Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberCN1510881 A
Publication typeApplication
Application numberCN 200310118074
Publication date7 Jul 2004
Filing date20 Nov 2003
Priority date20 Dec 2002
Also published asEP1432196A1, US20040165527
Publication number200310118074.7, CN 1510881 A, CN 1510881A, CN 200310118074, CN-A-1510881, CN1510881 A, CN1510881A, CN200310118074, CN200310118074.7
Inventors谷晓原, ・黑肯伯格, 罗尔夫黑肯伯格, 宫崎秋弘,
Applicant松下电器产业株式会社
Export CitationBiBTeX, EndNote, RefMan
External Links: SIPO, Espacenet
Control traffic compression method
CN 1510881 A
Abstract  translated from Chinese
本发明涉及一种用于压缩媒体数据传输中的控制通信量的方法,所述媒体数据传输在可以获得的会话带宽的分配部分内使用实时传送协议(RTP)和实时控制协议(RTCP),所述实时传送协议(RTP)和实时控制协议(RTCP)特别用于在因特网协议(IP)网络中的实时或近乎实时的多媒体数据传送。 The present invention relates to a method for compressing media data transmission control traffic, said media data transmission using real-time transport protocol (RTP) can be obtained within the allocated portion of the session bandwidth and real-time control protocol (RTCP), the said real-time transport protocol (RTP) and Real Time Control Protocol (RTCP) are particularly useful in Internet Protocol (IP) network in real-time or near real-time multimedia data transmission. 为了优化RTCP通信量的带宽效率和降低RTCP报告/反馈间隔,本发明提供了一种方法,包括步骤:通过初始发送场境参数来初始化控制通信流的场境;如果必要的话,使用压缩的控制分组在会话期间更新所述场境。 To optimize bandwidth efficiency and reduce traffic RTCP RTCP reports / feedback interval, the present invention provides a method comprising the steps of: sending through initial context initialization parameters to control the traffic of context; if necessary, use compressed control packet during a session updating the context.
Claims(34)  translated from Chinese
1.一种用于压缩媒体数据传输中的控制通信量的方法,所述媒体数据传输在可以获得的会话带宽的分配部分内使用实时传送协议(RTP)和实时控制协议(RTCP),所述实时传送协议(RTP)和实时控制协议(RTCP)特别用于在因特网协议(IP)网络中的实时或近乎实时的多媒体数据传送,其中,所述方法包括步骤:通过初始发送场境参数来初始化控制通信流的场境;如果必要的话,使用压缩的控制分组在会话期间更新所述场境。 1. A method for compressing media data transmission control traffic, said media data transmission using real-time transport protocol (RTP) can be obtained within the allocated portion of the session bandwidth and real-time control protocol (RTCP), the real-time or near real-time multimedia data transmission in real-time transport protocol (RTP) and Real Time Control Protocol (RTCP) are particularly useful in Internet Protocol (IP) network, wherein the method comprises the steps of: sending through initial context initialization parameters context traffic flow control; if necessary, use compressed control packets during a session updating the context.
2.如权利要求1所述的方法,其中,场境参数被分类为静态场境参数和动态场境参数。 2. The method of claim 1, wherein the context parameter is a context parameter classified as static and dynamic context parameters.
3.如权利要求1或2所述的方法,还包括步骤:省略先验已知的场境参数。 The method according to 1 or 2 claim, further comprising the steps of: a priori known context argument is omitted.
4.如权利要求1-3之一所述的方法,其中,在至少一个初始化分组内发送所述静态场境参数。 4. A method as claimed in one of claims 1-3, wherein, in at least one initialization packet transmitted within said static context parameters.
5.如权利要求1-4之一所述的方法,其中,在初始化分组、刷新分组或压缩控制分组中发送动态场境参数。 5. A method as claimed in one of claims 1-4, wherein, in the initialization packet, the refresh control packet or compressed packet transmitted dynamic context parameters.
6.如权利要求5所述的方法,其中,在源描述(SDES)分组和BYE分组中进一步发送动态场境参数。 6. The method of claim 5, wherein the source description (SDES) packets and BYE packets sent further dynamic context parameters.
7.如权利要求5或6所述的方法,其中,所述方法用于在压缩器和解压器之间的压缩控制数据传输,所述方法还包括步骤:在初始化步骤之前,限定所述初始化分组、所述刷新分组和所述压缩控制分组的分组格式以及压缩器和解压器场境参数。 7. A method as claimed in claim 5 or claim 6, wherein, the method for compressing between the compressor and the decompressor control data transmission, the method further comprising the steps of: prior to the initialization step, defining the initialization grouping, the refresh packets and compressed control packets the packet format and the compression and decompression device context parameters.
8.如权利要求7所述的方法,其中,所述方法还包括步骤:如果必要的话,在所述解压器修理和恢复不同步的场境。 8. The method of claim 7, wherein said method further comprises the step of: if necessary, in the repair and restoration of the decompressor context are not synchronized.
9.如权利要求2-8之一所述的方法,其中,在所述初始化步骤中发送用于所述动态场境参数的初始化值作为基准。 9. The method as claimed in one of claims 2-8, wherein, in the step of initializing transmitting said dynamic context parameter value is used as a reference to initialize.
10.如权利要求2-9之一所述的方法,还包括步骤:在发送所述静态场境参数之前,从所述静态场境参数形成至少一个初始化分组。 10. The method according to one of claims 2-9, further comprising the step of: prior to transmitting said static context parameters from said static context parameters forming at least one initialization packet.
11.如权利要求2-10之一所述的方法,还包括步骤:在发送所述动态场境参数之前,从所述动态场境参数形成刷新分组和压缩的控制分组。 11. The method as claimed in one of claims 2-10, further comprising the step of: before sending the dynamic context parameters, forming refresh packets and control packets from the compression dynamic context parameters.
