You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The assembly of ETI frames into transmission frames is not checked for alignment contiuously. It is only checked once at startup. This is why when using the ZeroMQ input, a zmq message must contain four ETI frames, so that this alignment is never disturbed.
But in an scenario with timestamped transmission, once the modulator is streaming, it is expected that no incoming ETI frames get dropped. However, ZeroMQ doesn't guarantee that all frames reach the modulator. If we only lose a few, and the buffer is still large enough to sustain streaming, the UHD output will not interrupt, and the system will be transmitting timestamped frames at the wrong point in time.
This breaks the SFN feature.
The text was updated successfully, but these errors were encountered:
This is the first step to fix issue #6.
The goal is to enable the OutputUHD to check
for FCT consistency. Once it can do that,
it will also be able to reset the UHD streamer if
necessary.
The assembly of ETI frames into transmission frames is not checked for alignment contiuously. It is only checked once at startup. This is why when using the ZeroMQ input, a zmq message must contain four ETI frames, so that this alignment is never disturbed.
But in an scenario with timestamped transmission, once the modulator is streaming, it is expected that no incoming ETI frames get dropped. However, ZeroMQ doesn't guarantee that all frames reach the modulator. If we only lose a few, and the buffer is still large enough to sustain streaming, the UHD output will not interrupt, and the system will be transmitting timestamped frames at the wrong point in time.
This breaks the SFN feature.
The text was updated successfully, but these errors were encountered: