So now we are back in the 'can-we-find-the-ULB-before-the-cheap-batteries-run-down' game again. All this extra work being proposed for streaming DFDR data is because the existing DFDRs are poorly designed their location systems are only useful in an inland lake or river (and perhaps not even then).
In the 'old days' before solid state memory there were all sorts of survivability issues; but with fast solid state memory with significant capacity this is not a problem. The real problem is the use of batteries that just do not have the power driving ULBs that are unsuitable for crashes in the ocean. Obviously, the beancounters at airlines are not interested, but it is time for an international mandate to have ULBs that actually work.
I would suggest that the ULBs requirements are:
Detection Range At least 25 nautical miles in open ocean
Battery Life At least 6 months, possibly by reducing number of pulses and using smart transponder that does not go into regular short location signals until it receives a search request sonar signal
Encoded Location Signals The signals from the ULBs should be encoded with airframe ID, their position (last GPS position of the aircraft) and their depth in the water.
These requirements appear to be asking too much and all that is being offered is a slightly longer life battery.