12.如权利要求11所述的方法,其中,形成刷新分组和压缩的控制分组的步骤还形成源描述分组和BYE分组。 12. The method of claim 11, wherein the forming refresh packets and control packets compression step of further forming a source description packets and BYE packet.
13.如权利要求2-12之一所述的方法,还包括步骤:将所述动态场境参数分类为偶尔改变的场境参数、随机字符的场境参数、类似计数器的场境参数、经常改变的场境参数和规则地改变固定的增量的场境参数。 13. The method according to claim one of 2-12, further comprising the step of: the dynamic context parameters are classified as occasional change of context parameters, context parameters of random characters, context parameters similar counter, often change in fixed increments of change of context parameters and context parameters regularly.
14.如权利要求13所述的方法,还包括步骤:利用最小有效位(LSB)编码来编码所述类似计数器的场境参数、所述经常改变的场境参数和所述规则地改变固定的增量的场境参数。 14. The method of claim 13, further comprising the step of: utilizing the least significant bit (LSB) encoding to encode the context parameters similar counter, the constantly changing context parameters change regularly and the fixed context parameter increment.
15.如权利要求12-14之一所述的方法,其中,形成刷新分组和压缩的控制分组的步骤将所述偶尔改变的场境参数和所述随机字符的场境参数以未编码的形式集成到所述形成的分组,并且将所述类似计数器的场境参数、所述经常改变的场境参数和所述规则地改变固定的增量的场境参数以编码的形式集成到所述分组中。 15. The method as claimed in one of claims 12-14, wherein the step of forming the context parameter refresh control packets and compressed packets will occasionally change the parameters of the context and the random characters to form uncoded integrated into the formed packet, and the context of the similar counter parameter, the constantly changing context parameters and the rules changed in fixed increments of context parameters coded form integrated into the packet in.
16.如权利要求3-14之一所述的方法,其中,所述先验已知的场境参数包括控制协议版本。 16. The method as claimed in one of claims 3-14, wherein said a priori known context parameters include control protocol version.
17.如权利要求2-16之一所述的方法,其中,所述静态场境参数包括:填充标志,用于指示是否发送器报告分组在发送器报告分组的尾部包含一个附加填充字段,它不作为场境参数的一部分;至少一个同步源(SSRC)标识符,用于标识媒体数据传输的分组发送器或源;至少一个贡献源(CSRC)标识符,用于识别是至少一个向数据分组增加内容的源。 17. The method as claimed in one of claims 2-16, wherein said static context parameters include: filling flag that indicates whether the sender report packet reporting at the end of the packet transmission device comprises an additional padding field, it not part of the context parameters; at least one synchronization source (SSRC) identifier for the media data transmission packet transmitter or source; at least one contributing source (CSRC) identifier for identifying at least one data packet to increase source content.
18.如权利要求13-17之一所述的方法,其中,所述偶尔改变的场境参数包括分组的下列字段:接收报告计数(RC)字段,它指示在分组中报告块的数量;源计数(SC)字段,它们指示在源描述(SDES)分组中的同步源(SSRC)或贡献源(CSRC)的数量,或者识别在BYE分组中的同步源(SSRC)或贡献源(CSRC)的数量;有效载荷类型(PT)字段,它识别分组类型;源描述(SDES)项目,包括描述分组源属性的信息;子类型字段,在应用限定的(APP)分组内,允许在唯一名称下定义一组应用限定的(APP)分组。 18. The method as claimed in one of claims 13-17, wherein said context parameters occasionally changed packet includes the following fields: receiving a report count (RC) field, which indicates the number of blocks in the packet report; source count (SC) fields, which indicates the number of synchronization source (SSRC) or contributing source (CSRC) in a source description (SDES) packets, or identify the BYE packet synchronization source (SSRC) or contributing source (CSRC) of quantity; payload type (PT) field, which identifies the packet type; source description (SDES) project, including a description of the property information packet source; subtype field in the application defined (APP) packets, allows you to define in a unique name a set of application defined (APP) packets.
19.如权利要求18所述的方法,其中,所述源描述(SDES)项目包括:正规的端点标识符(CNAME)项目,用于描述用户和源的域名,用户名称(NAME)项目,用于描述源的通用名称,电子邮件地址(EMAIL)项目,用于描述源的电子邮件地址,电话号码(PHONE)项目,用于描述源的电话号码,地理用户位置(LOC)项目,用于描述源的地理位置,应用或工具名称(TOOL)项目,用于描述产生媒体数据的源应用的名称,注意或状态(NOTE)项目,用于描述源的状态的瞬态消息,专用扩展(PRIV)项目,用于定义实验或应用特定的扩展。 19. The method of claim 18, wherein said source description (SDES) include: a formal endpoint identifier (CNAME) project, and the source is used to describe the domain the user, the user name (NAME) project, with to describe the source of the common name, e-mail address (EMAIL) project to describe the source of the e-mail address, telephone number (PHONE) project to describe the source of the telephone number, geographic user location (LOC) project to describe source location, the name of the application or tool (TOOL) project, used to describe the name of the source application generating the media data, attention or status (NOTE) items for transient messages describing the source of state specific extensions (PRIV) project to define experimental or application specific extensions.
20.如权利要求13-19之一所述的方法,其中,所述随机字符的场境参数包括:部分损耗字段,用于指示被容许接收的分组的数量除损耗的分组的数量,和包括损耗分组的比特屏蔽的字段(BLP)。 20. The method according to one of claims 13-19, wherein said context parameters of random characters include: partial loss field for indicating the number of received packets is allowed in addition to the number of losses of packets, and including packet loss bits masked field (BLP).
