Network Working Group J. Lennox Internet-Draft Vidyo Intended status: Informational K. Gross Expires: July 26, 2015 AVA S. Nandakumar G. Salgueiro Cisco Systems B. Burman Ericsson January 22, 2015 A Taxonomy of Grouping Semantics and Mechanisms for Real-Time Transport Protocol (RTP) Sources draft-ietf-avtext-rtp-grouping-taxonomy-05 Abstract The terminology about, and associations among, Real-Time Transport Protocol (RTP) sources can be complex and somewhat opaque. This document describes a number of existing and proposed relationships among RTP sources, and attempts to define common terminology for discussing protocol entities and their relationships. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on July 26, 2015. Copyright Notice Copyright (c) 2015 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents Lennox, et al. Expires July 26, 2015 [Page 1] Internet-Draft RTP Grouping Taxonomy January 2015 (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 2. Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.1. Media Chain . . . . . . . . . . . . . . . . . . . . . . . 4 2.1.1. Physical Stimulus . . . . . . . . . . . . . . . . . . 8 2.1.2. Media Capture . . . . . . . . . . . . . . . . . . . . 8 2.1.3. Raw Stream . . . . . . . . . . . . . . . . . . . . . 8 2.1.4. Media Source . . . . . . . . . . . . . . . . . . . . 8 2.1.5. Source Stream . . . . . . . . . . . . . . . . . . . . 9 2.1.6. Media Encoder . . . . . . . . . . . . . . . . . . . . 10 2.1.7. Encoded Stream . . . . . . . . . . . . . . . . . . . 11 2.1.8. Dependent Stream . . . . . . . . . . . . . . . . . . 11 2.1.9. Media Packetizer . . . . . . . . . . . . . . . . . . 11 2.1.10. RTP Stream . . . . . . . . . . . . . . . . . . . . . 12 2.1.11. RTP-based Redundancy . . . . . . . . . . . . . . . . 13 2.1.12. Redundancy RTP Stream . . . . . . . . . . . . . . . . 13 2.1.13. Media Transport . . . . . . . . . . . . . . . . . . . 13 2.1.14. Media Transport Sender . . . . . . . . . . . . . . . 14 2.1.15. Sent RTP Stream . . . . . . . . . . . . . . . . . . . 15 2.1.16. Network Transport . . . . . . . . . . . . . . . . . . 15 2.1.17. Transported RTP Stream . . . . . . . . . . . . . . . 15 2.1.18. Media Transport Receiver . . . . . . . . . . . . . . 15 2.1.19. Received RTP Stream . . . . . . . . . . . . . . . . . 15 2.1.20. Received Redundancy RTP Stream . . . . . . . . . . . 16 2.1.21. RTP-based Repair . . . . . . . . . . . . . . . . . . 16 2.1.22. Repaired RTP Stream . . . . . . . . . . . . . . . . . 16 2.1.23. Media Depacketizer . . . . . . . . . . . . . . . . . 16 2.1.24. Received Encoded Stream . . . . . . . . . . . . . . . 16 2.1.25. Media Decoder . . . . . . . . . . . . . . . . . . . . 16 2.1.26. Received Source Stream . . . . . . . . . . . . . . . 17 2.1.27. Media Sink . . . . . . . . . . . . . . . . . . . . . 17 2.1.28. Received Raw Stream . . . . . . . . . . . . . . . . . 17 2.1.29. Media Render . . . . . . . . . . . . . . . . . . . . 17 2.2. Communication Entities . . . . . . . . . . . . . . . . . 18 2.2.1. Endpoint . . . . . . . . . . . . . . . . . . . . . . 19 2.2.2. RTP Session . . . . . . . . . . . . . . . . . . . . . 19 2.2.3. Participant . . . . . . . . . . . . . . . . . . . . . 20 2.2.4. Multimedia Session . . . . . . . . . . . . . . . . . 20 2.2.5. Communication Session . . . . . . . . . . . . . . . . 21 Lennox, et al. Expires July 26, 2015 [Page 2] Internet-Draft RTP Grouping Taxonomy January 2015 3. Concepts of Inter-Relations . . . . . . . . . . . . . . . . . 21 3.1. Synchronization Context . . . . . . . . . . . . . . . . . 21 3.1.1. RTCP CNAME . . . . . . . . . . . . . . . . . . . . . 22 3.1.2. Clock Source Signaling . . . . . . . . . . . . . . . 22 3.1.3. Implicitly via RtcMediaStream . . . . . . . . . . . . 22 3.1.4. Explicitly via SDP Mechanisms . . . . . . . . . . . . 22 3.2. Endpoint . . . . . . . . . . . . . . . . . . . . . . . . 22 3.3. Participant . . . . . . . . . . . . . . . . . . . . . . . 23 3.4. RtcMediaStream . . . . . . . . . . . . . . . . . . . . . 23 3.5. Single- and Multi-Session Transmission of Dependent Streams . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.6. Multi-Channel Audio . . . . . . . . . . . . . . . . . . . 24 3.7. Simulcast . . . . . . . . . . . . . . . . . . . . . . . . 24 3.8. Layered Multi-Stream . . . . . . . . . . . . . . . . . . 25 3.9. RTP Stream Duplication . . . . . . . . . . . . . . . . . 27 3.10. Redundancy Format . . . . . . . . . . . . . . . . . . . . 27 3.11. RTP Retransmission . . . . . . . . . . . . . . . . . . . 28 3.12. Forward Error Correction . . . . . . . . . . . . . . . . 30 3.13. RTP Stream Separation . . . . . . . . . . . . . . . . . . 31 3.14. Multiple RTP Sessions over one Media Transport . . . . . 32 4. Mapping from Existing Terms . . . . . . . . . . . . . . . . . 32 4.1. Telepresence Terms . . . . . . . . . . . . . . . . . . . 32 4.1.1. Audio Capture . . . . . . . . . . . . . . . . . . . . 32 4.1.2. Capture Device . . . . . . . . . . . . . . . . . . . 32 4.1.3. Capture Encoding . . . . . . . . . . . . . . . . . . 32 4.1.4. Capture Scene . . . . . . . . . . . . . . . . . . . . 33 4.1.5. Endpoint . . . . . . . . . . . . . . . . . . . . . . 33 4.1.6. Individual Encoding . . . . . . . . . . . . . . . . . 33 4.1.7. Media Capture . . . . . . . . . . . . . . . . . . . . 33 4.1.8. Media Consumer . . . . . . . . . . . . . . . . . . . 33 4.1.9. Media Provider . . . . . . . . . . . . . . . . . . . 33 4.1.10. Stream . . . . . . . . . . . . . . . . . . . . . . . 33 4.1.11. Video Capture . . . . . . . . . . . . . . . . . . . . 33 4.2. Media Description . . . . . . . . . . . . . . . . . . . . 33 4.3. Media Stream . . . . . . . . . . . . . . . . . . . . . . 34 4.4. Multimedia Conference . . . . . . . . . . . . . . . . . . 34 4.5. Multimedia Session . . . . . . . . . . . . . . . . . . . 34 4.6. Multipoint Control Unit (MCU) . . . . . . . . . . . . . . 34 4.7. Recording Device . . . . . . . . . . . . . . . . . . . . 34 4.8. RtcMediaStream . . . . . . . . . . . . . . . . . . . . . 35 4.9. RtcMediaStreamTrack . . . . . . . . . . . . . . . . . . . 35 4.10. RTP Sender . . . . . . . . . . . . . . . . . . . . . . . 35 4.11. RTP Session . . . . . . . . . . . . . . . . . . . . . . . 35 4.12. SSRC . . . . . . . . . . . . . . . . . . . . . . . . . . 35 5. Security Considerations . . . . . . . . . . . . . . . . . . . 35 6. Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . 36 7. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 36 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 36 Lennox, et al. Expires July 26, 2015 [Page 3] Internet-Draft RTP Grouping Taxonomy January 2015 9. Informative References . . . . . . . . . . . . . . . . . . . 36 Appendix A. Changes From Earlier Versions . . . . . . . . . . . 38 A.1. Modifications Between WG Version -04 and -05 . . . . . . 38 A.2. Modifications Between WG Version -03 and -04 . . . . . . 38 A.3. Modifications Between WG Version -02 and -03 . . . . . . 39 A.4. Modifications Between WG Version -01 and -02 . . . . . . 39 A.5. Modifications Between WG Version -00 and -01 . . . . . . 40 A.6. Modifications Between Version -02 and -03 . . . . . . . . 40 A.7. Modifications Between Version -01 and -02 . . . . . . . . 41 A.8. Modifications Between Version -00 and -01 . . . . . . . . 41 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 41 1. Introduction The existing taxonomy of sources in RTP is often regarded as confusing and inconsistent. Consequently, a deep understanding of how the different terms relate to each other becomes a real challenge. Frequently cited examples of this confusion are (1) how different protocols that make use of RTP use the same terms to signify different things and (2) how the complexities addressed at one layer are often glossed over or ignored at another. This document attempts to provide some clarity by reviewing the semantics of various aspects of sources in RTP. As an organizing mechanism, it approaches this by describing various ways that RTP sources can be grouped and associated together. All non-specific references to ControLling mUltiple streams for tElepresence (CLUE) in this document map to [I-D.ietf-clue-framework] and all references to Web Real-Time Communications (WebRTC) map to [I-D.ietf-rtcweb-overview]. 2. Concepts This section defines concepts that serve to identify and name various transformations and streams in a given RTP usage. For each concept an attempt is made to list any alternate definitions and usages that co-exist today along with various characteristics that further describes the concept. These concepts are divided into two categories, one related to the chain of streams and transformations that media can be subject to, the other for entities involved in the communication. 