Internet Draft J. Rey draft-ietf-avt-rtp-3gpp-timed-text-06.txt Y. Matsui Matsushita Expires: March 10, 2005 September 10, 2004 RTP Payload Format for 3GPP Timed Text Status of this Memo By submitting this Internet-Draft, we certify that any applicable patent or other IPR claims of which we are aware have been disclosed, and any of which we become aware will be disclosed, in accordance with RFC 3668 (BCP 79). By submitting this Internet-Draft, we accept the provisions of Section 3 of RFC 3667 (BCP 78). Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. 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." The list of current Internet-Drafts can be accessed at http://www.ietf.org/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html Abstract This document specifies an RTP payload format for the transmission of 3GPP (3rd Generation Partnership Project) timed text. 3GPP timed text is a time-lined decorated text media format with defined storage in a 3GP file. Timed Text can be synchronized with audio/video contents. As of today, 3GP files containing timed text contents can only be downloaded via HTTP. There is no available mechanism for streaming 3GPP timed text contents neither out of 3GP files nor directly from live content. In the following sections the problems of streaming timed text are addressed and a payload format for streaming 3GPP timed text over RTP is specified. IETF draft - Expires March 10, 2005 [Page 1] Table of Contents 1. Terminology.....................................................3 2. Introduction....................................................5 2.1. General Overview of the 3GPP Timed Text format...............5 2.2. Requirements for a payload format for 3GPP timed text........7 2.3. General Remarks..............................................8 2.3.1. Character Counting........................................8 2.3.2. On the length indication in the units.....................8 2.3.3. Fragmentation of Timed Text Samples.......................8 2.3.4. On aggregate payloads....................................10 2.3.5. Reassembling text samples at the receiver................12 2.3.6. Live streaming vs. Streaming from a 3GP file.............14 3. RTP Payload Format for 3GPP Timed Text.........................14 3.1. Payload Header Definitions..................................17 3.1.1. Common Payload Header Fields.............................18 3.1.2. TYPE 1 Header............................................20 3.1.3. TYPE 2 Header............................................22 3.1.4. TYPE 3 Header............................................23 3.1.5. TYPE 4 Header............................................24 3.1.6. TYPE 5 Header............................................25 4. Resilient Transport............................................25 5. Congestion control.............................................26 6. Scene Description..............................................26 6.1. Text rendering position and composition.....................26 6.2. SMIL usage..................................................27 7. MIME Type usage Registration...................................27 7.1. 3GPP Timed Text MIME Registration...........................27 8. SDP usage......................................................30 8.1. Mapping to SDP..............................................30 8.2. Parameter Usage in the SDP Offer/Answer Model...............31 8.2.1. Unicast Usage............................................31 8.2.2. Multicast Usage..........................................33 8.3. Offer/Answer Examples.......................................34 8.4. Parameter Usage outside of Offer/Answer.....................36 9. IANA Considerations............................................36 10. Security considerations.......................................36 11. References....................................................37 11.1. Normative References.......................................37 11.2. Informative References.....................................37 12. Annexes.......................................................38 12.1. Dynamic SIDX wrap-around mechanism.........................38 12.2. Basics of the 3GP File Structure...........................39 12.3. Usage of 3GP file information for transport in RTP.........40 13. Acknowledgements..............................................42 14. Author's Addresses............................................42 15. IPR Notices...................................................42 IETF draft - Expires March 10, 2005 [Page 2] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 16. Full Copyright Statement......................................43 [Note to the RFC Editor: - please delete the Change Log section upon publication of this document as RFC, - please replace "RFCXXXX" with the RFC designation of this document when published, - please substitute "draft-ietf-..." references with the corresponding RFC number if available at the time of publication] Change Log Since version -05: - sendonly example in section 8: corrected offer. - formatting issues with the O/A examples: wraparound of long lines - MIME registration: - added cross-reference to section 3 for choosing other than default timestamp rate values. - changed MIME type registration from 'video' to 'text' as agreed on AVT ML. - MIME registration template includes "Restrictions on usage:" field, new in draft-freed-media-type-reg-01.txt. - removed the "File extension" from MIME registration template since the payload format does not define a file storing format. - editorial nits. 1. Terminology The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [5]. Furthermore, the following terms are used and have specific meaning within the context of this document: text sample or whole text sample or text access unit: In the 3GPP Timed Text media format [1] this refers to a unit of timed text data as contained in the source file. Its equivalent in audio/video would be a frame. A text sample contains text strings followed by zero or more modifier boxes. In MPEG the equivalent of a text sample is referred as a 'text access unit'. Rey & Matsui [Page 3] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 In this document, like in 3GPP Timed Text media format [1], the a text sample comprises a text string (note definition below) followed by zero or more modifier boxes. fragment or text sample fragment: a fraction of a text sample. A fragment may contain either text strings or modifier (decoration) contents, but not both at the same time. sample contents: general term to identify timed text data transported when using this payload format. text strings: text strings is the term used to denote the actual text characters encoded either as UTF-8 [18] or UTF-16 [19]. This text string MUST NOT contain any byte order mark (BOM). This differs from the definition in the 3GPP Timed Text media format [1], where UTF-16 text strings must include it. In this payload format the BOM is not needed as explained in Section 3.1.1. decoration/modifiers: the terms "decoration" and "modifiers" are used interchangeably throughout the document to denote the contents of the text sample that modify the default text formatting. Modifiers may, for example, specify different font size for a particular sequence of characters or define karaoke timing for the sample. sample description: this term is used to denote information which is potentially shared by more than one text sample. In a 3GP file a sample description is stored in a place where it can be shared. It contains setup and default information such as scrolling direction, text box position, delay value, default font, background color, etc. units or transport units: the payload headers specified in this document encapsulate text samples, fragments thereof and sample descriptions by prepending a specific payload header and so building what is here called a (transport) unit. Rey & Matsui [Page 4] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 aggregation / aggregate packet An aggregate RTP packet consists of several units. track / stream 3GP files contain audio/video and text tracks. This document enables to stream text tracks using RTP. Therefore both terms are exchanged in this document in the context of 3GP files. Media Header Box / Track Header Box / ... the 3GP file format makes use of these structures defined in the ISO Base File Format [2]. When referring to these in this document, initials are capitalized for clarity. 2. Introduction 3GPP timed text is a media format for time-lined decorated text specified in [1]. 3GPP Timed text contents may be stored in 3GP files or may be generated in real time. The 3GP file format itself is based on the ISO Base Media File Format recommendation [2]. Section 12.2 gives some insight into the 3GP file structure. The purpose of this draft is to provide a means to stream 3GPP timed text contents using RTP. This includes the streaming of timed text being read out of a 3GP file as well as the streaming of timed text generated in real time, a.k.a. live streaming. 2.1. General Overview of the 3GPP Timed Text format The 3GPP timed text format was developed for use in the services specified in the 3GPP Transparent End-to-end Packet-switched Streaming Services (3GPP PSS) [16]. Besides plain text, the 3GPP timed text format allows the display of decorated text: like for karaoke applications, scrolling text for newscasts or hyperlinked text. Furthermore, these contents may or may not be synchronized with other media, like audio or video. The scope of the 3GPP PSS includes both downloading and streaming of multimedia content over 3G packet-switched networks. However, due to the lack of an appropriate RTP payload format, the current usage of the 3GPP timed text file format is limited to downloading via HTTP. The 3GPP PSS adopts multimedia codecs (such as MPEG-4 Visual, AMR, MPEG-4 AAC, and JPEG) and protocols like SMIL [9] for presentation layouts or RTP [3] for streaming. In general, a multimedia presentation might consist of several audio/video/text streams (or tracks in ISO file format jargon). Different streams may have Rey & Matsui [Page 5] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 different formats. The media may be spatially synchronized either using the information within the streams or a scene description language like SMIL. An example of this would be a media session with three different media streams: 1 audio, 1 video and 1 timed text that reproduces a music video with karaoke subtitles. For each stream some information is needed, which defines the regions where each media is displayed, how the media looks like and its synchronization, among other things. In karaoke, for example, the song lyrics may be displayed below the music video and the words are highlighted synchronized with the music track. Four differentiated functional components might be identified in the 3GPP timed text media format: o initial spatial layout information related to the text track: these are the height and width of the text region where text is displayed, the position of the text region in the display and the layer or proximity of the text to the user. These pieces of information are contained in the Track Header Box. Sections 6.1 and 12 provide further details. o default settings for formatting and positioning of text: style (font, size, colour,...), background colour, horizontal and vertical justification, line width, scrolling, etcetera. Sample descriptions contain such settings. o the actual text: encoded characters using either UTF-8 [18] or UTF-16 [19] encoding and, o the decoration inside the modifier boxes: if some characters have different style, some delay, blink, etcetera... this needs to be indicated by appending the modifier boxes to the text strings. Modifier boxes are only present in the text samples if they are actually needed. Otherwise, the default settings in the corresponding sample description apply. At the time of writing this payload format the following decorations or modifiers are specified in the 3GPP timed text media format specification [1]: - text highlight, - highlight color, - blinking text, - karaoke feature, - hyperlink, - text delay, - text style and, - positioning of the text box and, - text wrap indication. Rey & Matsui [Page 6] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 Section 12.3 specifies where to find these values in the 3GP file and how these are mapped to the payload format. For live streaming, appropriate values using the same formats and units shall be used. For further details on the 3GPP Timed Text media format, refer to [1]. 