21.如权利要求13-20之一所述的方法,其中,所述经常改变的场境参数包括:实时传送协议(RTP)时间标记字段,用于指示自所接收的最后一个发送器报告的延迟,最后的发送器报告的时间标志字段,到达之间抖动字段,用于指示实时协议(RTP)数据分组到达之间的时间的统计变化的估计值,长度字段,指示分组的长度。 21. The method according to one of claims 13-20, wherein the constantly changing context parameters include: Real-time Transport Protocol (RTP) timestamp field, for indicating the received since the last report of the transmitter delay, the final report of the time stamp transmitter jitter field, to reach the field, used to estimate the value of the statistics indicate a change in real-time protocol (RTP) packet inter-arrival time, the length field indicating the length of the packet.
22.如权利要求13-21之一所述的方法,其中,所述类似计数器的场境参数包括:实时协议(RTP)序号,指示所接收的分组的扩展的最高序号的字段,分组发送器的分组计数,用于指示在会话开始和包括发送器分组计数的分组的产生之间的时间帧中、发送器已经发送的实时协议(RTP)分组的总数;分组发送器的八比特组计数,用于指示在实时协议(RTP)分组中由发送器在会话开始和包括发送器的八比特组计数的分组的产生之间的时间帧中发送的有效载荷八比特组的总数;指示在发送期间损耗的分组的累计数量的字段。 22. The method according to one of claims 13-21, wherein the context parameters, the similar counter comprising: an extended highest sequence number packet field real-time protocol (RTP) sequence number, indicating the received packet transmitter The packet count for indicating a time frame in real time protocol session start and including a packet sender packet count generated between the transmitter has been transmitted (RTP) of the total number of packets; octet count packet transmitter, indicates the total number of real-time protocol (RTP) packet transmitted by the transmitter of the payload frame at the time the session starts and comprises a transmitter octet count packets generated between the groups of eight bits for; indicating during transmission the cumulative amount of packet loss fields.
23.如权利要求11-22之一所述的方法,其中,所述压缩的控制分组可以是发送器报告分组、接收器报告分组和应用限定(APP)分组。 23. The method according to one of claims 11-22, wherein the control packet may be compressed sender report packet, the receiver report packet and the application is defined (APP) packet.
24.如权利要求10或11所述的方法,其中,所述形成初始化分组的步骤形成初始化分组,包括:场境标识符,用于识别要用于解压分组的首标解压器的状态,分组标识符,用于使得分组接收器能够识别分组类型,规范概要信息,包括分组发送器的规范概要信息,循环冗余查看(CRC)字段,用于查看更新的分组的数据完整性;静态信息链,包括静态场境参数,和动态信息链,包括动态场境参数。 24. The method according to claim 10 or claim 11, wherein said step of forming the initialization packet initialization packet, comprising: a context identifier used to identify the state to be used for extracting a packet header decompressor, the packet identifiers for causing the receiver to identify the packet packet type, standardized summary information, including packet transmitter specification profiles, Cyclic Redundancy Check (CRC) field, for data integrity check for updated packet; static information chain including static context parameters, and dynamic information chain, including dynamic context parameters.
25.如权利要求11所述的方法,其中,所述形成刷新分组和压缩控制分组的步骤形成刷新分组,包括:场境标识符,用于识别要用于解压分组的首标解压器的状态,分组标识符,用于使得分组接收器能够识别分组类型,规范概要信息,包括分组发送器的规范概要信息,循环冗余查看(CRC)字段,用于查看更新的分组的数据完整性;动态信息链,包括动态场境参数。 25. The method of claim 11, wherein said step of controlling refresh packets and compressed packets forming refresh packets, comprising: a context identifier used to identify the state to be used for extracting a packet header decompressor , packet identifier, so that the packet for the receiver to identify the packet type, standardized summary information, including packet transmitter specification profiles, Cyclic Redundancy Check (CRC) fields for data integrity check for updated packet; dynamic information chain, including dynamic context parameters.
26.如权利要求11所述的方法,其中,所述形成刷新分组和压缩控制分组的步骤形成发送器报告分组,包括一个发送器报告分组首标和至少一个报告块。 26. The method of claim 11, wherein said step of controlling refresh packets and compressed packets forming sender report packet, a sender report packet comprises a header block and at least one report.
27.如权利要求26所述的方法,其中,所述发送器报告分组首标包括:分组标识符,用于识别发送器报告分组类型,接收报告计数字段,用于指示包括在发送器报告分组中的报告块的数量;有效发送器标志,用于指示产生报告块的会话参与者是否有效,循环冗余查看(CRC)字段,用于查看发送器报告分组的数据完整性,填充标志,用于指示发送器报告分组在发送器报告分组尾部是否包含附加的填充字段。 27. The method of claim 26, wherein said sender report packet header comprising: a packet identifier for identifying the sender report packet type, receiving the report count field for indicating the sender report packet comprising the number of reports in the block; effective transmitter flag indicating generate conversation participants reported block is valid, a cyclic redundancy Check (CRC) field, for viewing data integrity reports sent packets filled mark with to indicate the sender report packet sender report packet contains an additional padding trailing field. 附加的填充字段不是所述场境参数的一部分,最小有效位(LSB)编码的实时协议(RTP)时间标记,扩展标志,用于指示发送器报告分组还包括扩展字段,最小有效位(LSB)编码的发送器的分组计数字段,用于指示在会话开始和发送器报告分组的产生之间的时间帧中、发送器已经发送的实时协议(RTP)分组的总数,最小有效位(LSB)编码的发送器的八比特组计数字段,用于指示在会话开始和发送器报告分组的产生之间的时间帧中、在实时协议(RTP)数据分组中由发送器发送的有效载荷八比特组的总数,和长度字段,用于指示以最小有效位(LSB)编码格式的发送器报告的长度。 