2.1. Media Chain In the context of this memo, Media is a sequence of synthetic or Physical Stimuli (Section 2.1.1) (sound waves, photons, key-strokes), Lennox, et al. Expires July 26, 2015 [Page 4] Internet-Draft RTP Grouping Taxonomy January 2015 represented in digital form. Synthesized Media is typically generated directly in the digital domain. This section contains the concepts that can be involved in taking Media at a sender side and transporting it to a receiver, which may recover a sequence of physical stimuli. This chain of concepts is of two main types, streams and transformations. Streams are time-based sequences of samples of the physical stimulus in various representations, while transformations changes the representation of the streams in some way. The below examples are basic ones and it is important to keep in mind that this conceptual model enables more complex usages. Some will be further discussed in later sections of this document. In general the following applies to this model: o A transformation may have zero or more inputs and one or more outputs. o A stream is of some type, such as audio, video, real-time text, etc. o A stream has one source transformation and one or more sink transformations (with the exception of Physical Stimulus (Section 2.1.1) that may lack source or sink transformation). o Streams can be forwarded from a transformation output to any number of inputs on other transformations that support that type. o If the output of a transformation is sent to multiple transformations, those streams will be identical; it takes a transformation to make them different. o There are no formal limitations on how streams are connected to transformations, this may include loops if required by a particular transformation. It is also important to remember that this is a conceptual model. Thus real-world implementations may look different and have different structure. To provide a basic understanding of the relationships in the chain we first introduce the concepts for the sender side (Figure 1). This covers physical stimuli until media packets are emitted onto the network. Lennox, et al. Expires July 26, 2015 [Page 5] Internet-Draft RTP Grouping Taxonomy January 2015 Physical Stimulus | V +--------------------+ | Media Capture | +--------------------+ | Raw Stream V +--------------------+ | Media Source |<- Synchronization Timing +--------------------+ | Source Stream V +--------------------+ | Media Encoder | +--------------------+ | Encoded Stream +------------+ V | V +--------------------+ | +----------------------+ | Media Packetizer | | | RTP-based Redundancy | +--------------------+ | +----------------------+ | | | +------------+ Redundancy RTP Stream Source RTP Stream | V V +--------------------+ +--------------------+ | Media Transport | | Media Transport | +--------------------+ +--------------------+ Figure 1: Sender Side Concepts in the Media Chain In Figure 1 we have included a branched chain to cover the concepts for using redundancy to improve the reliability of the transport. The Media Transport concept is an aggregate that is decomposed in Section 2.1.13. In Figure 2 we review a receiver media chain matching the sender side, to look at the inverse transformations and their attempts to recover identical streams as in the sender chain, subject to what may be lossy compression and imperfect Media Transport. Note that the streams out of a reverse transformation, like the Source Stream out the Media Decoder are in many cases not the same as the corresponding ones on the sender side, thus they are prefixed with a "Received" to denote a potentially modified version. The reason for not being the same lies in the transformations that can be of irreversible type. Lennox, et al. Expires July 26, 2015 [Page 6] Internet-Draft RTP Grouping Taxonomy January 2015 For example, lossy source coding in the Media Encoder prevents the Source Stream out of the Media Decoder to be the same as the one fed into the Media Encoder. Other reasons include packet loss or late loss in the Media Transport transformation that even RTP-based Repair, if used, fails to repair. However, some transformations are not always present, like RTP-based Repair that cannot operate without Redundancy RTP Streams. +--------------------+ +--------------------+ | Media Transport | | Media Transport | +--------------------+ +--------------------+ | | Received RTP Stream Received Redundancy RTP Stream | | | +-------------------+ V V +--------------------+ | RTP-based Repair | +--------------------+ | Repaired RTP Stream V +--------------------+ | Media Depacketizer | +--------------------+ | Received Encoded Stream V +--------------------+ | Media Decoder | +--------------------+ | Received Source Stream V +--------------------+ | Media Sink |--> Synchronization Information +--------------------+ | Received Raw Stream V +--------------------+ | Media Renderer | +--------------------+ | V Physical Stimulus Figure 2: Receiver Side Concepts of the Media Chain Lennox, et al. Expires July 26, 2015 [Page 7] Internet-Draft RTP Grouping Taxonomy January 2015 2.1.1. Physical Stimulus The physical stimulus is a physical event that can be sampled and converted to digital form by an appropriate sensor or transducer. This include sound waves making up audio, photons in a light field, or other excitations or interactions with sensors, like keystrokes on a keyboard. 2.1.2. Media Capture Media Capture is the process of transforming the Physical Stimulus (Section 2.1.1) into digital Media using an appropriate sensor or transducer. The Media Capture performs a digital sampling of the physical stimulus, usually periodically, and outputs this in some representation as a Raw Stream (Section 2.1.3). This data is due to its periodical sampling, or at least being timed asynchronous events, some form of a stream of media data. The Media Capture is normally instantiated in some type of device, i.e. media capture device. Examples of different types of media capturing devices are digital cameras, microphones connected to A/D converters, or keyboards. Characteristics: o A Media Capture is identified either by hardware/manufacturer ID or via a session-scoped device identifier as mandated by the application usage. o A Media Capture can generate an Encoded Stream (Section 2.1.7) if the capture device support such a configuration. o The nature of the Media Capture may impose constraints on the clock handling in some of the subsequent steps. For example, many audio or video capture devices are not completely free in selecting the sample rate. 2.1.3. Raw Stream The time progressing stream of digitally sampled information, usually periodically sampled and provided by a Media Capture (Section 2.1.2). A Raw Stream can also contain synthesized Media that may not require any explicit Media Capture, since it is already in an appropriate digital form. 2.1.4. Media Source A Media Source is the logical source of a reference clock synchronized, time progressing, digital media stream, called a Source Stream (Section 2.1.5). This transformation takes one or more Raw Lennox, et al. Expires July 26, 2015 [Page 8] Internet-Draft RTP Grouping Taxonomy January 2015 Streams (Section 2.1.3) and provides a Source Stream as output. The output is synchronized with a reference clock (Section 3.1), which can be as simple as a system local wall clock or as complex as NTP synchronized. The output can be of different types. One type is directly associated with a particular Media Capture's Raw Stream. Others are more conceptual sources, like an audio mix of multiple Source Streams (Figure 3). Mixing multiple streams typically requires that the input streams are possible to relate in time, meaning that they have to be Source Streams (Section 2.1.5) rather than Raw Streams. In Figure 3, the generated Source Stream is a mix of the three input Source Streams. Source Source Source Stream Stream Stream | | | V V V +--------------------------+ | Media Source |<-- Reference Clock | Mixer | +--------------------------+ | V Source Stream Figure 3: Conceptual Media Source in form of Audio Mixer Another possible example of a conceptual Media Source is a video surveillance switch, where the input is multiple Source Streams from different cameras, and the output is one of those Source Streams based on some selection criteria, like a round-robin or based on some video activity measure. Characteristics: o At any point, it can represent a physical captured source or conceptual source. 2.1.5. Source Stream A time progressing stream of digital samples that has been synchronized with a reference clock and comes from particular Media Source (Section 2.1.4). Lennox, et al. Expires July 26, 2015 [Page 9] Internet-Draft RTP Grouping Taxonomy January 2015 2.1.6. Media Encoder A Media Encoder is a transform that is responsible for encoding the media data from a Source Stream (Section 2.1.5) into another representation, usually more compact, that is output as an Encoded Stream (Section 2.1.7). The Media Encoder step commonly includes pre-encoding transformations, such as scaling, resampling etc. The Media Encoder can have a significant number of configuration options that affects the properties of the Encoded Stream. This include properties such as bit-rate, start points for decoding, resolution, bandwidth or other fidelity affecting properties. The actually used codec is also an important factor in many communication systems. Scalable Media Encoders need special attention as they produce multiple outputs that are potentially of different types. As shown in Figure 4, a scalable Media Encoder takes one input Source Stream and encodes it into multiple output streams of two different types; at least one Encoded Stream that is independently decodable and one or more Dependent Streams (Section 2.1.8). Decoding requires at least one Encoded Stream and zero or more Dependent Streams. A Dependent Stream's dependency is one of the grouping relations this document discusses further in Section 3.8. Source Stream | V +--------------------------+ | Scalable Media Encoder | +--------------------------+ | | ... | V V V Encoded Dependent Dependent Stream Stream Stream Figure 4: Scalable Media Encoder Input and Outputs There are also other variants of encoders, like so-called Multiple Description Coding (MDC). Such Media Encoder produce multiple independent and thus individually decodable Encoded Streams. However, (logically) combining multiple of these Encoded Streams into a single Received Source Stream during decoding leads to an improvement in perceptual reproduced quality when compared to decoding a single Encoded Stream. Lennox, et al. Expires July 26, 2015 [Page 10] Internet-Draft RTP Grouping Taxonomy January 2015 Creating multiple Encoded Streams from the same Source Stream, where the Encoded Streams are neither in a scalable nor in an MDC relationship is commonly utilized in Simulcast environments. Characteristics: o A Media Source can be multiply encoded by different Media Encoders to provide various encoded representations. 