2.2. Requirements for a payload format for 3GPP timed text In this section a set of requirements is listed. A justification for each of them is also given. An RTP Payload Format for 3GPP timed text SHALL: 1. Keep the 3GP text sample structure. This requirement means that it SHALL be possible for an RTP receiver using this payload format to rebuild the text samples from the received RTP packets. 2. Transmit the text sample size, sample duration and sample description index in-band. In RTP it is important to transmit it in- band because this information might change from sample to sample. 3. Enable the transmission of the sample descriptions both by out-of-band and in-band means. Typically, a set of default formatting settings is transmitted out-of-band (reliably) once at the initialization phase. If sample descriptions are needed in the course of a session, these may be sent also out-of-band or in-band. In-band transmission, although unreliable, may be more appropriate for large sample descriptions or if these are sent frequently. It is also useful in cases where an out-of-band channel may not be available and for live streaming, where contents are not known a priori. In order to cover this wide range of scenarios, the payload format SHALL enable both in-band and out-of-band transmission of sample descriptions. 4. Enable the aggregation of units into an RTP packet. In a mobile communication environment a typical text sample size is around 100-200 bytes. If the available bit rate and the packet size allow it, units SHOULD be aggregated into one RTP packet. This makes the transport more efficient. 5. Enable the fragmentation of a text sample into several RTP packets in order to cover a wide range of applications and network environments. In general, fragmentation should be a rare event given the low bit rates and text sample sizes. However, the 3GPP Timed Text media format does allow for larger text samples. The payload format SHALL take this into account and provide a means for coping with fragmentation and reassembly. 6. Enable the use of resilient transport mechanisms, such as repetition, retransmission [11] and FEC [7]. These mechanisms may be used to protect the information. RFC 2354 [8] discusses available mechanisms for stream repair. Rey & Matsui [Page 7] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 2.3. General Remarks Before going into the details of the payload headers, some general observations are made in this section. In order to understand the next sections a rough functional description of the units is needed: o a TYPE 1 unit contains one complete text sample, o a TYPE 2 unit contains a complete text string or a fragment thereof, o a TYPE 3 unit contains the complete modifiers or only the first fragment thereof, o a TYPE 4 unit contains one modifier fragment other than the first and, o a TYPE 5 unit contains one sample description. 2.3.1. Character Counting This payload format does not enable an RTP receiver to find out the exact number of text characters lost. The fragment size included in the payload headers does not help in finding the number of lost characters, because the UTF-8/UTF-16 [18][19] encodings used yield a variable number of bytes per character. Informative note: for finding out the exact number of lost characters, an additional field in TYPE 2 units and character counting upon fragmentation would be required. This overhead does not seem reasonable given the expected low rate of fragmentation. 2.3.2. On the length indication in the units Usually, RTP applications use the information on packet size from UDP or lower layers to find out the length of the RTP payload. While this information can still be used, this payload format includes an explicit length indication for each unit in the payload as a fixed field in the payload headers. This design choice allows easy interoperability with the RTP Payload Format for Transport of MPEG-4 Elementary Streams, RFC 3640 [12], which does require an explicit length indication for each unit (see AU-header in RFC 3640). 2.3.3. Fragmentation of Timed Text Samples This section describes why text samples may have to be fragmented and discusses some of the possible approaches to do it. A solution is proposed together with rules and recommendations for fragmenting and transporting text samples using this payload format. Rey & Matsui [Page 8] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 3GPP Timed Text applications are expected to operate at low bit rates. This fact added to the small size of timed text samples (typically one or two hundred bytes) makes fragmentation of text samples a rare event. Samples should usually fit into the MTU size of the used network path. Nevertheless, some text strings (e.g. ending roll in a movie) and some modifier boxes (i.e. for hyperlinks, for karaoke or for styles) might become large. This may also apply for future modifier boxes. In such cases, the first option to consider is whether it is possible to adjust the encoding (e.g. the size of sample) in such a way that fragmentation is avoided. If so, this is preferred to fragmentation and SHOULD be done. Otherwise, if this is not possible or other constraints avoid doing this, fragmentation MAY be used and the basic guidelines given in this document MUST be followed: o It is RECOMMENDED that text samples are fragmented as seldom as possible, i.e. the least possible number of fragments is created out of a text sample. o If there is some bitrate and space in the payload available, sample descriptions (if at hand) SHOULD be aggregated. Sample descriptions (TYPE 5 units) MAY be placed anywhere in an aggregate payload, since the sample index (SIDX) is used to associate them to their text samples (explained further in this document). o Text strings MUST split at character boundaries. Otherwise, it is not possible to display the text contents of a fragment if a previous fragment was lost. As a consequence, text string fragmentation requires knowledge of the UTF-8/UTF-16 encoding formats to determine character boundaries. o Unlike text strings, the modifier boxes are NOT REQUIRED to split at meaningful boundaries. However, it is RECOMMENDED to do so whenever possible. This decreases the effects of packet loss. This payload format does not ensure that partially received modifiers can be applied to text strings. If only part of the modifiers is received, it is an application issue how to deal with these, i.e. whether to use them or not. Informative note: ensuring that partially received modifiers can be applied to text strings in all cases (for all modifier types and for all fragment loss constellations) would place additional requirements on the payload format. In particular this would require that: a) senders understand the semantics of the modifier boxes and b) specific fragment headers for each of the modifier boxes are defined, in addition to the payload formats defined below. Given the low probability of fragmentation and the desire to keep the requirements low, it does not seem reasonable to include such additional headers. Rey & Matsui [Page 9] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 o Modifier and text string fragments SHOULD be protected against packet losses, i.e. using FEC [7], retransmission [11], repetition (Section 4) or an equivalent technique. This minimizes the effects of packet loss. o An additional requirement when fragmenting text samples is that the start of the modifiers MUST be indicated using the payload header defined for that purpose, i.e. a TYPE 3 unit MUST be used (see Section 3.1.4). Otherwise, if packets are lost, a client may be unable to identify where the modifiers start and the text ends or whether either text strings or modifiers were received completely or not. Note that the text string length included in the text samples can be used to easily find the modifier start byte. o Finally, sample descriptions SHALL NOT be fragmented, because they contain important information that may affect several text samples. 2.3.4. On aggregate payloads An RTP payload using this payload format MUST contain, at least, one unit of any type (TYPEs 1-5). Units SHOULD be aggregated to avoid overhead, whenever possible. The aggregate payloads MUST comply with one of the following configurations: 1. zero or more whole text samples (TYPE 1 units) and zero or more sample descriptions (TYPE 5) or, 2. zero or one modifier fragment (either TYPE 3 or TYPE 4) and zero or more sample descriptions. 