Additional padding field is not part of the field of the environment parameters, the least significant bit (LSB) encoded in real-time protocol (RTP) timestamps, extended flag that indicates the sender report packet also includes the extension field, the least significant bit (LSB) transmitting the encoded packet count field for indicating the total number of frames in the session start time and the sender report packet is generated between the real-time protocol (RTP) transmitter has transmitted the packet, the least significant bit (LSB) encoding The octet count field transmitter, used to indicate the time frame the session start and send reports generated between the packets in real-time protocol (RTP) data packet payload octets transmitted by the transmitter group The total number, and length field, for indicating to the least significant bit (LSB) of the transmitter-length coding format reports.
28.如权利要求11-27之一所述的方法,其中,所述形成刷新分组和压缩控制分组的步骤形成压缩的接收器报告分组,包括一个接收器报告分组首标和至少一个报告块。 28. The method according to one of claims 11-27, wherein said step of forming refresh packets and control packets form a compressed compression receiver report packet, a receiver report packet comprises a header block and at least one report.
29.如权利要求27所述的方法,其中,所述接收器报告分组首标包括:分组标识符,用于识别接收器报告分组类型,接收报告计数字段,用于指示在接收器报告分组中包括的报告块的数量;有效发送器标志,用于指示产生报告块的会话参与者是否有效,循环冗余查看(CRC)字段,用于查看接收器报告分组的数据完整性,填充标志,用于指示压缩的实时控制协议(RTCP)接收器报告分组在压缩的实时控制协议(RTCP)的接收器报告分组尾部是否包含附加的填充字段。 29. The method of claim 27, wherein said receiver report packet header comprising: a packet identifier for identifying the type of the receiver report packet, receiving report count field for indicating the receiver report packet block includes a number of reports; effective transmitter flag indicating generate conversation participants reported block is valid, Cyclic Redundancy Check (CRC) field, for viewing data integrity reports received packet, filling mark with to indicate compression of Real Time Control Protocol (RTCP) packets received reports in real-time compression control protocol (RTCP) receiver report packet tail contains additional padding field. 附加的填充字段不是所述场境参数的一部分,长度字段,用于指示以最小有效位(LSB)编码格式的发送器报告的长度。 Additional padding field is not part of the field, the length of the field throughout the argument, the least significant bit is used to indicate to the sender report (LSB) encoding format length.
30.如权利要求26-29之一所述的方法,其中,所述报告块包括:部分损耗字段,用于指示被容许接收的分组的数量除损耗的分组的数量,最小有效位(LSB)编码的累计损耗字段,用于指示在发送期间损耗的分组的累计数量,最小有效位(LSB)编码的序号周期字段,用于指示所接收的分组的扩展的最高序号的序号周期,最小有效位(LSB)编码的最高序号,用于指示分组的发送器接收的最高序号,最小有效位(LSB)编码的到达之间抖动的字段,用于指示实时协议(RTP)数据分组到达之间时间的统计变化的估计值,最小有效位(LSB)编码的实时协议(RTP)时间标记,和最小有效位(LSB)编码的自最后的发送器报告字段的延迟,用于指示自最后接收的发送器报告的延迟。 30. The method according to one of claims 26-29, wherein said report block comprises: Part loss field for indicating the number of the number of the received packet is allowed in addition to the loss of a packet, the least significant bit (LSB) encoding cumulative loss field for indicating the cumulative number of packet loss during transmission, the least significant bit (LSB) encoding sequence number field period, for indicating the received packet sequence number of the highest number of extended cycles, the least significant bit (LSB) encoding the highest sequence number, used to indicate the highest sequence number of the packet received by the transmitter, the jitter between the least significant bit (LSB) encoding arrival field for indicating real time protocol (RTP) data packets between the time of arrival Statistical estimates change, the least significant bit (LSB) encoded real-time protocol (RTP) time stamp, and the least significant delay bit (LSB) encoded by the sender since the last report fields for the sender received instructions from the last Delayed reports.
31.如权利要求28-30之一所述的方法,其中,所述发送器报告分组和接收器报告分组还包括用于规范概要扩展的字段。 31. The method according to one of claims 28-30, wherein said transmitter and receiver report packet reporting packet further comprising means for expansion of the field profile specification.
32.如权利要求11-31之一所述的方法,其中,所述形成刷新分组和压缩控制分组的步骤形成应用限定的(APP)分组,包括: 32. The method according to one of claims 11-31, wherein said step of forming refresh packets and control packets forming the compressed defined application (APP) packet, comprising:
分组标识符,用于指示应用限定(APP)分组类型,反馈类型字段,用于指示在应用限定(APP)分组中包括的反馈类型,最小有效位(LSB)编码的反馈长度字段,用于指示应用限定(APP)分组的长度,和比特屏蔽(BLP)字段,用于指示损耗的分组。 Packet identifier to indicate the application is defined (APP) packet type, feedback type field, for feedback that indicates the type included in the application is defined (APP) packet, the least significant bit (LSB) encoded feedback length field that indicates Applications Limited (APP) packet length, and bit masking (BLP) field for packet loss indication.
33.一种计算机程序,包括程序代码手段,用于当所述程序在计算机中运行时执行权利要求1-32之中的任一项的所有步骤。 33. A computer program comprising program code means for executing when the program is run on the computer all the steps of any one of claims 1-32 among.
Description  translated from Chinese
控制通信量压缩方法 Control traffic compression method