2.1.7. Encoded Stream A stream of time synchronized encoded media that can be independently decoded. Characteristics: o Due to temporal dependencies, an Encoded Stream may have limitations in where decoding can be started. These entry points, for example Intra frames from a video encoder, may require identification and their generation may be event based or configured to occur periodically. 2.1.8. Dependent Stream A stream of time synchronized encoded media fragments that are dependent on one or more Encoded Streams (Section 2.1.7) and zero or more Dependent Streams to be possible to decode. Characteristics: o Each Dependent Stream has a set of dependencies. These dependencies must be understood by the parties in a Multimedia Session that intend to use a Dependent Stream. 2.1.9. Media Packetizer The transformation of taking one or more Encoded (Section 2.1.7) or Dependent Streams (Section 2.1.8) and put their content into one or more sequences of packets, normally RTP packets, and output Source RTP Streams (Section 2.1.10). This step includes both generating RTP payloads as well as RTP packets. The Media Packetizer can use multiple inputs when producing a single RTP Stream. One such example is SRST packetization when using Scalable Video Coding (SVC) (Section 3.5). The Media Packetizer can also produce multiple RTP Streams, for example when Encoded and/or Dependent Streams are distributed over Lennox, et al. Expires July 26, 2015 [Page 11] Internet-Draft RTP Grouping Taxonomy January 2015 multiple RTP Streams. One example of this is MRMT packetization when using SVC (Section 3.5). Characteristics: o The Media Packetizer will select which Synchronization source(s) (SSRC) [RFC3550] in which RTP Sessions that are used. o Media Packetizer can combine multiple Encoded or Dependent Streams into one or more RTP Streams. 2.1.10. RTP Stream A stream of RTP packets containing media data, source or redundant. The RTP Stream is identified by an SSRC belonging to a particular RTP Session. The RTP Session is identified as discussed in Section 2.2.2. A Source RTP Stream is a RTP Stream containing at least some content from an Encoded Stream (Section 2.1.7). Source material is any media material that is produced for transport over RTP without any additional RTP-based redundancy applied. Note that RTP-based redundancy excludes the type of redundancy that most suitable Media Encoders (Section 2.1.6) may add to the media format of the Encoded Stream that makes it cope better with inevitable RTP packet losses. This is further described in RTP-based Redundancy (Section 2.1.11) and Redundancy RTP Stream (Section 2.1.12). Characteristics: o Each RTP Stream is identified by a Synchronization source (SSRC) [RFC3550] that is carried in every RTP and RTP Control Protocol (RTCP) packet header. The SSRC is unique in a specific RTP Session context. o At any given point in time, a RTP Stream can have one and only one SSRC, but SSRCs for a given RTP Stream can change over time. SSRC collision and clock rate change [RFC7160] are examples of valid reasons to change SSRC for an RTP Stream. In those cases, the RTP Stream itself is not changed in any significant way, only the identifying SSRC number. o Each SSRC defines a unique RTP sequence numbering and timing space. o Several RTP Streams, each with their own SSRC, may represent a single Media Source. Lennox, et al. Expires July 26, 2015 [Page 12] Internet-Draft RTP Grouping Taxonomy January 2015 o Several RTP Streams, each with their own SSRC, can be carried in a single RTP Session. 2.1.11. RTP-based Redundancy RTP-based Redundancy is defined here as a transformation that generates redundant or repair packets sent out as a Redundancy RTP Stream (Section 2.1.12) to mitigate network transport impairments, like packet loss and delay. The RTP-based Redundancy exists in many flavors; they may be generating independent Repair Streams that are used in addition to the Source Stream (like RTP Retransmission (Section 3.11) and some special types of Forward Error Correction, like RTP stream duplication (Section 3.9)), they may generate a new Source Stream by combining redundancy information with source information (Using XOR FEC (Section 3.12) as a redundancy payload (Section 3.10)), or completely replace the source information with only redundancy packets. 2.1.12. Redundancy RTP Stream A RTP Stream (Section 2.1.10) that contains no original source data, only redundant data that may be combined with one or more Received RTP Stream (Section 2.1.19) to produce Repaired RTP Streams (Section 2.1.22). 2.1.13. Media Transport A Media Transport defines the transformation that the RTP Streams (Section 2.1.10) are subjected to by the end-to-end transport from one RTP sender to one specific RTP receiver (an RTP Session (Section 2.2.2) may contain multiple RTP receivers per sender). Each Media Transport is defined by a transport association that is normally identified by a 5-tuple (source address, source port, destination address, destination port, transport protocol), but a proposal exists for sending multiple transport associations on a single 5-tuple [I-D.westerlund-avtcore-transport-multiplexing]. Characteristics: o Media Transport transmits RTP Streams of RTP Packets from a source transport address to a destination transport address. o Each Media Transport contains only a single RTP Session. o A single RTP Session can span multiple Media Transports. Lennox, et al. Expires July 26, 2015 [Page 13] Internet-Draft RTP Grouping Taxonomy January 2015 The Media Transport concept sometimes needs to be decomposed into more steps to enable discussion of what a sender emits that gets transformed by the network before it is received by the receiver. Thus we provide also this Media Transport decomposition (Figure 5). RTP Stream | V +--------------------------+ | Media Transport Sender | +--------------------------+ | Sent RTP Stream V +--------------------------+ | Network Transport | +--------------------------+ | Transported RTP Stream V +--------------------------+ | Media Transport Receiver | +--------------------------+ | V Received RTP Stream Figure 5: Decomposition of Media Transport 2.1.14. Media Transport Sender The first transformation within the Media Transport (Section 2.1.13) is the Media Transport Sender. The sending Endpoint (Section 2.2.1) takes an RTP Stream and emits the packets onto the network using the transport association established for this Media Transport, thereby creating a Sent RTP Stream (Section 2.1.15). In the process, it transforms the RTP Stream in several ways. First, it generates the necessary protocol headers for the transport association, for example IP and UDP headers, thus forming IP/UDP/RTP packets. In addition, the Media Transport Sender may queue, pace or otherwise affect how the packets are emitted onto the network, thereby potentially introducing delay, jitter and inter packet spacings that characterize the Sent RTP Stream. Lennox, et al. Expires July 26, 2015 [Page 14] Internet-Draft RTP Grouping Taxonomy January 2015 2.1.15. Sent RTP Stream The Sent RTP Stream is the RTP Stream as entering the first hop of the network path to its destination. The Sent RTP Stream is identified using network transport addresses, like for IP/UDP the 5-tuple (source IP address, source port, destination IP address, destination port, and protocol (UDP)). 2.1.16. Network Transport Network Transport is the transformation that subjects the Sent RTP Stream (Section 2.1.15) to traveling from the source to the destination through the network. This transformation can result in loss of some packets, varying delay on a per packet basis, packet duplication, and packet header or data corruption. This transformation produces a Transported RTP Stream (Section 2.1.17) at the exit of the network path. 2.1.17. Transported RTP Stream The RTP Stream that is emitted out of the network path at the destination, subjected to the Network Transport's transformation (Section 2.1.16). 2.1.18. Media Transport Receiver The receiver Endpoint's (Section 2.2.1) transformation of the Transported RTP Stream (Section 2.1.17) by its reception process, which results in the Received RTP Stream (Section 2.1.19). This transformation includes transport checksums being verified. Sensible system designs typically either discard packets with mis-matching checksums, or pass them on while somehow marking them in the resulting Received RTP Stream so to alarm subsequent transformations about the possible corrupt state. In this context it is worth noting that there is typically some probability for corrupt packets to pass through undetected (with a seemingly correct checksum). Other transformations can compensate for delay variations in receiving a packet on the network interface and providing it to the application (de-jitter buffer). 2.1.19. Received RTP Stream The RTP Stream (Section 2.1.10) resulting from the Media Transport's transformation, i.e. subjected to packet loss, packet corruption, packet duplication and varying transmission delay from sender to receiver. Lennox, et al. Expires July 26, 2015 [Page 15] Internet-Draft RTP Grouping Taxonomy January 2015 2.1.20. Received Redundancy RTP Stream The Redundancy RTP Stream (Section 2.1.12) resulting from the Media Transport transformation, i.e. subjected to packet loss, packet corruption, and varying transmission delay from sender to receiver. 