3. zero or one text string fragment (TYPE 2) and zero or one TYPE 3 unit and zero or more sample descriptions. Moreover, if a TYPE 2 unit and a TYPE 3 unit are present, then they MUST belong to the same text sample. Different aggregates than the ones listed above SHALL NOT be used. Some observations: o TYPE 5 units MAY be placed anywhere in the aggregate and they SHALL NOT be regarded for calculating the timestamp of the subsequent units. This is because they usually do not belong to any text sample in particular, but may apply to several. For timestamp calculations, TYPE 5 units MUST simply be ignored, i.e. by jumping to the next unit. o as per rule 3 above, a payload MAY contain fragments of one (and only one) text sample. If this is the case, then exactly one TYPE 2 unit followed by one TYPE 3 unit are allowed in the same payload. This is inline with RFC 3640 [12], Section 2.4, which explicitly disallows combining fragments of different (text) Rey & Matsui [Page 10] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 samples in the same RTP payload. Note that, in this special case, no timestamp calculation is needed. Some examples of aggregate payloads are illustrated in Figure 1 below (although not scaled PAYLOAD 1, 2 and 3 correspond to the same length): TS1 N/A TS2 TS3 +------+-----+------+-----+ |TYPE1 |TYPE5|TYPE1 |TYPE1| +------+-----+------+-----+ sdur1 N/A sdur2 sdur3 TS4 N/A +-----+-------+ |TYPE1| TYPE 5| a) +-----+-------+ sdur4 N/A TS4 TS4 TS4 +--------------+ +--------------+ | TYPE2 | |TYPE2 |TYPE 3 | b) +--------------+ +--------------+ sdur4 sdur4 sdur4 TS4 TS4 +--------------+ +--------------+ | TYPE2| TYPE 3| | TYPE4 | c) +--------------+ +--------------+ sdur4 sdur4 sdur4 |----------PAYLOAD 1------| |--PAYLOAD 2---| |--PAYLOAD 3---| rtpts1 rtpts2 rtpts3 Figure 1. Example aggregate payloads. In Figure 1 four text samples (TS1-4) are sent using three different RTP packets. For illustrative purposes, three different possibilities for the last text sample, TS4, are depicted: a), b) and c). TS1 N/A TS2 TS3 TS4 TS4 TS4 +------+-----+------+-----+ +--------------+ +--------------+ |TYPE1 |TYPE5|TYPE1 |TYPE1| | TYPE2 | |TYPE2 |TYPE 3 | +------+-----+------+-----+ +--------------+ +--------------+ sdur1 N/A sdur2 sdur3 sdur4 sdur4 sdur4 (#1) (#2) (#3) (#4) (#5) (#6) (#7) Rey & Matsui [Page 11] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 |----------PAYLOAD 1------| |--PAYLOAD 2---| |--PAYLOAD 3---| rtpts1 rtpts2 rtpts3 Using a particular configuration from Figure 1 to we will illustrate how the timestamp for each unit is found. Assuming TSx means Text Sample x, rtptsy represents the standard RTP timestamp for PAYLOAD y and sdurz the duration of unit z, the timestamp for unit #z (ts#z) can be found as the sum of rtptsy plus the cumulative sum of the durations of preceding units in that payload (except in the case of PAYLOAD 3 as per rule 3 above). Thus, we have: 1. for the units in the first aggregate payload, PAYLOAD 1: ts(#1)= rtpts1, ts(#2)= N/A ts(#3)= rtpts1 + sdur1, ts(#4)= rtpts1 + sdur1 + sdur2, Note that no sdur value is assigned to TYPE 5 units, and they are taken into account in the timestamp calculation. 2. for PAYLOAD 2: ts(#5)= rtpts2, 3. for PAYLOAD 3: ts(#6)= ts(#7)= rtpsts2= rtpts3 In this case rtpts3 must be equal to rtpts2 in order to be able to assign this fragment to the correct sample (TS4). Additionally, note that units 6 and 7 have the same timestamp since they belong to the same text sample (also TS4). 2.3.5. Reassembling text samples at the receiver The payload headers defined in this document allow reassembling fragmented text samples. For this purpose the standard RTP timestamp, the duration indication (SDUR) and the total and subtotal counters (see TOTAL, THIS) field of the payload headers are used. The process for collecting the different fragments (units) of a text sample is as follows: 1. Search for units having the same timestamp value, i.e. belonging to the same text sample. 2. Check within this set whether any of the units from the text sample is missing. This is done using the TOTAL and THIS fields; the TOTAL field indicates how many fragments were created out of the text sample and the THIS field indicates the Rey & Matsui [Page 12] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 position of this fragment in the text sample. As result of this operation two outcomes are possible: a. No fragment is missing. Then the THIS field SHALL be used to order the fragments and reassemble the text sample before forwarding it to the decoding application. Special care SHALL be taken when reassembling the text string as indicated in bullet 4 below. b. One or more fragments are missing: check whether this fragment belongs to the text string or to the modifiers: TYPE 2 units identify text string fragments, TYPE 3 and 4 modifier fragments: i. if the fragment or fragments missing belong to the text string and the modifiers were received complete, then the received text characters MAY, at least, be displayed as plain text. Some modifiers MAY only be applied as long as it is possible to identify the character numbers, e.g. if only last text string fragment is lost. This is the case for modifiers defining specific font styles ('styl'), highlighted characters ('hlit'), karaoke feature ('krok)' and blinking characters ('blnk'). Other modifiers such as 'dlay' or 'tbox' can be applied without the knowledge of the character number. It is an application issue to decide whether to use apply the modifiers or not. ii. if the fragment missing belongs to the modifiers and the text strings were received complete, then the incomplete modifiers MAY be used. The text string SHOULD at least be displayed as plain text. As mentioned in Section 2.3.3, modifiers MAY split without observing meaningful boundaries. Hence, it may not always be possible to make use of partially received modifiers. Again, to avoid this case, it is RECOMMENDED that the modifiers do split at meaningful boundaries. iii. a third possibility is that it is not possible to discern whether modifiers or text strings were received complete. E.g. if the TYPE 3 unit of a sample plus the following or preceding packet is lost, there is no way for the RTP receiver to know if one if both packets lost belong to the modifiers or there is also some text strings. FEC [7], retransmission [11] or other protection mechanisms as per section 4 are RECOMMENDED to avoid this situation. iv. finally, if it is sure that neither text strings nor modifiers were received complete, then the text strings and the modifiers MAY be rendered partially or MAY be discarded. This is an application choice. Rey & Matsui [Page 13] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 3. Sample descriptions can be directly associated with the reassembled text samples, via the sample description index (SIDX). 4. Reassembling of text strings: since the text strings transported in RTP packets MUST NOT include any byte order mark (BOM), the receiver MUST prepend it to the reassembled string (if needed) before handling it to the timed text decoder. This is needed for UTF-16 encoded strings (i.e. "U" bit is set to 1). The value of the BOM is 0xFEFF (see [1]) and it is used by the 3GPP timed text decoder to recognize the UTF encoding. 2.3.6. Live streaming vs. Streaming from a 3GP file This section addresses the differences between streaming live content and streaming text tracks from a 3GP file. For the purpose of this document, the term live streaming refers to those scenarios where the timed text stream is sent from a live encoder. Upon reception the content may or may not be stored in a 3GP file. At the sender, timed text content SHALL be encapsulated in RTP packets following the guidelines given in this document. At the receiving side, a buffer is typically used to cancel the network delay and delay jitter. If receiver and sender support packet loss resilience mechanisms (i.e. retransmission [11], packet FEC [7] or other) it may also be possible to recover from packet losses. Note that how sender and receiver actually manage and dimension the buffers are implementation design choices. Section 12.3 specifies how the 3GP file parameters are mapped to the fields of the payload header. For live streaming, appropriate values complying with the format and units described in [1] shall be used. Where needed, clarifications on appropriate values are given in this document. 3. RTP Payload Format for 3GPP Timed Text The format of an RTP packet containing 3GPP timed text is shown below: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |V=2|P|X| CC |M| PT | sequence number | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | timestamp | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | synchronization source (SSRC) identifier | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Rey & Matsui [Page 14] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 | | + RTP payload | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Marker bit (M): the marker bit SHALL be set to 1 if the RTP packet includes one or more whole text samples or the last fragment of a text sample; otherwise set to zero (0). Timestamp: the timestamp MUST indicate the sampling instant of the earliest (or only) unit contained in the RTP packet. The initial value SHOULD be randomly determined, as specified in RTP [3]. In an aggregate payload, units MUST be placed in play-out order, i.e. earliest first in the payload. If TYPE 1 units are aggregated, the timestamp of the subsequent units MUST be obtained by adding the timed text sample duration of previous samples to the RTP timestamp value. Refer to the details on the timestamp calculation for units in an aggregate payload in Section 2.3.4. Note that TYPE 5 units do not make use of the timestamp and therefore the above does not apply for this particular case. The timestamp clockrate of the samples in each text track is the value of the "timescale" parameter in the Media Header Box for that text track. Note that each track in a 3GP file MAY have its own clockrate as specified in the Media Header Box. For live streaming an appropriate timestamp clockrate SHALL be used. A default value of 1000 Hz is RECOMMENDED. This value should provide enough timing resolution for expressing the duration of text samples, for synchronizing text with other media and for performing RTCP measurements such as the interarrival delay jitter or the RTCP Packet Receipt Times Report Block (Section 4.3 of RFC 3611 [20]). This is compliant to RTP [3], section 5.1: "The resolution of the clock MUST be sufficient for the desired synchronization accuracy and for measuring packet arrival jitter (one tick per video frame is typically not sufficient)" Other timestamp clockrates MAY be used. However, using too low clockrates may turn the RTCP measurements useless or may not provide enough synchronization accuracy. If this is the case, then such clockrate values SHALL NOT be used. Timestamp clockrates MUST be signaled by out-of-band means at session setup, e.g. using the "rate" attribute in SDP. See Section 8 for details. Unlike in other media like audio or video, no default sample size or sampling rate is defined for timed text. For 3GP files the length of the text samples is calculated beforehand and included in the track itself and for live encoding, it is the real time encoder that SHALL choose an appropriate size for each text sample. In general, the Rey & Matsui [Page 15] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 amount of text 'captured' in a sample depends on the text source and the particular application (see examples below). Samples may, e.g., be tailored to match the packet MTU as close as possible or to provide a given redundancy for the available bit rate. The encoding application MUST also take into account the delay constraints of the real-time session and assess whether FEC, retransmission or other similar techniques are reasonable options for stream repair. The following examples shall illustrate how a real-time encoder may choose its settings to adapt to the scenario constraints. Imagine a newscast scenario, where the spoken news is transcribed and synchronized with the image and voice of the reporter. We assume that the