技术领域 Technical Field

本发明涉及一种用于在媒体数据传输会话中的RTCP通信量的压缩方法。 The present invention relates to a method for data transmission in a media session RTCP traffic compression method. 具体上,所述方法意欲被用于在因特网协议(IP)网络中的实时或近乎实时的数据分组传输中,所述因特网协议(IP)网络使用用于媒体数据传送的实时协议(RTP)和用于控制媒体传送的实时控制协议(RTCP)。 Real-time or near real-time transmission of data packets on the concrete, the method intended to be used in Internet Protocol (IP) networks, the Internet Protocol (IP) network for media data transmission using real-time protocol (RTP), and Real-time control protocol for controlling media transmission (RTCP). 协议RTP或RTCP的每种按照在RFC 1889中给出的规格被分配了可以获得的会话带宽的一部分。 RTP or RTCP protocol according to the specifications of each are given in RFC 1889 is allocated a portion of available session bandwidth.

背景技术 Background

在RFC 1889中定义的实时传送协议(RTP)是提供适合于通过多播或单播网络服务发送实时数据的应用的、端到端的网络传送功能的实际标准。 In the Real-time Transport Protocol (RTP) RFC 1889 is defined to provide adapted to transmit real-time data over multicast or unicast network services applications, the de facto standard end network transport functions. 它被增加了实时控制协议(RTCP),以便使得能够以可调节到大的多播网络的方式监控数据传送的服务质量(QoS),并且提供最小的控制和识别功能。 It was an increase of Real Time Control Protocol (RTCP), so that makes it possible to adjust to large multicast network to monitor the way data transmission quality of service (QoS), and provides a minimum of control and recognition. RTP不处理资源保留,并且不保证实时服务的服务质量(QoS)。 RTP does not handle resource reservation and does not guarantee real-time service quality of service (QoS).

RTP使用两个规则来限制控制通信量:首先,推荐5%的会话带宽被分配到RTCP通信量,并且它被在会话中的所有参与者共享。 RTP uses two rules to limit control traffic: First, Top 5% of the session bandwidth is allocated to the RTCP traffic, and it is shared by all participants in the session. 其次,推荐用于反馈传送的最小报告间隔是5秒。 Secondly, the minimum recommended for transmitting feedback reporting interval is 5 seconds. 在会话中的所有接收器使用它们自己的部分,利用这个5%来计算它们的报告间隔。 All receivers in the session using their own part, the use of the 5% to calculate their reporting interval. RTCP报告间隔T定义了需要满足的、在来自一个来源的两个RTCP数据分组之间的时间间隔。 RTCP report interval T defines the need to meet in time between two RTCP packet data from a source interval. 这个间隔很大程度上依赖于平均的RTCP分组大小。 This interval is largely dependent on the average RTCP packet size.

虽然这些规则使得RTP稳定和可用于大多播组,但是它对于单播或小多播情况不是最佳的。 While these rules make RTP stable and most can be used for the multicast group, but it is for unicast or small multicast situation is not optimal. 对于后者,每个用户更多的反馈是有益的并且是很可能的。 In the latter case, each user feedback is more useful and it is possible. 这个问题已经被确认,并且IETF的音频视频传送工作组正在标准化一个新的RTP规范概要RTP-AVPF。 This problem has been identified, and audio and video transmission IETF working group is a new RTP specification standardized summary RTP-AVPF. 利用这个新的规范概要,所推荐的5秒的最小反馈间隔不适用。 With this new specification summary, the recommended minimum feedback interval of 5 seconds is not applicable. 因此,接收器可以根据当前的会话参数发送一些早期的RTCP分组作为对于分组损耗的反馈。 Thus, the receiver may send some early RTCP packets as a packet loss of feedback for the current session parameters.

基于IP的实时多媒体应用程序由于在实时数据流中单个分组的通常小的有效载荷大小而带来了大的第三层、第四层和上层首标开销。 IP-based real-time multimedia applications due to the generally small size of the payload data in real time in a single packet stream brought a big third layer, the fourth layer and an upper header overhead. 因为无线链路的受限的带宽,首标压缩是移动因特网的必要先决条件,即每当基于IP的移动端器件必须与基于IP的基础设施通信的时候如此。 Because the limited bandwidth of the wireless link, header compression is a necessary prerequisite for a mobile Internet, that whenever IP-based mobile terminal device must be IP-based communications infrastructure when the case. 在RFC 3095中定义的鲁棒首标压缩(ROHC)是由IETF标准化的现代首标压缩方案。 Robust defined in RFC 3095 header compression (ROHC) is standardized by the IETF modern header compression scheme. 它提供了复杂的框架,使得可以根据不同的链路条件细调压缩效率,还是压缩抵制链接差错的鲁棒性。 It provides a complex framework, making it possible to link different conditions according to fine-tune the compression efficiency and compression resistance link error robustness. 所述协议通过保持在第一跳跃或最后跳跃无线链路的终端的状态和通过消除分组首标的冗余以及通过以有效的方式编码信息而工作。 The protocol by maintaining the packet header and by eliminating redundancy in an efficient manner and by coding information in the first or the last skip skip terminal radio link state. 在发送端(或终端)的压缩器和在接收端的解压器的状态也被称为“场境”。 At the transmitting end (or terminal) compressor and decompressor at the receiving end of state also called "context." 所述场境包含来自分组流中的在前的首标的相关信息,诸如静态字段(staticfield)和用于压缩和解压的可能的参考值。 The context contains the header preceding the relevant information from the packet stream, such as static fields (staticfield) and for the compression and decompression of possible reference value. 描述分组流的附加信息也可以是场境的一部分,例如关于IP标识符字段如何改变的信息和在序号或时间标记中的典型的分组间增加。 Additional information describing the packet stream may also be part of the context, for example, on how IP identifier field to change between information and typical packet sequence number or time stamp of the increase.

虽然在RFC 3095中存在四个用于无压缩、RTP/UDP/IP、UDP/IP和ESP的规范概要以及仅仅用于IP的规范概要草案,但是没有关于如何可以利用压缩来处理RTCP分组和首标的规范。 Although there are four for no compression in RFC 3095, the specification summary RTP / UDP / IP, UDP / IP and ESP and the only draft specification for IP's outline, but not about how you can use compression to handle RTCP packets and the first target specification.

作为使用RTP/RTCP的关键应用之一的在移动因特网上的视频流正在获得动力。 As video streams on mobile Internet use RTP / RTCP, one of the key applications is gaining momentum. 但是,在无线链路中的有损耗的行为和长的往返时间使得使用这种应用足够复杂。 However, in the wireless link lossy behavior and long round-trip time complex enough to make use of this application. 一个原因是诸如MPEG-4的帧间视频压缩算法,它使用在帧之间的时间相关性以实现极高的压缩增益,但是它们也受到公知的差错传播的影响,因为参考帧的差错传播到所有的从属的不同帧。 One reason is that MPEG-4, such as inter-frame video compression algorithm, which uses a time correlation between frames to achieve high compression gain, but they are also known to be affected by error propagation, since the error propagation to the reference frame All dependent different frames.