2.1.21. RTP-based Repair RTP-based Repair is a Transformation that takes as input one or more Received RTP Streams (Section 2.1.19) and Received Redundancy RTP Streams (Section 2.1.20), and produces one or more Repaired RTP Streams (Section 2.1.22) that are as close to the corresponding sent Source RTP Streams (Section 2.1.10) as possible, using different RTP- based repair methods, for example the ones referred in RTP-based Redundancy (Section 2.1.11). 2.1.22. Repaired RTP Stream A Received RTP Stream (Section 2.1.19) for which Received Redundancy RTP Stream (Section 2.1.20) information has been used to try to recover the Source RTP Stream (Section 2.1.10) as it was before Media Transport (Section 2.1.13). 2.1.23. Media Depacketizer A Media Depacketizer takes one or more RTP Streams (Section 2.1.10), depacketizes them, and attempts to reconstitute the Encoded Streams (Section 2.1.7) or Dependent Streams (Section 2.1.8) present in those RTP Streams. In practical implementations, the Media Depacketizer and the Media Decoder may be tightly coupled and share information to improve or optimize the overall decoding and error concealment process. It is, however, not expected that there would be any benefit in defining a taxonomy for those detailed (and likely very implementation- dependent) steps. 2.1.24. Received Encoded Stream The received version of an Encoded Stream (Section 2.1.7). 2.1.25. Media Decoder A Media Decoder is a transformation that is responsible for decoding Encoded Streams (Section 2.1.7) and any Dependent Streams (Section 2.1.8) into a Source Stream (Section 2.1.5). Lennox, et al. Expires July 26, 2015 [Page 16] Internet-Draft RTP Grouping Taxonomy January 2015 In practical implementations, the Media Decoder and the Media Depacketizer may be tightly coupled and share information to improve or optimize the overall decoding process in various ways. It is however not expected that there would be any benefit in defining a taxonomy for those detailed (and likely very implementation- dependent) steps. Characteristics: o A Media Decoder has to deal with any errors in the Encoded Streams that resulted from corruption or failure to repair packet losses. Therefore, it commonly is robust to error and losses, and includes concealment methods. 2.1.26. Received Source Stream The received version of a Source Stream (Section 2.1.5). 2.1.27. Media Sink The Media Sink receives a Source Stream (Section 2.1.5) that contains, usually periodically, sampled media data together with associated synchronization information. Depending on application, this Source Stream then needs to be transformed into a Raw Stream (Section 2.1.3) that is conveyed to the Media Render (Section 2.1.29), synchronized with the output from other Media Sinks. The Media Sink may also be connected with a Media Source (Section 2.1.4) and be used as part of a conceptual Media Source. Characteristics: o The Media Sink can further transform the Source Stream into a representation that is suitable for rendering on the Media Render as defined by the application or system-wide configuration. This include sample scaling, level adjustments etc. 2.1.28. Received Raw Stream The received version of a Raw Stream (Section 2.1.3). 2.1.29. Media Render A Media Render takes a Raw Stream (Section 2.1.3) and converts it into Physical Stimulus (Section 2.1.1) that a human user can perceive. Examples of such devices are screens, and D/A converters connected to amplifiers and loudspeakers. Characteristics: Lennox, et al. Expires July 26, 2015 [Page 17] Internet-Draft RTP Grouping Taxonomy January 2015 o An Endpoint can potentially have multiple Media Renders for each media type. 2.2. Communication Entities This section contains concept for entities involved in the communication. +------------------------------------------------------------+ | Communication Session | | | | +----------------+ +----------------+ | | | Participant A | +------------+ | Participant B | | | | | | Multimedia | | | | | | +-------------+|<==>| Session |<==>|+-------------+ | | | | | Endpoint A || | | || Endpoint B | | | | | | || +------------+ || | | | | | | +-----------++----------------------++-----------+ | | | | | | | | | | | | | | | | | RTP Session|---Media Transport--->| | | | | | | | | Audio |<---Media Transport---| | | | | | | | | | ^ | | | | | | | | +-----------++----------|-----------++-----------+ | | | | | | || v || | | | | | | || +-----------------+ || | | | | | | || | Synchronization | || | | | | | | || | Context | || | | | | | | || +-----------------+ || | | | | | | || ^ || | | | | | | +-----------++----------|-----------++-----------+ | | | | | | | | v | | | | | | | | | RTP Session|<---Media Transport---| | | | | | | | | Video |---Media Transport--->| | | | | | | | | | | | | | | | | | +-----------++----------------------++-----------+ | | | | | +-------------+| |+-------------+ | | | +----------------+ +----------------+ | +------------------------------------------------------------+ Figure 6: Example Point to Point Communication Session with two RTP Sessions Figure 6 shows a high-level example representation of a very basic point-to-point Communication Session between Participants A and B. It uses two different audio and video RTP Sessions between A's and B's Endpoints, using separate Media Transports for those RTP Sessions. The Multimedia Session shared by the Participants can, for example, be established using SIP (i.e., there is a SIP Dialog Lennox, et al. Expires July 26, 2015 [Page 18] Internet-Draft RTP Grouping Taxonomy January 2015 between A and B). The terms used in Figure 6 are further elaborated in the sub-sections below. 2.2.1. Endpoint A single addressable entity sending or receiving RTP packets. It may be decomposed into several functional blocks, but as long as it behaves as a single RTP stack entity it is classified as a single "Endpoint". Characteristics: o Endpoints can be identified in several different ways. While RTCP Canonical Names (CNAMEs) [RFC3550] provide a globally unique and stable identification mechanism for the duration of the Communication Session (see Section 2.2.5), their validity applies exclusively within a Synchronization Context (Section 3.1). Thus one Endpoint can handle multiple CNAMEs, each of which can be shared among a set of Endpoints belonging to the same Participant (Section 2.2.3). Therefore, mechanisms outside the scope of RTP, such as application defined mechanisms, must be used to ensure Endpoint identification when outside this Synchronization Context. o An Endpoint can be associated with at most one Participant (Section 2.2.3) at any single point in time. o In some contexts, an Endpoint would typically correspond to a single "host", for example a computer using a single network interface and being used by a single human user. 2.2.2. RTP Session An RTP Session is an association among a group of Participants communicating with RTP. It is a group communications channel which can potentially carry a number of RTP Streams. Within an RTP Session, every Participant can find meta-data and control information (over RTCP) about all the RTP Streams in the RTP Session. The bandwidth of the RTCP control channel is shared between all Participants within an RTP Session. Characteristics: o An RTP Session can carry one ore more RTP Streams. o An RTP Session shares a single SSRC space as defined in RFC3550 [RFC3550]. That is, the Endpoints participating in an RTP Session can see an SSRC identifier transmitted by any of the other Endpoints. An Endpoint can receive an SSRC either as SSRC or as a Lennox, et al. Expires July 26, 2015 [Page 19] Internet-Draft RTP Grouping Taxonomy January 2015 Contributing source (CSRC) in RTP and RTCP packets, as defined by the Endpoints' network interconnection topology. o An RTP Session uses at least two Media Transports (Section 2.1.13), one for sending and one for receiving. Commonly, the receiving Media Transport is the reverse direction of the Media Transport used for sending. An RTP Session may use many Media Transports and these define the session's network interconnection topology. o A single Media Transport always carries a single RTP Session. o Multiple RTP Sessions can be conceptually related, for example originating from or targeted for the same Participant (Section 2.2.3) or Endpoint (Section 2.2.1), or by containing RTP Streams that are somehow related (Section 3). 2.2.3. Participant A Participant is an entity reachable by a single signaling address, and is thus related more to the signaling context than to the media context. Characteristics: o A single signaling-addressable entity, using an application- specific signaling address space, for example a SIP URI. o A Participant can participate in several Multimedia Sessions (Section 2.2.4). o A Participant can be comprised of several associated Endpoints (Section 2.2.1). 