news speaker talks at an average speed of 5 words per second with an average word length of 5 characters plus one space per word, i.e. 30 characters per second. We assume an available IP MTU of 576 bytes and an available bitrate of 576*8bits per second=4.6Kbps. We assume each character can be encoded using 2-bytes in UTF-16. Several constraints may come in this scenario into place, for example: available IP MTU, available bandwidth, allowable delay and required redundancy. If the target were to minimize the packet overhead, a text sample covering 8 seconds of text would be closest to the IP MTU: IP/UDP/RTP/TYPE1 Header + (8s text sample)=20+8+12+8+(~6 chars/word * 5 word/s * 8s *2 chars/word)= 528 bytes < 576 bytes. For other scenarios, a delay of 8 seconds may be too much and just one packet per sample too low of a redundancy. If lower delay and higher redundancy is required, a choice could be that the encoder 'collects' text every second; this yields text samples (TYPE 1 units) of 68 bytes, TYPE 1 header included. Taking a smaller delay of 3s, three contiguous text samples could be aggregated in one RTP payload: the current and last two text samples. This accounts to a total IP packet size of 20+8+12+3*(8+60)= 244 bytes. Now, with the same available bitrate of 4.6Kbps, these 244-byte packets can be sent redundantly up two times per second, without exceeding the available bandwidth: RTP payload (1,2,3),(1,2,3) (2,3,4),(2,3,4) (3,4,5),(3,4,5) ... Time: <-----1s------> <-----1s------> <-----1s------> ... This means that each text sample is sent at least six times, which should provide enough redundancy. Although not as bandwidth efficient (488*8 < 528*8 < 576*8 bps) as the previous packetization, this option increases the stream redundancy while still meeting the delay and bandwidth constraints. Another example would be a user sending timed text from a type- in area in the display. In this case, the text sample is created as soon as the user clicks the 'send' button. Depending on the packet length, fragmentation may be needed. Rey & Matsui [Page 16] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 In a video conferencing application, text is synchronized with audio and video. Thus, the text samples shall be displayed long enough to be read by a human, shall fit in the video screen and shall 'capture' the audio contents rendered during the time the corresponding video and audio is rendered. Payload Type (PT): the payload type is set dynamically and sent by out-of-band means. The usage of the remaining RTP header fields follows the rules of RTP [3] and the profile in use. 3.1. Payload Header Definitions An RTP packet using the payload headers defined in this document has the following format: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |V=2|P|X| CC |M| PT | sequence number | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | timestamp | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | synchronization source (SSRC) identifier | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R | TYPE| LEN | : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- : : (variable header fields depending on TYPE : : : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | : SAMPLE CONTENTS : : : : : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Figure 2 RTP Packet Format. The payload headers specified in this document consist of a set of common fields followed by specific fields for each header type and sample contents. See Figure 3. In this manner, the structure of the payload headers resembles that of the 'access units' (AU) in RFC 3640 [12]. This similarity is intentional to improve interoperability. The 'AU header' of that document finds an equivalent in the common header fields: U, R, TYPE and LEN. Similarly, the specific fields plus the sample contents would be equivalent to the 'AU data section' of RFC 3640. This is illustrated in the figure below. =~'AU header' | =~'AU data' Rey & Matsui [Page 17] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-....+ |U| R |TYPE | LEN | (variable) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-....+ Figure 3 Unit Format. An aggregate RTP payload containing two complete text samples and a payload with one text sample fragment would schematically look like this: +----------------------+ | | | RTP Header | | | --------_+----------------------+ | | | _ | | TYPE 1 Header | | ........................ UNIT 1 - | | | | Text Sample | | _ | | |-------\........................ -------/| | | | TYPE 1 Header | | ........................ UNIT 2 - | | | | Text Sample | | | | | _ | | --------------------------------+ +----------------------+ | | | RTP Header | | | --------_+----------------------+ | |TYPE 2(or 3 or 4)Hdr | | ........................ UNIT 3 - | | | | Text Sample Fragment | |_ | | | _ | | ---------+----------------------+ Figure 4 Example RTP packets. 3.1.1. Common Payload Header Fields The fields common to all payload headers have the following format: Rey & Matsui [Page 18] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 0 1 2 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Figure 5 Common payload header fields. Where: o U (1 bit) "UTF Transformation flag": indicates whether the text characters are encoded using UTF-8 (U=0) or UTF-16 (U=1). This is used to inform RTP receivers whether UTF-8 or UTF-16 was used to encode the text string. Since this bit is used, no byte order mark (BOM) is needed inside the text string itself. For the payload formats defined in this document, the U bit is only used in TYPE 1 and TYPE 2 headers. Senders MUST set the U bit to zero in TYPE 3, TYPE 4 and TYPE 5 headers. Receivers MUST ignore the U bit in TYPE 3, TYPE 4 and TYPE 5 headers. o R (4 bits) "Reserved bits": for future extensions. This field MUST be set to zero (0x0) and MUST be ignored by receivers. o TYPE (3 bits) "Type Field": this field specifies which specific header fields follow. The following TYPE values are defined: - TYPE 1, for a whole text sample - TYPE 2, for a text string fragment (without modifiers) - TYPE 3, for a whole modifier box or the first fragment of a modifier box - TYPE 4, for a modifier fragment other than first. - TYPE 5, for a sample description. One header per sample description. - TYPE 0, 6 and 7 are reserved. o Finally, the LEN (16 bits) "Length Field": indicates the size (in bytes) of this header field and all the fields following, i.e. the LEN field followed by the unit payload (text strings and modifiers, if any). In order to calculate the LEN value, the text sample length in the Sample Size Box of 3GP files is used (see Section 12.3). For live streaming, both sample length and the LEN value for the current fragment MUST be calculated during the sampling process or during fragmentation. In general, LEN may take the following values: - TYPE = 1, LEN >= 8, - TYPE = 2, LEN >= 9, - TYPE = 3, LEN > 6, Rey & Matsui [Page 19] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 - TYPE = 4, LEN > 6 and, - TYPE = 5, LEN > 3. In the next subsection the different payload headers for the values of TYPE are specified. 3.1.2. TYPE 1 Header 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN (always >=8) | SIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | SDUR | TLEN | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | TLEN | +-+-+-+-+-+-+-+-+ This header type is used to transport whole text samples. If several text samples are sent in an RTP packet, every sample is preceded by its own header (see Figure 4). Note that also empty text samples are considered whole text samples, although they do not contain sample contents. Empty text samples may be used to clear the display or to put an end to samples of unknown duration, for example. Units without sample contents SHALL have a LEN field value of 8 (0x0008). The fields above have the following meaning: o U, R and TYPE as defined above. o SIDX (8 bits) "Text Sample Entry Index": this is an index used to identify the sample descriptions. The SIDX field is used to find the sample description corresponding to the unit's payload. There are two types of SIDX values: static and dynamic. Static SIDX values are used to identify sample descriptions that MUST be sent out-of-band and MUST remain active during the whole session. The transport of sample descriptions out-of-band is a MANDATORY feature. A static SIDX value is unequivocally linked to one particular sample description during the whole session. It SHOULD be avoided that many sample descriptions are carried out- of-band, since these may become large and, ultimately, transport is not the goal of the out-of-band channel. Thus, this feature is RECOMMENDED for transporting those sample descriptions that provide a set of minimum default format settings. Static SIDX values MUST fall in the (inclusive) interval [129,254]. Dynamic SIDX values are used for sample descriptions sent in-band. Sample descriptions MAY be sent in-band for several reasons: Rey & Matsui [Page 20] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 because they are generated in real time, for transport resiliency or both. A dynamic SIDX value is unequivocally linked to one particular sample description during the period in which this is active in the session and it SHALL NOT be modified during that period. This period MAY be smaller than or equal to the session duration. This period is not known a priori. A maximum of 64 dynamic active SIDX is allowed at any moment. Dynamic SIDX values MUST fall in the inclusive interval [0,127]. This should be enough for both, recorded content and live streaming applications. Nevertheless, a wrap-around mechanism is provided in Section 12.1 to handle sessions where more than 64 SIDX values might be needed. Servers MAY make use of dynamic sample descriptions. Clients MUST be able to receive and interpret dynamic sample descriptions. Finally, SIDX values 128 and 255 are reserved for future use. o SDUR (24 bits) "Text Sample Duration": indicates the sample duration in RTP timestamp units of the text sample. For this field, a length of 3 bytes is preferred to 2 bytes. This is because, for a typical clockrate of 1000 Hz, 16 bits would allow for a maximum duration of just 65 seconds, which might be too short for some streams. Apart from defining the time period during which the text is displayed, the duration field is also used to find the timestamp of any subsequent units within the RTP packet. Text samples have generally a known duration at the time of transmission. However, in some cases like live streaming, the time at which a text piece shall be presented might not be known a priori. For this case, the value zero SDUR=0 (0x000000) is reserved to signal unknown duration. However, upon storing a text sample with SDUR=0 in a 3GP file, the SDUR value MUST be changed to the effective duration of the text sample, which MUST be always greater than zero (note that the ISO file format [2] explicitly forbids a sample duration of zero). The effective duration MUST be calculated as the timestamp difference between the current sample (with unknown duration) and the next text sample that is displayed. The next example illustrates how units of unknown duration MUST be presented. If no text sample following is available, it is an implementation issue what should be displayed. E.g. a server could send an empty sample to clear the text box. Let us revisit a previous example, imagine now you are in an airport watching the latest news report while you wait for your plane. Airports are loud, so the news report is transcribed in the lower area of the screen. This area displays two lines of text: the headlines and the words spoken by the news speaker. As usual, the headlines are shown for a longer time than the rest. This time is, in principle, unknown to the stream server, Rey & Matsui [Page 21] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 which is streaming live. A headline is just replaced when the next headline is received. Note that samples of unknown duration SHALL NOT use features, such as scrolling or karaoke, which require knowledge of the duration of the sample up front. Furthermore, only sample descriptions (TYPE 5) MAY follow units of unknown duration in the same aggregate payload. Otherwise, it would not be possible to calculate the timestamp. For text stored in 3GP files, see Section 12.3 for details on how to extract the duration value. For live streaming, live encoders SHALL assign appropriate values and units according to [1] and later releases. o TLEN (16 bits), "Text String Length", is a byte-count of the text string. The text string length is needed by the decoder to know where the modifiers in the payload start. TLEN is not present in text string fragments (TYPE 2) since it can be deductively calculated from the LEN values of each fragment. Please refer to Section 12.3 about how to obtain the text string length from a 3GP file. o Finally, the sample contents following the TLEN field consists of a string of characters encoded using either UTF-8 or UTF-16, followed by zero or more modifiers. 3.1.3. TYPE 2 Header 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN( always >9) | TOTAL | THIS | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | SDUR | SIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | SLEN | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ This header type is used to transport either a whole text string or a fragment of it. TYPE 2 units SHALL NOT contain modifiers. In detail: o The U, R, TYPE, SIDX, and SDUR fields have similar interpretation as above. The U, SIDX and SDUR fields are meaningful since partial text strings can also be displayed. o The LEN field (16 bits) has the same meaning as above. For this header, it MUST always be greater than nine (0x0009). As mentioned in Section 2.3.3 text strings MUST be split at character boundaries to allow the display of text fragments. Rey & Matsui [Page 22] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 Hence, as a minimum a text fragment MUST contain one character in either UTF-8 or UTF-16. Actually, this is just a formalism since by following the fragmentation guidelines much larger fragments should be created. Note also, that TYPE 2 units do not contain an explicit text string length (like TLEN in TYPE 1). If all TYPE 2 units of a sample are received, then the original string length can be obtained deductively from the LEN of the fragments. o The SLEN field (16 bits) indicates the size (in bytes) of the original (whole) text sample to which this fragment belongs. This length comprises the text string plus any modifier boxes present (and does not include the byte order mark and the text string length as mentioned in the terminology section). For stored content, see Section 12.3 for details on how to find the SLEN value in a 3GP file. For live content, the SLEN MUST be obtained during the sampling process. Clients MAY use SLEN to buffer space for the remaining fragments of a text sample. o The fields TOTAL (4 bits) and THIS (4 bits) indicate the total number of fragments in which the original text sample (i.e. text string and its modifiers) has been fragmented and which order occupies the current fragment in that sequence, respectively. The usual "byte offset" field is not used here for two reasons: a) it would take one more byte and b) it does not provide any information on the character offset. UTF-8/UTF-16 text strings have, in general, a variable character length ranging from 1 to 6 bytes. Therefore, the TOTAL/THIS solution is preferred. It could also be argued that the LEN and SLEN fields be used for this purpose, but while they would provide information about the completeness of the text sample, they do not specify the order of the fragments. o Finally, the sample contents following the SLEN field consists of a fragment of the UTF-8/UTF-16 character string. Again, no modifiers follow. 3.1.4. TYPE 3 Header 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN( always >6) |TOTAL | THIS | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | SDUR | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ This header type is used to transport either the entire modifier contents present in a text sample or just the first fragment of these. Rey & Matsui [Page 23] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 This depends on whether the modifier boxes fit in the current RTP payload. If a text sample containing modifiers is fragmented this header MUST be used to transport the first fragment or, if possible, the complete modifiers. In detail: o The U, R, TOTAL/THIS and LEN fields are used as above. The LEN field MUST be greater than six (0x0006). o The TOTAL/THIS field has the same meaning as for TYPE 2. The THIS field is counting the number of units (TYPE2, TYPE 3, TYPE 4) used for fragmenting a text sample. Therefore, the last (trailing) modifier fragment are transported in a unit in which TOTAL=THIS. In this case, TOTAL=THIS MUST be greater than one, because TOTAL indicates the total number of fragments of the text sample, which is logically, always larger than one. Otherwise, if TOTAL is different from THIS in a TYPE 3 unit, this unit just contains the first fragment of the modifiers. o The SDUR has the same definition as above. Since the fragments are always transported in own RTP packets, this field is only needed to know how long this fragment is valid. This may, e.g., be used to determine how long it should be kept in the display buffer. Note that the SLEN and SIDX fields are not present. This is because: a) these fragments do not contain text strings and b) these types of fragments are applied over text string fragments, which already contain this information. 3.1.5. TYPE 4 Header 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN( always >6) |TOTAL | THIS | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | SDUR | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ This header type is prepended to modifier fragments, other than the first one. The U, R, TOTAL/THIS and LEN fields are used as above. The LEN field MUST be greater than six (0x0006). Regarding the SDUR field and the absence of the SLEN and SIDX fields, the same reasoning as for TYPE 3 applies. Rey & Matsui [Page 24] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 3.1.6. TYPE 5 Header 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |U| R |TYPE | LEN( always >3) | SIDX | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ This header type is used to transport (dynamic) sample descriptions. The LEN field MUST be greater than three (0x0003). Every sample description MUST have its own TYPE 5 header. The SIDX value is in the (inclusive) interval [0,127], as this unit contains a dynamic sample description. Senders MAY make use of TYPE 5 units. All receivers MUST implement support for TYPE 5 units, since it adds minimum complexity and it may increase the robustness of the streaming session. Finally, if sample descriptions for a given SIDX value are not available at the receiver, it is a matter of implementation whether the text sample contents are displayed. For example, an application MAY provide a static default sample description to be used for these cases. This is, however, an implementation issue and out of the scope of this document. 4. Resilient Transport Apart from the basic fragmentation guidelines described in the section above, the simplest option for packet loss resilient transport is repetition. A server MAY decide to use repetition as a measure for packet loss resilience. Thereby, a server MAY send the same RTP packet payloads or just parts of it, i.e. single units. As for the case of complete payloads, single repeated units MUST match exactly the same units sent in the first transmission, i.e. if fragmentation is needed it SHALL be performed only once for each text sample Only then, a receiver can use the already received and the repeated units to reconstruct the original text samples. Since the RTP timestamp is used to group together the fragments of a sample, care must taken to preserve the timing of units when constructing new RTP packets. E.g. if a text sample was originally sent as a single non- fragmented text sample (one TYPE 1 unit), a repetition of that sample MUST be sent also as a single non-fragmented text sample in one unit. Likewise, if the original text sample was fragmented and spread over several RTP packets, say a total of 3 units, then the repeated fragments SHALL also have the same byte boundaries and use the same unit headers and bytes per fragment. Rey & Matsui [Page 25] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 With repetition, repeated units resolve to the same timestamp as their originals. Where redundant units are available, the receiver SHOULD use those units received in the RTP packet with the highest sequence number. Regarding the RTP header fields: o if the whole RTP payload is repeated, all payload-specific fields in the RTP header (the M, TS and PT fields) MUST keep their original values except the sequence number that MUST be incremented to comply with RTP. o in packets containing single repeated units, the general rules in Section 3 for assigning values to the RTP header fields apply. Particularly relevant here is to keep the value of the RTP timestamp to preserve the timing of the units. Apart from repetition other mechanisms such as FEC [7], retransmission [11] or similar techniques SHOULD be used to cope with packet losses. 5. Congestion control Congestion control for RTP SHALL be implemented in accordance with RTP [3], and the applicable RTP profile, e.g. RTP/AVP [17]. The RTP profile under which this payload format is used defines an appropriate congestion control mechanism in different environments. Following the rules under the profile, an RTP application can determine its acceptable bitrate and packet rate in order to be fair to other TCP or RTP flows. 