[Live-devel] RTP Slicing with ProxyServer

[Ross Finlayson]

> Can I grab the RTPSink object from my running ProxyServerMediaSession, create a new ProxyMediaSubsession to my new RTSPServer and 'simply' reuse the RTPSink object?

What you’re describing should be possible, I think.  There’s no reason why a “MediaSink” object (including a “RTPSink”) cannot call, in succession, “getNextFrame()” on different “FramedSource” objects.  However, this would probably require modifying our existing ‘proxy server’ code (not just subclassing it), so you’d be pretty much on your own.

However, I’m not convinced that this is the best way to solve your problem, given that:

> The infrastructure we use is very dynamic and can change at any time (Wireless Meshed Network). Our video source is also not static and so may the corresponding RTSPServer change it's location (Flying drone).

If you have just a single video source (a drone), and a single receiver (perhaps a “LIVE555 Proxy Server”), then would it not be better to solve this problem at a much lower level in the network protocol stack - in particular, below the IP level?  I.e., why not allow the endpoints to keep the same IP address, and use a mesh routing protocol to handle the communication between them.  This is the way that mobile phone networks, work, for example.

If you did this, then you wouldn’t need to change any of our software (or any other application-level communication software).  You could run a standard RTSP server on the drone, and a standard RTSP client (or perhaps a “LIVE555 Proxy Server”) on the receiving node.


Ross Finlayson
Live Networks, Inc.
http://www.live555.com/