2.2.4. Multimedia Session A Multimedia Session is an association among a group of Participants (Section 2.2.3) engaged in the communication via one or more RTP Sessions (Section 2.2.2). It defines logical relationships among Media Sources (Section 2.1.4) that appear in multiple RTP Sessions. Characteristics: o A Multimedia Session can be composed of several RTP Sessions with potentially multiple RTP Streams per RTP Session. o Each Participant in a Multimedia Session can have a multitude of Media Captures and Media Rendering devices. Lennox, et al. Expires July 26, 2015 [Page 20] Internet-Draft RTP Grouping Taxonomy January 2015 o A single Multimedia Session can contain media from one or more Synchronization Contexts (Section 3.1). An example of that is a Multimedia Session containing one set of audio and video for communication purposes belonging to one Synchronization Context, and another set of audio and video for presentation purposes (like playing a video file) with a separate Synchronization Context that has no strong timing relationship and need not be strictly synchronized with the audio and video used for communication. 2.2.5. Communication Session A Communication Session is an association among two or more Participants (Section 2.2.3) communicating with each other via one or more Multimedia Sessions (Section 2.2.4). Characteristics: o Each Participant in a Communication Session is identified via an application-specific signaling address. o A Communication Session is composed of Participants that share at least one Multimedia Session, involving one or more parallel RTP Sessions with potentially multiple RTP Streams per RTP Session. For example, in a full mesh communication, the Communication Session consists of a set of separate Multimedia Sessions between each pair of Participants. Another example is a centralized conference, where the Communication Session consists of a set of Multimedia Sessions between each Participant and the conference handler. 3. Concepts of Inter-Relations This section uses the concepts from previous sections, and looks at different types of relationships among them. These relationships occur at different abstraction levels and for different purposes, but the reason for the needed relationship at a certain step in the media handling chain may exist at another step. For example, the use of Simulcast (Section 3.7)) implies a need to determine relations at RTP Stream level, but the underlying reason is that multiple Media Encoders use the same Media Source, i.e. to be able to identify a common Media Source. 3.1. Synchronization Context A Synchronization Context defines a requirement on a strong timing relationship between the Media Sources, typically requiring alignment of clock sources. Such a relationship can be identified in multiple ways as listed below. A single Media Source can only belong to a Lennox, et al. Expires July 26, 2015 [Page 21] Internet-Draft RTP Grouping Taxonomy January 2015 single Synchronization Context, since it is assumed that a single Media Source can only have a single media clock and requiring alignment to several Synchronization Contexts (and thus reference clocks) will effectively merge those into a single Synchronization Context. 3.1.1. RTCP CNAME RFC3550 [RFC3550] describes Inter-media synchronization between RTP Sessions based on RTCP CNAME, RTP and Network Time Protocol (NTP) [RFC5905] formatted timestamps of a reference clock. As indicated in [RFC7273], despite using NTP format timestamps, it is not required that the clock be synchronized to an NTP source. 3.1.2. Clock Source Signaling [RFC7273] provides a mechanism to signal the clock source in Session Description Protocol (SDP) [RFC4566] both for the reference clock as well as the media clock, thus allowing a Synchronization Context to be defined beyond the one defined by the usage of CNAME source descriptions. 3.1.3. Implicitly via RtcMediaStream WebRTC defines "RtcMediaStream" with one or more "RtcMediaStreamTracks". All tracks in a "RtcMediaStream" are intended to be synchronized when rendered, implying that they must be generated such that synchronization is possible. 3.1.4. Explicitly via SDP Mechanisms The SDP Grouping Framework [RFC5888] defines an m= line (Section 4.2) grouping mechanism called "Lip Synchronization" (with LS identification-tag) for establishing the synchronization requirement across m= lines when they map to individual sources. Source-Specific Media Attributes in SDP [RFC5576] extends the above mechanism when multiple Media Sources are described by a single m= line. 3.2. Endpoint Some applications requires knowledge of what Media Sources originate from a particular Endpoint (Section 2.2.1). This can include such decisions as packet routing between parts of the topology, knowing the Endpoint origin of the RTP Streams. Lennox, et al. Expires July 26, 2015 [Page 22] Internet-Draft RTP Grouping Taxonomy January 2015 In RTP, this identification has been overloaded with the Synchronization Context (Section 3.1) through the usage of the RTCP source description CNAME (Section 3.1.1). This works for some usages, but in others it breaks down. For example, if an Endpoint has two sets of Media Sources that have different Synchronization Contexts, like the audio and video of the human Participant as well as a set of Media Sources of audio and video for a shared movie, CNAME would not be an appropriate identification for that Endpoint. Therefore, an Endpoint may have multiple CNAMEs. The CNAMEs or the Media Sources themselves can be related to the Endpoint. 3.3. Participant In communication scenarios, it is commonly needed to know which Media Sources originate from which Participant (Section 2.2.3). One reason is, for example, to enable the application to display Participant Identity information correctly associated with the Media Sources. This association is handled through the signaling solution to point at a specific Multimedia Session where the Media Sources may be explicitly or implicitly tied to a particular Endpoint. Participant information becomes more problematic due to Media Sources that are generated through mixing or other conceptual processing of Raw Streams or Source Streams that originate from different Participants. This type of Media Sources can thus have a dynamically varying set of origins and Participants. RTP contains the concept of CSRC that carry information about the previous step origin of the included media content on RTP level. 3.4. RtcMediaStream An RtcMediaStream in WebRTC is an explicit grouping of a set of Media Sources (RtcMediaStreamTracks) that share a common identifier and a single Synchronization Context (Section 3.1). 3.5. Single- and Multi-Session Transmission of Dependent Streams Scalable media coding formats such as, for example, H.264 based SVC [RFC6190] has two modes of operation: 1. In Single Session Transmission (SST), the SVC Media Encoder sends Encoded Streams (Section 2.1.7) and Dependent Streams (Section 2.1.8) as a single RTP Stream (Section 2.1.10) in a single RTP Session (Section 2.2.2), using the SVC RTP Payload format. Lennox, et al. Expires July 26, 2015 [Page 23] Internet-Draft RTP Grouping Taxonomy January 2015 2. In Multi-Session Transmission (MST), the SVC Media Encoder sends Encoded Streams and Dependent Streams distributed across multiple RTP Streams in one or more RTP Sessions. SST denotes one RTP Stream (SSRC) per Media Source in a single RTP Session. MST denotes one or more RTP Streams (SSRC) per Media Source in each of multiple RTP Sessions. The above is not unambiguously specified in the SVC payload format text [RFC6190], but it is what existing deployments of that RFC have implemented. The use of the term "RTP Session" in the SST/MST definition is somewhat misleading, since a single RTP Session can contain multiple RTP Streams. Also, it is sometimes useful to make a distinction between using a single Media Transport or multiple separate Media Transports when (in both cases) using multiple RTP Streams to carry Encoded Streams and Dependent Streams for a Media Source. Therefore, herein the following new terminology is defined: SRST: Single RTP Stream on a Single Media Transport MRST: Multiple RTP Streams on a Single Media Transport MRMT: Multiple RTP Streams on Multiple Media Transports 3.6. Multi-Channel Audio There exist a number of RTP payload formats that can carry multi- channel audio, despite the codec being a mono encoder. Multi-channel audio can be viewed as multiple Media Sources sharing a common Synchronization Context. These are independently encoded by a Media Encoder and the different Encoded Streams are packetized together in a time synchronized way into a single Source RTP Stream, using the used codec's RTP Payload format. Examples of codecs that support multi-channel audio are PCMA and PCMU [RFC3551], AMR [RFC4867], and G.719 [RFC5404]. 3.7. Simulcast A Media Source represented as multiple independent Encoded Streams constitutes a Simulcast or MDC of that Media Source. Figure 7 shows an example of a Media Source that is encoded into three separate Simulcast streams, that are in turn sent on the same Media Transport flow. When using Simulcast, the RTP Streams may be sharing RTP Session and Media Transport, or be separated on different RTP Sessions and Media Transports, or any combination of these two. It is other considerations that affect which usage is desirable, as discussed in Section 3.13. Lennox, et al. Expires July 26, 2015 [Page 24] Internet-Draft RTP Grouping Taxonomy January 2015 +----------------+ | Media Source | +----------------+ Source Stream | +----------------------+----------------------+ | | | V V V +------------------+ +------------------+ +------------------+ | Media Encoder | | Media Encoder | | Media Encoder | +------------------+ +------------------+ +------------------+ | Encoded | Encoded | Encoded | Stream | Stream | Stream V V V +------------------+ +------------------+ +------------------+ | Media Packetizer | | Media Packetizer | | Media Packetizer | +------------------+ +------------------+ +------------------+ | Source | Source | Source | RTP | RTP | RTP | Stream | Stream | Stream +-----------------+ | +-----------------+ | | | V V V +-------------------+ | Media Transport | +-------------------+ Figure 7: Example of Media Source Simulcast The Simulcast relation between the RTP Streams is the common Media Source. In addition, to be able to identify the common Media Source, a receiver of the RTP Stream may need to know which configuration or encoding goals that lay behind the produced Encoded Stream and its properties. This to enable selection of the stream that is most useful in the application at that moment. 