I have professional concerns about the reliability of alarm systems that depend entirely on one way wireless communication on licence exempt radio frequencies. These concerns are based on my 12 years employed designing equipment that had no alternative but to use radio due to one or more items being mobile.
I can see now why you are in effect overcompensating in relation to this issue.
A home alarm has a very static environment where the sensors range is fixed at the point of installation and the systems tested in situ.
You are using your experience of designing systems that needed to work critically in a mobile scenario where the levels of inteference and signal direction/strength and local topography could change at a moments notice and assuming the same level of uncertainty would exist in a static home environment.
Yes its possible that if you were to randomly keep movong your yale sensors around the house you would be capable of putting a sensor somewhere it would not work due to local conditions but that is why the system is tested to ensure it works properly.
The only outside influence then would be the local environment and yes if a taxi cab firm suddenly opens up next door and sticks up a 15mtre antenae yes you would be screwed.
Otherwise the threats are hypothetical and not in my experience of any significance.
As JohnD says you are saying that you should not fit a one way wireless diy system because it will fail when in actual fact all you should be saying is that there is a remote possibility that it will fail.
There is not a single home I visit where I am happy to let the home security be handled by one single device alone. Even if a hallway has a door sensor I promote a pir sensor as well as a backup and the same with all the ground floor rooms. I would never assume that a hallway sensor would be fine to protect your lounge as they would have to go through the hall to get in because they could come in the lounge window instead so again I promote a pir in every room with door or window access.
This reduces the chance of a system failing to operate to virtually zero as the likelihood of every sensor being blocked at that particular instance becomes a mathmeticians puzzle to actually calculate the probability of failure.
The chances of someone leaving home and forgetting to set the alarm in the first place is significantly expotentially higher than any probability of a properly installed one way wireless system failing.