6. Scene Description 6.1. Text rendering position and composition In order to set up a timed text session, regardless of the stream being stored in a 3GP file or streamed live, some initial layout information is needed by the communicating peers. Some parameters are used to provide the client with information on the stream properties. These are the "width" and "height" of the text area, its position and its "layer" or proximity to the user. At the same time, the server needs to know the client's capabilities, i.e. the maximum allowable values for the text track height and width: "max-h" and "max-w". These pieces of information MUST be conveyed in a reliable form previous to the start of the session, e.g. during session announcement or in an O/A exchange. An example of a reliable Rey & Matsui [Page 26] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 transport may be the out-of-band channel used for SDP. Section 7 and 8 provide details on the usage in SDP descriptions. For stored content, the layout values expressing stream properties MUST be obtained from the Track Header Box. See Section 12.3 for details on finding these values in a 3GP file. For live streaming appropriate values SHALL be used. 6.2. SMIL usage Note that the attributes contained in the Track Header Boxes of a 3GP file only specify the spatial relationship of the tracks within the given 3GP file. If several media streams are sent, they require spatial synchronization. For such purpose, SMIL [9] SHOULD be used. SMIL assigns regions in the display to each of those files and places the tracks within those regions. The original track header information is used for each track within its region. Therefore, even if SMIL scene description is used, the track header information pieces SHOULD be sent anyway as they represent the intrinsic media properties. See [1] and the 3GPP SMIL Language Profile in [16] for details. 7. MIME Type usage Registration 7.1. 3GPP Timed Text MIME Registration The MIME subtype for the 3GPP Timed Text codec is allocated from the IETF tree. The MIME top-level type under which this payload format is registered is 'text'. The receiver MUST ignore any unspecified parameter. MIME Type: text MIME subtype: 3gpp-tt Required parameters rate: the RTP timestamp clockrate is equal to the clockrate of the media. If RTP packets are generated out of a 3GP file, the clockrate of the text media MUST be copied from the 3GP file, i.e. the clockrate is the value of "timescale" parameter in the Media Header Box describing that text track. Other tracks (audio/video/text) in the 3GP file may have their own clockrates as indicated in their corresponding Media Header Box. For live encoding, a clockrate of 1000 Hz is RECOMMENDED but other values MAY be used. If a different timestamp rate Rey & Matsui [Page 27] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 value is used, this MUST be chosen carefully as per Section 3 in RFCXXXX. sver: The parameter "sver" contains a list of supported backwards-compatible versions of the timed text format specification (3GPP TS 26.245) that the sender accepts to receive (and which are the same that it would be willing to send). The first value is the value preferred to be received ( or preferred to send). The first value MAY be followed by a comma-separated list of versions that SHOULD be used as alternatives. The order is meaningful, being first the most preferred and last the least preferred. Each entry has the format Zi(xi*256+yi), where "Zi" is the number of the Release, "xi" and "yi" are taken from the 3GPP specification version, i.e. vZi.xi.yi. For example, for 3GPP TS 26.245 v6.0.0, Zi(xi*256+yi)=6(0), the version value is "60". (Note that "60" is the concatenation of the values Zi=6 and (xi*256+yi)=0 and not its product.) width: This parameter indicates the width in pixels of the text track or area of the text being sent. This is a 16 bit integer. height: This parameter indicates the height in pixels of the text track being sent. This is a 16 bit integer. max-w: This parameter indicates display capabilities. This is the maximum "width" value that the sender of this parameter supports. This is a 16 bit unsigned integer. max-h: This parameter indicates display capabilities. This is the maximum "height" value that the sender of this parameter supports. This is a 16 bit unsigned integer. tx: This parameter indicates the horizontal translation offset in pixels of the text track with respect to the origin of the video track. This is a 16 bit integer. ty: This parameter indicates the vertical translation offset in pixels of the text track with respect to the origin of the video track. This is a 16 bit integer. layer: This parameter indicates the proximity of the text track to the viewer. More negative values mean closer to the Rey & Matsui [Page 28] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 viewer. This parameter has no units. This is a 16 bit integer. Optional parameters: tx3g: This parameter MUST be used for conveying sample descriptions out-of-band. It contains a comma-separated list of base64-encoded entries. The entries of this list that MAY follow any particular order and the list MAY be empty. The absence of this parameter is equivalent to an empty list of sample descriptions. Each entry is the result of running base64 encoding over the concatenation of the SIDX and the sample description for that SIDX, in this order. The format of a sample description entry can be found in 3GPP TS 26.245 Release 6 and later releases. All servers and clients MUST understand this parameter and MUST be capable of using the sample description(s) contained in it. Please refer to RFC 3548 [6] for details on the base64 encoding. Encoding considerations: RTP payloads complying with this payload format contain binary data. Note that this type is incompatible with the use of text media types in other protocols, e.g. text/html. This is because in order to extract and decode any of the timed text media it is necessary understand the (binary) payload headers defined in RFCXXXX. Restrictions on usage: This type is only defined for transfer via RTP. Security considerations: Please refer to Section 10 of RFCXXXX. Interoperability considerations: The 3GPP Timed Text media format for which this payload format is defined is specified in Release 6 of 3GPP TS 26.245 "Transparent end-to-end packet switched streaming service (PSS); Timed Text Format (Release 6)". The 3GPP file format (3GP) and the SMIL language profile used can be found in Release 5 of 3GPP TS 26.234 and in the corresponding specifications for later Releases. Note also that 3GPP may in future Releases specify extensions or updates to the media format in a backwards- compatible way, e.g. new modifier boxes or extensions to the sample descriptions. The payload format defined in RFCXXXX Rey & Matsui [Page 29] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 allows for such extensions. For future 3GPP Releases of the Timed Text Format, the parameter "sver" is used to identify the exact specification used. Published specification: RFC XXXX Applications which use this media type: Multimedia streaming applications. Additional information: the 3GPP Timed Text media format is specified in 3GPP TS 26.245 "Transparent end-to-end packet switched streaming service (PSS); Timed Text Format (Release 6)". This document and future extensions to the 3GPP Timed Text format are publicly available at http://www.3gpp.org. Magic number(s): None. File extension(s): None. Macintosh File Type Code(s): None. Person & email address to contact for further information: Jose Rey, rey@panasonic.de Yoshinori Matsui, matsui.yoshinori@jp.panasonic.com Audio/Video Transport Working Group. Intended usage: COMMON Author/Change controller: Jose Rey Yoshinori Matsui IETF AVT WG 8. SDP usage 8.1. Mapping to SDP The information carried in the MIME media type specification has a specific mapping to fields in SDP [4]. If SDP is used to specify sessions using this payload format, the mapping is done as follows: o The MIME type ("text") goes in the SDP "m=" as the media name. m=text RTP/ Rey & Matsui [Page 30] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 o The MIME subtype ("3gpp-tt") and the timestamp clockrate "rate" (1000 Hz or other) go in SDP "a=rtpmap" line as the encoding name and rate, respectively: a=rtpmap: 3gpp-tt/1000 o The MANDATORY fmtp parameters "sver", "width", "height", "max-h", "max-w", "tx", "ty" and "layer" go in the SDP "a=fmtp" attribute by copying them directly from the MIME media type string as a semicolon separated list of parameter=value pairs. o The OPTIONAL parameter "tx3g" goes in the SDP "a=fmtp" attribute by copying them directly from the MIME media type string as a semicolon separated list of values: a=fmtp: =[, =] o Any unknown parameter SHALL be ignored. 8.2. Parameter Usage in the SDP Offer/Answer Model In this section the meaning of the SDP parameters defined in this document within the Offer/Answer (O/A) [13] context is explained. In unicast, sender and receiver typically negotiate the streams, i.e. which codecs and parameter values are used in the session. This is also possible in multicast to a lesser extend. As stated in the O/A model, some "fmtp" (payload-format-specific) parameters have a clear meaning and shall be included in the answer as present in the offer. Other parameters may need to be set among parties, because it is not clear that offerer and answerer shall use the same values. A third type of parameters may take different values for offerers and answerers. Additionally, the meaning of the parameter MAY vary depending on which direction it used. In the following sections, a " offer" means an offer that contains a stream set to . may take the values sendrecv, sendonly andrecvonly. Similar considerations apply for answers. E.g. an answer to sendonly offer is a recvonly answer. These considerations apply to both 3GP file streaming and live streaming scenarios. 8.2.1. Unicast Usage The following types of parameters are used in this payload format: 1. Declarative parameters: offerer and answerer declare the values they will use for the incoming (sendrecv/recvonly) or outgoing (sendonly) stream. They MAY use different values. Rey & Matsui [Page 31] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 a. "tx", "ty" and "layer": these are parameters describing where the text track is placed. Depending on the directionality: i. MUST appear in all sendrecv offers and answers and in all recvonly offers (thus applying to the incoming stream). ii. MUST appear in all recvonly answers. In this case, the answerer MAY accept the proposed values for the incoming stream or respond with different ones. The offerer MUST use the returned values. iii. MAY appear in sendonly offers and MUST appear in sendonly answers. In offers they specify the values that the offerer proposes for sending (useful for visually composed streams, see below). In answers these values MUST be copied from the corresponding recvonly offer upon accepting the stream (thus acknowledging the use of such). 2. Parameters describing the display capabilities, which MUST be included in all offers and answers where "tx" and "ty" refer to the incoming stream, thus excluding sendonly offers and answers (see examples below). "max-h" and "max-w" indicate the maximum dimensions of the text track (text display area) for the given "tx" and "ty" values. 