3.8. Layered Multi-Stream Layered Multi-Stream (LMS) is a mechanism by which different portions of a layered encoding of a Source Stream are sent using separate RTP Streams (sometimes in separate RTP Sessions). LMSs are useful for receiver control of layered media. A Media Source represented as an Encoded Stream and multiple Dependent Streams constitutes a Media Source that has layered dependencies. Figure 8 represents an example of a Media Source that is encoded into three dependent layers, where two layers are sent on the same Media Transport using different RTP Streams, i.e. SSRCs, and the third layer is sent on a separate Media Transport. Lennox, et al. Expires July 26, 2015 [Page 25] Internet-Draft RTP Grouping Taxonomy January 2015 +----------------+ | Media Source | +----------------+ | | V +---------------------------------------------------------+ | Media Encoder | +---------------------------------------------------------+ | | | Encoded Stream Dependent Stream Dependent Stream | | | V V V +----------------+ +----------------+ +----------------+ |Media Packetizer| |Media Packetizer| |Media Packetizer| +----------------+ +----------------+ +----------------+ | | | RTP Stream RTP Stream RTP Stream | | | +------+ +------+ | | | | V V V +-----------------+ +-----------------+ | Media Transport | | Media Transport | +-----------------+ +-----------------+ Figure 8: Example of Media Source Layered Dependency As an example, the SVC MRST and MRMT (Section 3.5) relations needs to identify the common Media Encoder origin for the Encoded and Dependent Streams. The SVC RTP Payload RFC [RFC6190] is not particularly explicit about how this relation is to be implemented. When using different RTP Sessions, thus different Media Transports (MRMT (Section 3.5)), and as long as there is only one RTP Stream per Media Encoder and a single Media Source in each RTP Session (MRMT), common SSRC and CNAMEs can be used to identify the common Media Source. When multiple RTP Streams are sent from one Media Encoder in the same RTP Session (MRST), then CNAME is the only currently specified RTP identifier that can be used. In cases where multiple Media Encoders use multiple Media Sources sharing Synchronization Context, and thus having a common CNAME, additional heuristics or identification need to be applied to create the MRST or MRMT relationships between the RTP Streams. Lennox, et al. Expires July 26, 2015 [Page 26] Internet-Draft RTP Grouping Taxonomy January 2015 3.9. RTP Stream Duplication RTP Stream Duplication [RFC7198], using the same or different Media Transports, and optionally also delaying the duplicate [RFC7197], offers a simple way to protect media flows from packet loss in some cases (see Figure 9). It is a specific type of redundancy and all but one Source RTP Stream (Section 2.1.10) are effectively Redundancy RTP Streams (Section 2.1.12), but since both Source and Redundant RTP Streams are the same it does not matter which one is which. This can also be seen as a specific type of Simulcast (Section 3.7) that transmits the same Encoded Stream (Section 2.1.7) multiple times. +----------------+ | Media Source | +----------------+ Source Stream | V +----------------+ | Media Encoder | +----------------+ Encoded Stream | +-----------+-----------+ | | V V +------------------+ +------------------+ | Media Packetizer | | Media Packetizer | +------------------+ +------------------+ Source | RTP Stream Source | RTP Stream | V | +-------------+ | | Delay (opt) | | +-------------+ | | +-----------+-----------+ | V +-------------------+ | Media Transport | +-------------------+ Figure 9: Example of RTP Stream Duplication 3.10. Redundancy Format The RTP Payload for Redundant Audio Data [RFC2198] defines a transport for redundant audio data together with primary data in the same RTP payload. The redundant data can be a time delayed version of the primary or another time delayed Encoded Stream using a Lennox, et al. Expires July 26, 2015 [Page 27] Internet-Draft RTP Grouping Taxonomy January 2015 different Media Encoder to encode the same Media Source as the primary, as depicted in Figure 10. +--------------------+ | Media Source | +--------------------+ | Source Stream | +------------------------+ | | V V +--------------------+ +--------------------+ | Media Encoder | | Media Encoder | +--------------------+ +--------------------+ | | | +------------+ Encoded Stream | Time Delay | | +------------+ | | | +------------------+ V V +--------------------+ | Media Packetizer | +--------------------+ | V RTP Stream Figure 10: Concept for usage of Audio Redundancy with different Media Encoders The Redundancy format is thus providing the necessary meta information to correctly relate different parts of the same Encoded Stream, or in the case depicted above (Figure 10) relate the Received Source Stream fragments coming out of different Media Decoders to be able to combine them together into a less erroneous Source Stream. 3.11. RTP Retransmission Figure 11 shows an example where a Media Source's Source RTP Stream is protected by a retransmission (RTX) flow [RFC4588]. In this example the Source RTP Stream and the Redundancy RTP Stream share the same Media Transport. Lennox, et al. Expires July 26, 2015 [Page 28] Internet-Draft RTP Grouping Taxonomy January 2015 +--------------------+ | Media Source | +--------------------+ | V +--------------------+ | Media Encoder | +--------------------+ | Retransmission Encoded Stream +--------+ +---- Request V | V V +--------------------+ | +--------------------+ | Media Packetizer | | | RTP Retransmission | +--------------------+ | +--------------------+ | | | +------------+ Redundancy RTP Stream Source RTP Stream | | | +---------+ +---------+ | | V V +-----------------+ | Media Transport | +-----------------+ Figure 11: Example of Media Source Retransmission Flows The RTP Retransmission example (Figure 11) illustrates that this mechanism works purely on the Source RTP Stream. The RTP Retransmission transform buffers the sent Source RTP Stream and, upon request, emits a retransmitted packet with an extra payload header as a Redundancy RTP Stream. The RTP Retransmission mechanism [RFC4588] is specified such that there is a one to one relation between the Source RTP Stream and the Redundancy RTP Stream. Therefore, a Redundancy RTP Stream needs to be associated with its Source RTP Stream. This is done based on CNAME selectors and heuristics to match requested packets for a given Source RTP Stream with the original sequence number in the payload of any new Redundancy RTP Stream using the RTX payload format. In cases where the Redundancy RTP Stream is sent in a separate RTP Session from the Source RTP Stream, these sessions are related, which is signaled by using the SDP Media Grouping's [RFC5888] Flow Identification (FID identification-tag) semantics. Lennox, et al. Expires July 26, 2015 [Page 29] Internet-Draft RTP Grouping Taxonomy January 2015 3.12. Forward Error Correction Figure 12 shows an example where two Media Sources' Source RTP Streams are protected by Forward Error Correction (FEC). Source RTP Stream A has a RTP-based Redundancy transformation in FEC Encoder 1. This produces a Redundancy RTP Stream 1, that is only related to Source RTP Stream A. The FEC Encoder 2, however, takes two Source RTP Streams (A and B) and produces a Redundancy RTP Stream 2 that protects them jointly, i.e. Redundancy RTP Stream 2 relates to two Source RTP Streams (a FEC group). FEC decoding, when needed due to packet loss or packet corruption at the receiver, requires knowledge about which Source RTP Streams that the FEC encoding was based on. In Figure 12 all RTP Streams are sent on the same Media Transport. This is however not the only possible choice. Numerous combinations exist for spreading these RTP Streams over different Media Transports to achieve the communication application's goal. +--------------------+ +--------------------+ | Media Source A | | Media Source B | +--------------------+ +--------------------+ | | V V +--------------------+ +--------------------+ | Media Encoder A | | Media Encoder B | +--------------------+ +--------------------+ | | Encoded Stream Encoded Stream V V +--------------------+ +--------------------+ | Media Packetizer A | | Media Packetizer B | +--------------------+ +--------------------+ | | Source RTP Stream A Source RTP Stream B | | +-----+---------+-------------+ +---+---+ | V V V | | +---------------+ +---------------+ | | | FEC Encoder 1 | | FEC Encoder 2 | | | +---------------+ +---------------+ | | Redundancy | Redundancy | | | RTP Stream 1 | RTP Stream 2 | | V V V V +----------------------------------------------------------+ | Media Transport | +----------------------------------------------------------+ Figure 12: Example of FEC Redundancy RTP Streams Lennox, et al. Expires July 26, 2015 [Page 30] Internet-Draft RTP Grouping Taxonomy January 2015 As FEC Encoding exists in various forms, the methods for relating FEC Redundancy RTP Streams with its source information in Source RTP Streams are many. The XOR based