two simple and easily obtained components which I won't name. In a container separated into two parts by a wax seal we placed the materials and sealed the ends.   Shock disrupted the seal and the end result was an explosion (plus fragments of the container).
Dangerous to the chemist? Oh, yes, but a suicide bomber wouldn't care. Could these substances be identifed? Easily enough, but I wonder if one of the components would be known to the 'sniffers'.

Of course there are many explosive substances that don't require generation of oxygen, as noted.

'Signature' needed for detection- correct. To do a proper job, a mass spectrograph would be needed, with a capable operator, and a comparative library in the computer. Would you be willing 1) to pay for such equipment at airports, etc., 2) pay trained operators and 3) Wait the time required to sample and accomplish the multiple analyses?
In a research lab where I worked, we used a scanning electron microscope with spectrographic analysis system to locate particles for analysis and identify them. The SEM may not be needed, but reading the 'signatures' (many different spectral lines with spacing and height characteristics and grouping) requires the analysis. Working in a particular field, we had an idea of what to expect on analysis. The problem is probably more complex today than I picture it.