发明内容 DISCLOSURE

本发明的目的在于优化RTCP通信量的带宽效率和降低RTCCP报告/反馈间隔。 The purpose of the present invention is to optimize bandwidth efficiency RTCP traffic and reduce RTCCP reports / feedback interval. 对于在一个会话中的所有参与者之间的和对于双向操作的、用于RTCP的共享会话带宽部分是有限的。 For all participants in a session between the two-way operation, for sharing part of RTCP session bandwidth is limited. 在稀少的和昂贵的无线链路中,频谱效率是关键的。 In the scarce and expensive radio link, spectral efficiency is critical. 因此,如何使用这个有限的带宽效率意味着应用经由无线链路使用RTP的需要。 Therefore, how to use the limited bandwidth efficiency means that the application needs to use the RTP via a wireless link. 本发明旨在最大化地将用于RTCP通信量的带宽效率用于这些用途中而不超过可以获得的RTCP带宽部分。 The present invention is intended to maximize the bandwidth efficiency for RTCP traffic for these applications without exceeding the portion of the bandwidth available RTCP. RTCP报告间隔是对于同一会话内在来自同一接收器的两个连续报告之间的时段。 RTCP report interval between two consecutive reports for the same session from the same receiver intrinsic period. 它受到两个方面的延迟时间的影响。 It is influenced by the delay time of two aspects. RTCP报告延迟时间是在接收器检测到分组损耗和发送报告/反馈之间的时段、以及由于链路的往返时间(RTT)的延迟时间。 RTCP report delay time in the receiver detects packet loss and send a report / feedback between periods, as well as due to the round trip time link (RTT) delay. 虽然后者难于避免,但是所述报告延迟时间可以被用于优化。 While the latter is difficult to avoid, but the report can be used to optimize the time delay. 用于计算被定义为报告间隔T的这个延迟时间的公式可以被表达如下:T=avg_rtcp_size*n/rtcp_bw对于其中在单播和小多播会话中使用RTP/RTCP的典型情况,参与者的数量n相对固定。 For calculating the formula is defined as the reporting interval delay time T can be expressed as follows: T = avg_rtcp_size * n / rtcp_bw wherein for unicast and multicast sessions using small typically RTP / RTCP, the number of participants n relatively fixed. 以降低参与者的数量为代价来降低延迟时间是不期望的。 To reduce the number of participants to reduce the cost of the delay time is not desirable. 这仅仅对于RTCP带宽部分rtcp_bw和平均RTCP分组大小avg_rtcp_size留下了实验性空间。 This is only part of the bandwidth for RTCP rtcp_bw and average RTCP packet size avg_rtcp_size left an experimental space. 如上所述,所述手段通过提高RTCP带宽部分来降低报告间隔,但是它们修改RTCP带宽部分的规则,所述RTCP带宽部分最大是总的可获得的会话带宽的5%。 As described above, by means of improving the RTCP reporting interval to reduce the portion of the bandwidth, RTCP bandwidth, they modify the rules portion, the largest part of the RTCP bandwidth is 5% of the total available bandwidth of the session. 而且这些手段可能遇到兼容性的问题。 And these means may encounter compatibility problems.

根据上述讨论,改进、即降低RTCP协议的报告间隔的唯一的可能性是降低平均RTCP分组大小avg_rtcp_size。 Based on the above discussion, the improvements that reduce the RTCP reporting interval of only possibility is to reduce the average RTCP packet size avg_rtcp_size. 由于RTCP报告间隔T与平均RTCP分组大小成正比,因此压缩RTCP分组可以将分组大小降低到高达原始RTCP分组的10%的水平。 Since RTCP reporting interval T is proportional to the average RTCP packet size, thus compressing the packet RTCP packet size can be reduced to a level as high as 10% of the original packet of RTCP. 这将导致控制协议分组的较小的平均分组大小,因此导致很小的报告间隔T。 This will result in smaller average packet size control protocol packets, thus resulting in a small reporting interval T.

据此,本发明提供了一种用于对控制RTP媒体数据传输会话的RTCP通信量的压缩方法。 Accordingly, the present invention provides a method for controlling RTP media data transmission session RTCP traffic compression method. 在此所述的压缩原理基本上可以适用于在有线/固定的网络或无线/移动中的、用于实时和近乎实时的媒体传送的任何种类的使用RTP的链路。 In this compression principle can be applied to substantially the wired / wireless network or a fixed / moving, real-time and near real-time for any kind of use of RTP media transmission link.

在按照本发明的数据传输中的终端保存压缩器和解压器的内容状态。 In the Save the compressor and decompressor according to the present invention, the data transmission terminal content state. 由于在场境中保持的结构,因此在解压器修理和恢复不同步的场境是可能的。 Due to the presence of territory held in the structure, repair and restore the decompressor context of sync is possible. 而且,有可能动态地限定分组格式和压缩器的和解压器的场境。 Moreover, it is possible to dynamically define packet formats and compressor and decompressor context.

所述压缩方法通过初始向接收终端发送场境参数而初始化控制通信流的场境。 The compression method to the receiving terminal sends an initial context initialization parameters to control traffic flow context. 如果必要的话,则在会话期间利用较小尺寸的分组(压缩的控制分组)更新场境。 If necessary, during a session using a smaller-sized packets (compressed control packets) to update the context. 后面的分组用于执行部分场境更新的情况下。 Packet back to the case of partial execution context updates. 也有可能利用初始化分组定期地更新场境。 It may also use the initialization packet context is updated regularly. 可以将场境参数分类为静态和动态参数。 The context parameter can be classified as static and dynamic parameters. 静态场境参数是先验已知的参数或在会话期间不改变的参数。 Static context parameters are known a priori parameters or parameter does not change during the session. 作为在会话期间改变的参数的动态场境参数在新定义的分组或压缩的控制分组中被发送到接收端。 As during a session to change the dynamic context parameters parameters are transmitted to the receiver in a new definition of control packet or a compressed packet.

