There is another thread on this. While maybe not technically impossible for NK to have put a nuke in orbit, it is so unlikely as to be practically impossible and certainly laughable. One satellite is tumbling and presumably lost, the other is in an unplanned, and deteriorating orbit with less solar energy than planned.
Ya'll keep expressing interest in EMP, so I'll ramble about that for a bit.
Optimal height for an EMP attack is of course relative. It needs to be above the atmosphere, but beyond that the EMP effect of the device is a function of the yield (roughly inverse cube). Think of the explosion as a growing sphere. When that sphere hits the upper atmosphere it pulls electrons from the atmosphere and for lots of reasons pushes them to ground. On the way they are attracted by conductors, including power lines and even short bits of wire in cars, computers, etc. creating a current/voltage spike and overloading them.
You can imagine that a small yield device detonated at a very high altitude will create a sphere that has relatively little energy per square meter compared to a high yield device at the same altitude. Remember that the spheres are the same size regardless of yield. The particles are relativistic, that is moving near the speed of light, regardless of yield, the difference is the energy density on the surface of the sphere. Yield also affects how big the sphere can and still be coherent, at some point the energy dissipates.
The area affected by an EMP is a function of altitude (the effect is limited to line of sight, although the power spike could travel over the horizon by way of power lines) so the optimal height for the detonation is based on available yield. Too low and the damaged is limited by line of sight, too high and the energy is too diffuse to be effective. To be effective the detonation must be above the atmosphere to get the downward EMP, but it must also have a high enough energy density to deliver an effective energy spike at ground.
For example, a 1kt explosion in a lunar orbit would affect fully half the earth, but the energy per square meter from the device would be so low when it hit the upper atmosphere that it would not create enough voltage at ground to be effective against anything near the ground.
The math on all this is not terribly complicated, but as far as I know nothing has been published since the early 1980's. At that time the estimates showed that an effective spike, assumed to be 25kv/m as I recall, would be generated in an area with a radius in kilometers roughly equal to the kt yield of the warhead when detonated at optimal height.
If you believe that and accept that NK's largest nuclear device is less than 25kt, then you see that optimal altitude would be 25km and it would effect an area with a radius of 25km. As a practical matter their satellites, at about 300km, can affect a very large portion of the US, but they can't generate enough energy at that altitude to create an EMP that would hurt us.
That's enough for this morning, but the short version is that if NK wanted to attack us then they would get far better results from a near ground detonation in a west coast sea port than they would from a satellite detonation given their available technology.
The risk from NK in the short term is proliferation, not attack, and certainly not an EMP attack.
Just one man's opinion based on the information I've seen.