RTP FEC Payload format [RFC5109] is defined in such a way that a Redundancy RTP Stream has a one to one relation with a Source RTP Stream. In fact, the RFC requires the Redundancy RTP Stream to use the same SSRC as the Source RTP Stream. This requires to either use a separate RTP Session or to use the Redundancy RTP Payload format [RFC2198]. The underlying relation requirement for this FEC format and a particular Redundancy RTP Stream is to know the related Source RTP Stream, including its SSRC. 3.13. RTP Stream Separation RTP Streams can be separated exclusively based on their SSRCs, at the RTP Session level, or at the Multi-Media Session level. When the RTP Streams that have a relationship are all sent in the same RTP Session and are uniquely identified based on their SSRC only, it is termed an SSRC-Only Based Separation. Such streams can be related via RTCP CNAME to identify that the streams belong to the same Endpoint. SSRC-based approaches [RFC5576], when used, can explicitly relate various such RTP Streams. On the other hand, when RTP Streams that are related but are sent in the context of different RTP Sessions to achieve separation, it is known as RTP Session-based separation. This is commonly used when the different RTP Streams are intended for different Media Transports. Several mechanisms that use RTP Session-based separation rely on it to enable an implicit grouping mechanism expressing the relationship. The solutions have been based on using the same SSRC value in the different RTP Sessions to implicitly indicate their relation. That way, no explicit RTP level mechanism has been needed, only signaling level relations have been established using semantics from Grouping of Media lines framework [RFC5888]. Examples of this are RTP Retransmission [RFC4588], SVC Multi-Session Transmission [RFC6190] and XOR Based FEC [RFC5109]. RTCP CNAME explicitly relates RTP Streams across different RTP Sessions, as explained in the previous section. Such a relationship can be used to perform inter-media synchronization. RTP Streams that are related and need to be associated can be part of different Multimedia Sessions, rather than just different RTP Sessions within the same Multimedia Session context. This puts further demand on the scope of the mechanism(s) and its handling of identifiers used for expressing the relationships. Lennox, et al. Expires July 26, 2015 [Page 31] Internet-Draft RTP Grouping Taxonomy January 2015 3.14. Multiple RTP Sessions over one Media Transport [I-D.westerlund-avtcore-transport-multiplexing] describes a mechanism that allows several RTP Sessions to be carried over a single underlying Media Transport. The main reasons for doing this are related to the impact of using one or more Media Transports (using a common network path or potentially have different ones). The fewer Media Transports used, the less need for NAT/FW traversal resources and number of flow based Quality of Service (QoS). However, Multiple RTP Sessions over one Media Transport imply that a single Media Transport 5-tuple is not sufficient to express in which RTP Session context a particular RTP Stream exists. Complexities in the relationship between Media Transports and RTP Session already exist as one RTP Session contains multiple Media Transports, e.g. even a Peer-to-Peer RTP Session with RTP/RTCP Multiplexing requires two Media Transports, one in each direction. The relationship between Media Transports and RTP Sessions as well as additional levels of identifiers need to be considered in both signaling design and when defining terminology. 4. Mapping from Existing Terms This section describes a selected set of terms from some relevant IETF RFC and Internet Drafts (at the time of writing), using the concepts from previous sections. 4.1. Telepresence Terms The terms in this sub-section are used in the context of CLUE [I-D.ietf-clue-framework]. 4.1.1. Audio Capture Describes an audio Media Source (Section 2.1.4). 4.1.2. Capture Device Identifies a physical entity performing a Media Capture (Section 2.1.2) transformation. 4.1.3. Capture Encoding Describes an Encoded Stream (Section 2.1.7) related to CLUE specific semantic information. Lennox, et al. Expires July 26, 2015 [Page 32] Internet-Draft RTP Grouping Taxonomy January 2015 4.1.4. Capture Scene Describes a set of spatially related Media Sources (Section 2.1.4). 4.1.5. Endpoint Describes exactly one Participant (Section 2.2.3) and one or more Endpoints (Section 2.2.1). 4.1.6. Individual Encoding Describes the configuration information needed to perform a Media Encoder (Section 2.1.6) transformation. 4.1.7. Media Capture Describes either a Media Capture (Section 2.1.2) or a Media Source (Section 2.1.4), depending on in which context the term is used. 4.1.8. Media Consumer Describes the media receiving part of an Endpoint (Section 2.2.1). 4.1.9. Media Provider Describes the media sending part of an Endpoint (Section 2.2.1). 4.1.10. Stream Describes an RTP Stream (Section 2.1.10). 4.1.11. Video Capture Describes a video Media Source (Section 2.1.4). 4.2. Media Description A single Session Description Protocol (SDP) [RFC4566] media description (or media block; an m-line and all subsequent lines until the next m-line or the end of the SDP) describes part of the necessary configuration and identification information needed for a Media Encoder transformation, as well as the necessary configuration and identification information for the Media Decoder to be able to correctly interpret a received RTP Stream. A Media Description typically relates to a single Media Source. This is for example an explicit restriction in WebRTC. However, nothing prevents that the same Media Description (and same RTP Session) is Lennox, et al. Expires July 26, 2015 [Page 33] Internet-Draft RTP Grouping Taxonomy January 2015 re-used for multiple Media Sources [I-D.ietf-avtcore-rtp-multi-stream]. It can thus describe properties of one or more RTP Streams, and can also describe properties valid for an entire RTP Session (via [RFC5576] mechanisms, for example). 4.3. Media Stream RTP [RFC3550] uses media stream, audio stream, video stream, and stream of (RTP) packets interchangeably, which are all RTP Streams. 4.4. Multimedia Conference A Multimedia Conference is a Communication Session (Section 2.2.5) between two or more Participants (Section 2.2.3), along with the software they are using to communicate. 4.5. Multimedia Session SDP [RFC4566] defines a Multimedia Session as a set of multimedia senders and receivers and the data streams flowing from senders to receivers, which would correspond to a set of Endpoints and the RTP Streams that flow between them. In this memo, Multimedia Session (Section 2.2.4) also assumes those Endpoints belong to a set of Participants that are engaged in communication via a set of related RTP Streams. RTP [RFC3550] defines a Multimedia Session as a set of concurrent RTP Sessions among a common group of Participants. For example, a video conference may contain an audio RTP Session and a video RTP Session. This would correspond to a group of Participants (each using one or more Endpoints) sharing a set of concurrent RTP Sessions. In this memo, Multimedia Session also defines those RTP Sessions to have some relation and be part of a communication among the Participants. 4.6. Multipoint Control Unit (MCU) This term is commonly used to describe the central node in any type of star topology [I-D.ietf-avtcore-rtp-topologies-update] conference. It describes a device that includes one Participant (Section 2.2.3) (usually corresponding to a so-called conference focus) and one or more related Endpoints (Section 2.2.1) (sometimes one or more per conference Participant). 4.7. Recording Device WebRTC specifications use this term to refer to locally available entities performing a Media Capture (Section 2.1.2) transformation. Lennox, et al. Expires July 26, 2015 [Page 34] Internet-Draft RTP Grouping Taxonomy January 2015 4.8. RtcMediaStream A WebRTC RtcMediaStreamTrack is a set of Media Sources (Section 2.1.4) sharing the same Synchronization Context (Section 3.1). 4.9. RtcMediaStreamTrack A WebRTC RtcMediaStreamTrack is a Media Source (Section 2.1.4). 4.10. RTP Sender RTP [RFC3550] uses this term, which can be seen as the RTP protocol part of a Media Packetizer (Section 2.1.9). 4.11. RTP Session Within the context of SDP, a singe m= line can map to a single RTP Session (Section 2.2.2) or multiple m= lines can map to a single RTP Session. The latter is enabled via multiplexing schemes such as BUNDLE [I-D.ietf-mmusic-sdp-bundle-negotiation], for example, which allows mapping of multiple m= lines to a single RTP Session. 4.12. SSRC RTP [RFC3550] defines this as "the source of a stream of RTP packets", which indicates that an SSRC is not only a unique identifier for the Encoded Stream (Section 2.1.7) carried in those packets, but is also effectively used as a term to denote a Media Packetizer (Section 2.1.9). 5. Security Considerations This document simply tries to clarify the confusion prevalent in RTP taxonomy because of inconsistent usage by multiple technologies and protocols making use of the RTP protocol. It does not introduce any new security considerations beyond those already well documented in the RTP protocol [RFC3550] and each of the many respective specifications of the various protocols making use of it. Hopefully having a well-defined common terminology and understanding of the complexities of the RTP architecture will help lead us to better standards, avoiding security problems. Lennox, et al. Expires July 26, 2015 [Page 35] Internet-Draft RTP Grouping Taxonomy January 2015 6. Acknowledgement This document has many concepts borrowed from several documents such as WebRTC [I-D.ietf-rtcweb-overview], CLUE [I-D.ietf-clue-framework], Multiplexing Architecture [I-D.westerlund-avtcore-transport-multiplexing]. The authors would like to thank all the authors of each of those documents. The authors would also like to acknowledge the insights, guidance and contributions of Magnus Westerlund, Roni Even, Paul Kyzivat, Colin Perkins, Keith Drage, Harald Alvestrand, Alex Eleftheriadis, Mo Zanaty, and Stephan Wenger. 