3. Parameters describing the sent stream properties, i.e. the sender of the stream decides upon the values of these: a. "width" and "height", specify the text track dimensions. They SHALL ALWAYS be present in sendrecv and sendonly offers and answers. For recvonly answers, the answerer MUST include the offered parameter values (if any) verbatim in the answer upon accepting the stream. b. "tx3g", static sample descriptions, this is an OPTIONAL parameter. It MAY only be present in sendrecv and sendonly offers and answers. 4. Negotiable parameters, which MUST be agreed on. This is the case of "sver". This parameter MUST be present in every offer and answer. The answerer SHALL choose one supported value from the offerer's list or else it MUST remove the stream or reject the session. 5. Symmetric parameters: "rate", timestamp clockrate, belongs to this class. It must be echoed verbatim in the answer. Otherwise the stream MUST be removed or the session rejected. Other observations regarding parameter usage: o Translation and transparency values: in sendonly offers "tx", "ty" and "layer" indicate proposed values. This is useful for visually composed sessions where the different streams occupy Rey & Matsui [Page 32] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 different parts of the display, e.g., a video stream and the captions. These are just suggested values because it is the peer rendering the text that ultimately decides where to place the text track. o Text track (area) dimensions, "height" and "width": in the case of sendonly (sendrecv) offers, an answerer accepting the offer MUST be prepared to render (and send) the stream with the same exact values. If any of these conditions are not met, the stream MUST be removed or the session rejected. o Display capabilities, "max-h" and "max-w": an answerer sending a stream SHALL ensure that the "height" and "width" values in the answer are compatible with the offerer's signalled capabilities. o Version handling via "sver": the idea is that offerer and answerer communicate using the same version. This is achieved by letting the answerer choose from a list of supported versions, "sver". For recvonly streams, the first value in the list is the preferred version to receive. Consequently, for sendonly (and sendrecv) streams the first value is the one preferred for sending (and receiving). The answerer MUST choose one value and return it in the answer. Upon receiving the answer, the offerer SHALL be prepared to send (sendonly and sendrecv) and receive (recvonly and sendrecv) a stream using that version. If none of the versions in the list is supported the stream MUST be removed or the session rejected. 8.2.2. Multicast Usage In multicast the parameter usage is similar to the unicast case, except in the following cases: o the parameters "tx", "ty" and "layer" in multicast offers only have meaning for sendrecv and recvonly streams. In order for all clients to have the same vision of the session, they MUST be used symmetrically. o for "height", "width" and the OPTIONAL "tx3g" (for sendrecv and sendonly), multicast offers specify which values of these parameters the participants MUST use for sending. Thus, if the stream is accepted, the answerer MUST also here include them verbatim in the answer. o The capability parameters, "max-h" and "max-w", SHALL NOT be used in multicast. If the offered text track should change in size, a new offer SHALL be used instead. o Regarding version handling: In the case of multicast offers, an answerer MAY accept a multicast offer as long as one of the versions listed in the "sver" is supported. Therefore, if the stream is accepted, the Rey & Matsui [Page 33] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 answerer MUST choose its preferred version but, unlike in unicast, the offerer SHALL NOT change the offered stream to this chosen version because there may be other session participants that do support the newer extensions. Consequently, different session participants may end up using different backwards-compatible media format versions. It is RECOMMENDED that the multicast offer contains a limited number of versions, in order for all participants to have the same view of the session. This is a responsibility of the session creator. If none of the offered versions is supported, the stream SHALL be removed or the session rejected. 8.3. Offer/Answer Examples In these unicast O/A examples the long lines are wrapped around. Static sample descriptions are shortened for clarity. Sendrecv offer: O -> A m=text RTP/AVP 98 a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=100; layer=0; height=80; width=100; max-h=120; max-w=160; sver=6256,60; tx3g=81... a=sendrecv A -> O m=text RTP/AVP 98.. a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=95; layer=0; height=90; width=100; max-h=100; max-w=160; sver=60; tx3g=82... a=sendrecv In this example the offerer is telling the answerer where it will place the received stream and what is the maximum height and width allowable for the stream that it will receive. Also, it tells the answerer the dimensions of the text track for the stream sent and which sample description it shall use. It offers two versions, 6256 and 60. The answerer responds with an equivalent set of parameters for the stream it receives. In this case the answerer's "max-h" and "max-w" are compatible with the offerer's "height" and "width". Otherwise, the answerer would have to remove this stream and the answerer issue a new offer taking the answerer's capabilities into account. Note also that the answerer's text box dimensions fit within the within the maximum values signalled in the offer. Finally, the answerer chooses to use version 60 of the timed text format. For recvonly: Rey & Matsui [Page 34] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 Offerer -> Answerer m=text RTP/AVP 98 a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=100; layer=0; max-h=120; max-w=160; sver=6256,60 a=recvonly A -> O m=text RTP/AVP 98.. a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=100; layer=0; height=90; width=100; sver=60; tx3g=82... a=sendonly In this case, the offer is different from the previous case: it does not include the stream properties: "height", "width" and "tx3g". The answerer copies the "tx", "ty" and "layer" values, thus acknowledging these. "max-h" and "max-w" are not present in the answer because the "tx" and "ty" (and "layer") in this special case do not apply to the received, but to the sent stream. Also, if offerer and answerer had very different displays sizes, it would not be possible to express the answerer's capabilities. In the example above and for an answerer with a 50x50 display, the translation values are already out of range. For sendonly: O -> A m=text RTP/AVP 98 a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=100; layer=0; height=80; width=100; sver=6256,60; tx3g=81... a=sendonly A -> O m=text RTP/AVP 98.. a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=100; ty=100; layer=0; height=80; width=100; max-h=100; max-w=160; sver=60 a=recvonly Note that "max-h" and "max-w" are not present in the offer. Also, with this answer, the answerer would accept the offer as is (thus echoing "tx", "ty", "height", "width" and "layer") and additionally inform the offerer about its capabilities: "max-h" and "max-w". Another possible answer for this case would be: Rey & Matsui [Page 35] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 A -> O m=text RTP/AVP 98.. a=rtpmap:98 3gpp-tt/1000 a=fmtp:98 tx=120; ty=105; layer=0; max-h=95; max-w=150; sver=60 a=recvonly In this case the answerer does not accept the values offered. The offerer MUST use these values or else remove the stream. 8.4. Parameter Usage outside of Offer/Answer SDP may also be employed outside of the Offer/Answer context, for instance for multimedia sessions that are announced through the Session Announcement Protocol (SAP) [14], or streamed through the Real Time Streaming Protocol (RTSP) [15]. In this case, the receiver of a session description MUST support the parameters and given values for the streams or else it MUST reject the session. It is the responsibility of the sender (or creator) of the session descriptions to define the session parameters so that the probability of unsuccessful session setup is minimized. This is out of the scope of this document. 9. IANA Considerations IANA is requested to register the MIME subtype name "3gpp-tt" for the media type "text" as specified in Section 7 of this document. 10. Security considerations RTP packets using the payload format defined in this specification are subject to the security considerations discussed in the RTP specification [3] and any applicable RTP profile, e.g. AVP [RFC3551]. In particular, an attacker may invalidate the current set of valid sample descriptions at the client by means of repeating a packet with an old sample description, i.e. replay attack. This would mean that the display of the text would be corrupted, if displayed at all. Another form of attack may consist in sending redundant fragments, whose boundaries do not match the exact boundaries of the originals. This may cause a decoder to crash. These types of attack may easily be avoided by using source authentication and integrity protection. Additionally, peers in a timed text session may desire to retain privacy in their communication, i.e. confidentiality. Rey & Matsui [Page 36] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 This payload format does not provide any mechanisms for achieving these. Confidentiality, integrity protection and authentication have to be solved by a mechanism external to this payload format, e.g. SRTP [10]. 11. References 11.1. Normative References [1] Transparent end-to-end packet switched streaming service (PSS); Timed Text Format (Release 6), TS 26.245 v 0.1.6, Working Draft, July 2003. [2] ISO/IEC 14496-1:2001/AMD5, "Information technology û Coding of audio-visual objects û Part 1: Systems, ISO Base Media File Format", 2003. [3] H. Schulzrinne, S. Casner, R. Frederick and V. Jacobson, "RTP: A Transport Protocol for Real-Time Applications", STD 64, RFC 3550, July 2003. [4] M. Handley, V. Jacobson, "SDP: Session Description Protocol", RFC 2327, April 1998. [5] S. Bradner, "Key words for use in RFCs to indicate requirement levels," BCP 14, RFC 2119, IETF, March 1997. [6] S. Josefsson (Ed.), "The Base16, Base32, and Base64 Data Encodings", RFC 3548, July 2003. 