SSRC stands for Synchronization Source.
An SSRC is a 32-bit identifier carried in every RTP packet header that uniquely identifies a media stream within an RTP session. Each audio track and video track in a WebRTC session has its own SSRC value.
How SSRC works in WebRTC
When a browser sends media, it assigns a random SSRC to each stream. The receiving end uses the SSRC to:
- Distinguish between different media streams (e.g., camera video vs. screen share)
- Associate RTCP feedback with the correct stream
- Identify simulcast layers (each layer gets its own SSRC)
- Track retransmission streams (RTX packets use a separate SSRC)
In SFU architectures, the server must manage SSRC mappings carefully. When forwarding media from one participant to another, the SFU may need to rewrite SSRC values to avoid collisions and maintain proper media identification.
SSRC and SDP
SSRCs are declared in the SDP using the a=ssrc attribute lines. In Unified Plan SDP, each m= line contains the SSRC(s) for that media track. Related identifiers include MID, RID, and RRID.