7. Contributors Magnus Westerlund has contributed the concept model for the media chain using transformations and streams model, including rewriting pre-existing concepts into this model and adding missing concepts. The first proposal for updating the relationships and the topologies based on this concept was also performed by Magnus. 8. IANA Considerations This document makes no request of IANA. 9. Informative References [I-D.ietf-avtcore-rtp-multi-stream] Lennox, J., Westerlund, M., Wu, W., and C. Perkins, "Sending Multiple Media Streams in a Single RTP Session", draft-ietf-avtcore-rtp-multi-stream-06 (work in progress), October 2014. [I-D.ietf-avtcore-rtp-topologies-update] Westerlund, M. and S. Wenger, "RTP Topologies", draft- ietf-avtcore-rtp-topologies-update-05 (work in progress), November 2014. [I-D.ietf-clue-framework] Duckworth, M., Pepperell, A., and S. Wenger, "Framework for Telepresence Multi-Streams", draft-ietf-clue- framework-20 (work in progress), January 2015. [I-D.ietf-mmusic-sdp-bundle-negotiation] Holmberg, C., Alvestrand, H., and C. Jennings, "Negotiating Media Multiplexing Using the Session Description Protocol (SDP)", draft-ietf-mmusic-sdp-bundle- negotiation-15 (work in progress), January 2015. Lennox, et al. Expires July 26, 2015 [Page 36] Internet-Draft RTP Grouping Taxonomy January 2015 [I-D.ietf-rtcweb-overview] Alvestrand, H., "Overview: Real Time Protocols for Browser-based Applications", draft-ietf-rtcweb-overview-13 (work in progress), November 2014. [I-D.westerlund-avtcore-transport-multiplexing] Westerlund, M. and C. Perkins, "Multiplexing Multiple RTP Sessions onto a Single Lower-Layer Transport", draft- westerlund-avtcore-transport-multiplexing-07 (work in progress), October 2013. [RFC2198] Perkins, C., Kouvelas, I., Hodson, O., Hardman, V., Handley, M., Bolot, J., Vega-Garcia, A., and S. Fosse- Parisis, "RTP Payload for Redundant Audio Data", RFC 2198, September 1997. [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson, "RTP: A Transport Protocol for Real-Time Applications", STD 64, RFC 3550, July 2003. [RFC3551] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and Video Conferences with Minimal Control", STD 65, RFC 3551, July 2003. [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session Description Protocol", RFC 4566, July 2006. [RFC4588] Rey, J., Leon, D., Miyazaki, A., Varsa, V., and R. Hakenberg, "RTP Retransmission Payload Format", RFC 4588, July 2006. [RFC4867] Sjoberg, J., Westerlund, M., Lakaniemi, A., and Q. Xie, "RTP Payload Format and File Storage Format for the Adaptive Multi-Rate (AMR) and Adaptive Multi-Rate Wideband (AMR-WB) Audio Codecs", RFC 4867, April 2007. [RFC5109] Li, A., "RTP Payload Format for Generic Forward Error Correction", RFC 5109, December 2007. [RFC5404] Westerlund, M. and I. Johansson, "RTP Payload Format for G.719", RFC 5404, January 2009. [RFC5576] Lennox, J., Ott, J., and T. Schierl, "Source-Specific Media Attributes in the Session Description Protocol (SDP)", RFC 5576, June 2009. [RFC5888] Camarillo, G. and H. Schulzrinne, "The Session Description Protocol (SDP) Grouping Framework", RFC 5888, June 2010. Lennox, et al. Expires July 26, 2015 [Page 37] Internet-Draft RTP Grouping Taxonomy January 2015 [RFC5905] Mills, D., Martin, J., Burbank, J., and W. Kasch, "Network Time Protocol Version 4: Protocol and Algorithms Specification", RFC 5905, June 2010. [RFC6190] Wenger, S., Wang, Y., Schierl, T., and A. Eleftheriadis, "RTP Payload Format for Scalable Video Coding", RFC 6190, May 2011. [RFC7160] Petit-Huguenin, M. and G. Zorn, "Support for Multiple Clock Rates in an RTP Session", RFC 7160, April 2014. [RFC7197] Begen, A., Cai, Y., and H. Ou, "Duplication Delay Attribute in the Session Description Protocol", RFC 7197, April 2014. [RFC7198] Begen, A. and C. Perkins, "Duplicating RTP Streams", RFC 7198, April 2014. [RFC7273] Williams, A., Gross, K., van Brandenburg, R., and H. Stokking, "RTP Clock Source Signalling", RFC 7273, June 2014. Appendix A. Changes From Earlier Versions NOTE TO RFC EDITOR: Please remove this section prior to publication. A.1. Modifications Between WG Version -04 and -05 o Editorial improvements A.2. Modifications Between WG Version -03 and -04 o Changed "Media Redundancy" and "Media Repair" to "RTP-based Redundancy" and "RTP-based Repair", since those terms are more specific and correct. o Changed "End Point" to "Endpoint" and removed Editor's Note on this. o Clarified that a Media Capture may impose constraints on clock handling. o Clarified that mixing multiple Raw Streams into a Source Stream is not possible, since that requires mixed streams to have a timing relation, requiring them to be Source Streams, and added an example. Lennox, et al. Expires July 26, 2015 [Page 38] Internet-Draft RTP Grouping Taxonomy January 2015 o Clarified that RTP-based Redundancy excludes the type of encoding redundancy found within the encoded media format in an Encoded Stream. o Clarified that a Media Transport contains only a single RTP Session, but a single RTP Session can span multiple Media Transports. o Clarified that packets with seemingly correct checksum that are received by a Media Transport Receiver may still be corrupt. o Clarified that a corrupt packet in a Media Transport Receiver is typically either discarded or somehow marked and passed on in the Received RTP Stream. o Added Synchronization Context to Figure 6. o Editorial improvements and clarifications. A.3. Modifications Between WG Version -02 and -03 o Changed section 3.5, removing SST-SS/MS and MST-SS/MS, replacing them with SRST, MRST, and MRMT. o Updated section 3.8 to align with terminology changes in section 3.5. o Added a new section 4.12, describing the term Multimedia Conference. o Changed reference from I-D to now published RFC 7273. o Editorial improvements and clarifications. A.4. Modifications Between WG Version -01 and -02 o Major re-structure o Moved media chain Media Transport detailing up one section level o Collapsed level 2 sub-sections of section 3 and thus moved level 3 sub-sections up one level, gathering some introductory text into the beginning of section 3 o Added that not only SSRC collision, but also a clock rate change [RFC7160] is a valid reason to change SSRC value for an RTP stream o Added a sub-section on clock source signaling Lennox, et al. Expires July 26, 2015 [Page 39] Internet-Draft RTP Grouping Taxonomy January 2015 o Added a sub-section on RTP stream duplication o Elaborated a bit in section 2.2.1 on the relation between End Points, Participants and CNAMEs o Elaborated a bit in section 2.2.4 on Multimedia Session and synchronization contexts o Removed the section on CLUE scenes defining an implicit synchronization context, since it was incorrect o Clarified text on SVC SST and MST according to list discussions o Removed the entire topology section to avoid possible inconsistencies or duplications with draft-ietf-avtcore-rtp- topologies-update, but saved one example overview figure of Communication Entities into that section o Added a section 4 on mapping from existing terms with one sub- section per term, mainly by moving text from sections 2 and 3 o Changed all occurrences of Packet Stream to RTP Stream o Moved all normative references to informative, since this is an informative document o Added references to RFC 7160, RFC 7197 and RFC 7198, and removed unused references A.5. Modifications Between WG Version -00 and -01 o WG version -00 text is identical to individual draft -03 o Amended description of SVC SST and MST encodings with respect to concepts defined in this text o Removed UML as normative reference, since the text no longer uses any UML notation o Removed a number of level 4 sections and moved out text to the level above A.6. Modifications Between Version -02 and -03 o Section 4 rewritten (and new communication topologies added) to reflect the major updates to Sections 1-3 o Section 8 removed (carryover from initial -00 draft) Lennox, et al. Expires July 26, 2015 [Page 40] Internet-Draft RTP Grouping Taxonomy January 2015 o General clean up of text, grammar and nits A.7. Modifications Between Version -01 and -02 o Section 2 rewritten to add both streams and transformations in the media chain. o Section 3 rewritten to focus on exposing relationships. A.8. Modifications Between Version -00 and -01 o Too many to list o Added new authors o Updated content organization and presentation Authors' Addresses Jonathan Lennox Vidyo, Inc. 433 Hackensack Avenue Seventh Floor Hackensack, NJ 07601 US Email: jonathan@vidyo.com Kevin Gross AVA Networks, LLC Boulder, CO US Email: kevin.gross@avanw.com Suhas Nandakumar Cisco Systems 170 West Tasman Drive San Jose, CA 95134 US Email: snandaku@cisco.com Lennox, et al. Expires July 26, 2015 [Page 41] Internet-Draft RTP Grouping Taxonomy January 2015 Gonzalo Salgueiro Cisco Systems 7200-12 Kit Creek Road Research Triangle Park, NC 27709 US Email: gsalguei@cisco.com Bo Burman Ericsson Kistavagen 25 SE-164 80 Stockholm Sweden Phone: +46 10 714 13 11 Email: bo.burman@ericsson.com Lennox, et al. Expires July 26, 2015 [Page 42]