I have heard a lot about the failures of even the best-funded anti-ballistic missile technology. The usual explanation is that ABM is very hard after the boost phase because of evasion techniques and the tiny margin for error. It seems reasonable that ABM would be relatively easy at the boost phase as getting the missile in the air typically requires an easily-identifiable and nearly-vertical path to be efficient.
I've read that the main obstacle to boost-phase ABM isn't intercepting the missile in boost phase but rather detecting the launch in time to attempt the interception.
It is my understanding that only a few known materials lie in the sweet spot of being stable enough to not immediately decay and unstable enough to require a low force to break the nucleus. It is also my understanding that it is possible to identify a material by its radiation profile given enough time.
So, I wonder whether it would be possible to identify specific radioactive materials at a distance - say, that of low-earth orbit - given sufficiently precise equipment and enough time to detect the profile. If it were possible, then ABM should be relatively easy - all you would need to do would be to take a survey of the radioactive material and determine whether the material is moving with a quick upward trajectory. Since material takes time to be developed (enriched) it seems like you would have enough time to spot the suspect material.
What are some problems with this kind of detection system and why, if possible, would it not be pursued?