Session Description Protocol

Description

The Session Description Protocol (SDP) is a textual protocol format, standardized by the IETF in RFC 4566 and widely adopted by 3GPP, for describing multimedia session initialization parameters. In 3GPP architectures, particularly the IP Multimedia Subsystem (IMS), SDP is not a standalone signaling protocol but a session description format that is encapsulated within the body of Session Initiation Protocol (SIP) messages. Its primary role is to enable endpoints to agree on a set of compatible media characteristics before a session (e.g., a voice or video call) is established.

An SDP description consists of a series of lines of text, each of the form `<type>=<value>`. The description is divided into two main sections: a session-level section and zero or more media-level sections. The session-level section (`v=`, `o=`, `s=`, `c=`, `t=` lines) describes the session as a whole, including the session originator, a session name, timing information, and optional connection data (e.g., a base IP address). The media-level sections (`m=`, `a=` lines) describe each media stream (e.g., audio, video) that is part of the session. A critical `m=` line specifies the media type (audio/video), port number, transport protocol (RTP/AVP), and a list of supported media formats (typically RTP payload types corresponding to codecs like AMR, EVS, or H.264).

The SDP offer/answer model, defined in RFC 3264, governs how SDP is used for negotiation. One endpoint (the offerer) generates an SDP offer listing the session parameters it is willing to receive and/or send. This offer is placed in a SIP INVITE or other SIP message. The answering endpoint examines the offer, selects from among the offered options based on its own capabilities and policies, and generates an SDP answer. The answer is sent back in the SIP response (e.g., 200 OK). The answer confirms the mutually agreeable parameters, including the selected codecs, IP addresses, and ports for media flow. This model allows for capabilities exchange and symmetric agreement without requiring a common baseline configuration.

Within 3GPP, SDP is extended with specific attributes (`a=` lines) to support telecommunications requirements. These include attributes for labeling media streams, indicating bandwidth, specifying used codec configurations, enabling/disabling media streams, and supporting network-related functions like Interactive Connectivity Establishment (ICE) for NAT traversal. SDP is also used in conjunction with the Real-time Transport Protocol (RTP) for the actual media delivery and with the Real-time Transport Control Protocol (RTCP) for quality feedback. Its simplicity and flexibility make it the cornerstone of session negotiation for IMS-based services like Voice over LTE (VoLTE), Video over LTE (ViLTE), and Rich Communication Services (RCS).

Purpose & Motivation

SDP exists to solve the problem of session negotiation in IP-based multimedia communications. Before a call or video session can begin, the participating endpoints must agree on a common 'language' for media exchange: which codecs to use, on which network addresses and ports to send media, and what type of session it is. SDP provides a standardized, human-readable, and machine-parsable format to convey this information. Its creation by the IETF was motivated by the growth of internet multimedia conferencing tools (like the MBONE) in the 1990s, which lacked a standard way to advertise session information.

3GPP adopted SDP for its IMS architecture to leverage an existing, well-understood internet standard rather than inventing a proprietary signaling method. This aligned with the IMS principle of using IETF protocols wherever possible. SDP addressed the limitations of earlier telecommunication signaling, which was often hard-coded for specific codecs (like G.711 in PSTN) or required out-of-band provisioning. In the dynamic, multi-vendor, multi-service environment of IMS, endpoints from different manufacturers need to discover mutual capabilities dynamically. SDP's offer/answer model perfectly fits this need, enabling efficient negotiation even when endpoints support different sets of codecs and features.

The use of SDP within SIP messages in IMS solves critical problems for mobile operators: it enables service flexibility (easy introduction of new codecs like EVS), supports complex sessions with multiple media streams (e.g., voice, video, and text), and facilitates interoperability with other IP-based networks. Furthermore, SDP's extensibility via new attributes allows 3GPP to define specific parameters for quality of service (QoS), network resource allocation, and emergency services, tailoring the internet protocol to meet carrier-grade reliability and regulatory requirements. Thus, SDP is the essential enabler for rich, negotiated multimedia sessions in modern IP-based mobile networks.