While there is complexity in being able to start a deflagration to detonation in a pulse-detonation engine, the V-1's engines used a detonation process so it clearly was doable some time in the past.
Regarding William Bollay's concept, which entailed utilizing a rotary-valved system (which looking at the diagrams, seemed to incorporate a whole bunch of small valves/tubes), what got in the way of it being effective?
I'd imagine the energy imparted to the flow by the rotation would help make the device work better. This could be achieved by using a small starter motor, or using some kind of liquefied-gas expander engine (which would use outside air to vaporize the liquefied gas, spinning the rotary-valve and imparting energy to the fuel/air mixture).