11.2. Informative References [7] J. Rosenberg, H. Schulzrinne, "An RTP Payload Format for Generic Forward Error Correction", RFC 2733, December 1999. [8] C. Perkins, O. Hodson, "Options for Repair of Streaming Media", RFC 2354, June 1998. [9] W3C, "Synchronised Multimedia Integration Language (SMIL 2.0)", August, 2001. [10] M. Baugher, D. A. McGrew, D. Oran, R. Blom, E. Carrara, M. Naslund, K. Norrman, "The Secure Real-Time Transport Protocol", RFC 3711, March 2004. [11] J. Rey et al., "RTP Retransmission Payload Format", draft-ietf- avt-rtp-retransmission-10.txt, work in progress, January 2004. [12] Van der Meer et al., "RTP Payload Format for Transport of MPEG-4 Elementary Streams ", RFC 3640, November 2003. Rey & Matsui [Page 37] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 [13] J. Rosenberg., H. Schulzrinne, " An Offer/Answer Model with the Session Description Protocol (SDP)", RFC 3264, June 2002. [14] M. Handley, et al. "Session Announcement Protocol", RFC 2974, October 2000. [15] H. Schulzrinne, et al.,"Real Time Streaming Protocol (RTSP)", RFC 2326, April 1998. [16] Transparent end-to-end packet switched streaming service (PSS); Protocols and codecs (Release 5), TS 26.234 v 5.6.0, Working Draft, September 2003. [17] H. Schulzrinne, S. Casner, "RTP Profile for Audio and Video Conferences with Minimal Control", STD 65, RFC 3551, July 2003. [18] F. Yergeau, "UTF-8, a transformation format of Unicode and ISO 10646", RFC 2044, October 1996. [19] P. Hoffman, F. Yergeau, "UTF-16, an encoding of ISO 10646", RFC 2781, February 2000. [20] Friedman, et al., "RTP Control Protocol Extended Reports (RTCP XR)", RFC 3611, November 2003. [21] Ott, et al., "Extended RTP Profile for RTCP-based Feedback (RTP/AVPF)", draft-ietf-avt-rtcp-feedback-09.txt, work in progress, July 2004. 12. Annexes 12.1. Dynamic SIDX wrap-around mechanism The use of dynamic sample descriptions by senders is OPTIONAL. However, if used, senders MUST implement this mechanism. Receivers MUST always implement it. As mentioned in Section 3.1.2, dynamic SIDX values remain active either during the entire duration of the session (if used just once) or in different intervals of it (if used once or more). Although 64 sample descriptions should cover the needs of most timed text applications, a wrap-around mechanism to handle the exception is described here. In the following, SIDX value means dynamic SIDX value. There is a sliding window of 64 active SIDX values. Values within the window are active, all others are considered inactive. An SIDX value becomes "active" if at least one sample description identified by that SIDX has been received. Since sample descriptions MAY be sent redundantly, it is possible that a client receives a given SIDX several times. However, the receiver SHALL ignore redundant sample descriptions and it MUST use the already cached copy. The guard Rey & Matsui [Page 38] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 range of inactive values ensures that always the correct association SIDX <-> sample description is used. The following algorithm is used to maintain the dynamic SIDX values: Let X be the SIDX of the last received sample description. Let Y be a value within the allowed range for dynamic SIDX: [0,127], and different from X. 1. Initialize all dynamic SIDX values as inactive. For stored content, read the sample description index in the Sample to Chunk box ("stsc") for that sample. For live streaming, the first value MAY be zero or any other value in the interval above. The initial value is SIDX=X. Go to step 2. 2. First in-band sample description with SIDX=X is received. Go to step 3. 3. Set all SIDX=Y inactive if inside the interval [X+1 modulo(128), X+64 modulo(128)]. Otherwise, set SIDX=Y as active. Go to step 4. 4. Wait for next sample description. Upon reception of a sample description with SIDX=X do: a. If X is currently active, then wait for next SIDX (do nothing). b. Else go to step 3. Example, if X=4, any SIDX in the interval [5,68] is inactive. Active SIDX values are in the complementary interval [69,127] plus [0,4]. Once the client is initialized, the interval of active SIDX values MUST change whenever a sample description with an inactive SIDX value is received. E.g., if the client receives a SIDX=6, then the active interval is now different: [0,6] plus [71,127]. However, if the received SIDX is in the current valid interval no change SHALL be applied. This means that at any instant a maximum of 64 SIDX values are valid, whereas the total of values used might be over 64. 12.2. Basics of the 3GP File Structure This section provides a coarse overview of the 3GP file structure. Each 3GP file consists of "Boxes". Boxes start with a header, which indicates both size and type contained. In general, a 3GP file contains the File Type Box (ftyp), the Movie Box (moov), and the Media Data Box (mdat). The Movie Box and the Media Data Box, serving as containers, include own boxes for each media. Similarly, each box type may include a number of boxes. See ISO Base Media file Format [2] for a complete list of possibilities. In the following, only those boxes are mentioned, which are useful for the purposes of this payload format. Rey & Matsui [Page 39] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 The File Type Box identifies the type and properties of a 3GP file. The File Type Box contents comprise the major brand, the minor version and the compatible brands. When streamed with RTP, these are communicated via out-of-band means, such as SDP. The Movie Box (moov) contains one or more Track Boxes (trak) which include information about each track. A Track Box contains, among others, the Track Header Box (tkhd), the Media Header Box (mdhd) and the Media Information Box (minf). The Track Header Box specifies the characteristics of a single track, where a track is, in this case, the streamed text during a session. Exactly one Track Header Box is present for a track. It contains information about the track, such as the spatial layout (width and height), the video transformation matrix and the layer number. Since these pieces of information are essential and static, i.e. constant for the duration of the session, they MUST be sent prior to the transmission of any text samples. See the ISO base media file format [2] for details about the definition of the conveyed information. The Media Header Box contains the timescale or number of time units that pass in one second, i.e. cycles per second or Hertz. The Media Information Box includes the Sample Table Box (stbl) which itself contains the Sample Description Box (stsd), the Decoding Time to Sample Box (stts), the Sample Size Box (stsz) and the Sample to Chunk Box (stsc). Sample descriptions for each text sample are encoded as "tx3g" sample entries in the Sample Description Box (stsd). The Sample Table Box (stbl) contains all the time and data indexing of the media samples in a track. Using the tables here, it is possible to locate samples in time, determine their type, and determine their size, container, and offset into that container. Finally, the Media Data Box contains the media data itself. In timed text tracks this box contains text samples. Its equivalent to audio and video is audio and video frames, respectively. The text sample consists of the text length, the text string, and one or several Modifier Boxes. The text length is the size of the text in bytes. The text string is plain text to render. The Modifier Box is information to render in addition to the text such as colour, font, etc. 12.3. Usage of 3GP file information for transport in RTP For the purpose of streaming timed text contents, some values in the boxes contained in a 3GP file are mapped to fields of this payload header. This section explains where to find and how to use those values. From the Track Header Box (tkhd): o tx,ty: these values are the second but last and third but last values in the unity matrix. These values are fixed- Rey & Matsui [Page 40] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 point 16.16 values, restricted to be integers (the lower 16 bits of each value shall be zero). Therefore, only the first 16 bits are used. o width, height: they also have the same name in the box and the payload header. Similarly as above, only the first 16 bits are used, the rest is zero. o layer: all 16 bits are used. From the Sample Table Box (stbl) the following information is carried in each RTP packet using this payload format: o the Sample Description Box (stsd): this stsd box provides information on the basic characteristics of text samples. Each entry is a sample entry box of type "tx3g". An example of the information contained in a sample entry could be the font size or the background color. These pieces of information are commonly used by many text samples during the session. Each sample entry "tx3g" is transported either in- band or out-of-band. o the Decoding Time to Sample Box (stts): the 24 least significant bits of the "sample_delta" are mapped to the field SDUR (Text Sample Duration), o the Sample Size Box (stsz): the 16 least significant bits of the "sample_size" or "entry_size" (depending on whether the sample size is fixed or variable) indicate the length (in bytes) of the text string plus any modifier boxes that may be in that text sample. In 3GP files this value includes the text string length and, if present, the byte order mark (BOM). For the purposes of this payload format, text samples include neither text string length nor BOM. For obtaining the text sample length as defined in this payload format, it MUST be accordingly adjusted to exclude those, i.e. 2 units SHALL be subtracted for UTF-8 encoded samples and 4 units for UTF-16 encoded samples. The resulting value MUST be directly mapped to the SLEN field defined in the TYPE 2 header. At the receiver, the reverse operation MUST be done to re- assemble the text string and restore the sample length as stored in the 3GP file. Note also that the text string length found in each of the text samples of the text track may not always be mapped directly to the TLEN. This value SHALL be reduced by 2 units to exclude the BOM in UTF-16 strings. o the Sample to Chunk Box (stsc): the value of the "sample_description_index" for that sample in the Sample to Chunk Box is mapped to the field SIDX (Text Sample Entry Index). The Sample to Chunk Box (stsc) associates the text sample and its corresponding sample description entry in the Rey & Matsui [Page 41] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 Sample Description Box (stsd, see below). The Sample to Chunk Box can be used to associate a text sample with a sample description entry. Since the sample description may vary during the session, the association SDIX is sent together with the text samples using this payload format. 13. Acknowledgements The authors would like to thank Dave Singer, Jan van der Meer, Magnus Westerlund and Colin Perkins for their comments and suggestions to this document. 14. Author's Addresses Jose Rey rey@panasonic.de Panasonic European Laboratories GmbH Monzastr. 4c D-63225 Langen, Germany Phone: +49-6103-766-134 Fax: +49-6103-766-166 Yoshinori Matsui matsui.yoshinori@jp.panasonic.com Matsushita Electric Industrial Co., LTD. 1006 Kadoma Kadoma-shi, Osaka, Japan Phone: +81 6 6900 9689 Fax: +81 6 6900 9699 15. IPR Notices The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79. Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement Rey & Matsui [Page 42] Internet Draft Payload Format for 3GPP Timed Text September 10, 2004 this standard. Please address the information to the IETF at ietf- ipr@ietf.org. 16. Full Copyright Statement Copyright (C) The Internet Society (2004). This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights. This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Rey & Matsui [Page 43]