因为不同的可能场境参数(也包括标准RTCP数据分组的所有字段)是已知的,因此它们可以便利地被划分为静态和动态场境参数。 Because different possible context parameters (including all standard RTCP packet fields) are known, so that they can be conveniently divided into static and dynamic context parameters. 根据这个分类,可以执行首标和数据压缩。 According to this classification, you can perform a header and data compression.

为了进一步降低通信量开销,因此先验已知的场境参数可以被省略,并且因此不被发送,虽然有可能通过利用在此说明的压缩和解压机制来执行这些分组的压缩。 In order to further reduce the amount of communication overhead, a priori known context parameters can be omitted, and thus is not transmitted, although it is possible through the use of compression and decompression mechanisms described herein performs these compressed packets.

为了初始化一个会话,至少一个包括这些场境参数的初始化分组被发送到接收节点。 To initiate a session, the at least one packet including the context initialization parameters are transmitted to the receiving node. 由于被包括的参数包括静态信息,这些信息仅仅需要被发送一次。 Due to the included parameters include static information that needs to be sent only once. 因此,可以大大地降低要被发送的总的通信量。 Therefore, it can greatly reduce the total amount of traffic to be transmitted. 动态场境参数例如在控制协议特定分组(压缩的控制分组)中被发送。 Dynamic context parameters such as being sent in the control protocol-specific packets (compressed control packets) in. 刷新的分组允许分组源可以在接收节点更新场境信息。 Refresh packet allows the packet to source information at the receiving node can update the context. 控制分组主要对应于从标准RTCP协议已知的那些分组。 Control packet grouping mainly correspond to those from the standard RTCP protocol known.

相反,压缩的控制分组在它们的分组结构上被改变,以便可以大大地减少它们的内容的大小(以比特为单位)并且引入诸如初始化分组和刷新分组的新的分组。 In contrast, the control packet is compressed changed in their packet structure, so that it can greatly reduce the size of their content (in bits) and initialize the packet, such as the introduction of a new packet and refresh packet. 因此,与标准的RTCP协议相比较,可以大大地降低控制协议的控制分组的总的平均分组大小。 Therefore, compared with the standard RTCP protocol, it can greatly reduce the overall average packet size control packet control protocol.

由于RTCP源描述分组和RTCP BYE分组的内容在会话期间不频繁地改变或不经常发生,因此在所公开的方法中将不压缩在控制协议中的对应的源描述分组和BYE分组,虽然有可能利用在此所述的压缩和解压机制执行这些分组的压缩。 Since RTCP source description packets and RTCP BYE packet content changes or infrequent infrequently during the session, so the method will be disclosed without compression source description packets and BYE packets corresponding control protocol, although there may be use this compression and decompression of the mechanisms to implement these compressed packets. 所述两种分组具有与对应的RTCP分组类似的格式。 It said two corresponding RTCP packet having a packet format similar.

在分类场境参数之后,在发送它们之前,从这些静态场境参数和——如果需要的话——从动态场境参数的初始化值形成至少一个初始化分组。 After classification context parameter, before sending them from these static context parameters and - if necessary - the formation of the initial value of dynamic context parameters from at least one initialization packet. 在发送相同的从前的动态场境参数形成刷新的分组。 Sending the same dynamic context parameters previously refresh packet form.

为了达到最大的压缩水平,动态场境参数被进一步分类为偶尔改变的场境参数、随机字符的场境参数、类似计数器(counter-like)的场境参数、经常改变的场境参数和规则地改变固定的增量的场境参数。 In order to achieve the maximum level of compression, dynamic context parameters are further classified as occasional change of context parameters, context parameters of random characters, similar to the counter (counter-like) in the context parameters, frequently changing context parameters and rules to changing context parameters fixed increment. 依赖于动态场境参数的种类,通过编码来压缩参数以便在将它们并入控制数据分组之前降低它们的大小。 It depends on the kind of dynamic context parameters, to compress the encoding parameters in order to be incorporated into the control data packets before they reduce their size. 特殊地,可以利用最低有效位(LSB)编码来编码类似计数器的场境参数、经常改变的场境参数和规则地改变固定的增量的场境参数。 Particularly, can use the least significant bit (LSB) encoding to encoding context parameters similar counter, incremental changes fixed context parameters change frequently and regularly context parameters.

利用最低有效位(LSB)编码来使用被编码的字段值的K个最低有效位而非原始的字段值,其中K是正整数。 The use of the least significant bit (LSB) encoding the K least significant bits of the field value instead of the original field value encoded, where K is a positive integer. 在接收K个比特之后,解压所压缩的数据分组的、在分组接收端的解压器使用作为基准的在前接收的值来得到原始值。 After receiving K bits, extracting the compressed data packets at the receiving end of the packet decompressor using as a reference value to obtain the first received the original value.

附图说明 Brief Description

图1示出了由压缩方法使用来初始化会话的初始化分组的分组格式,图2示出了由压缩方法使用来更新动态场境参数的刷新分组的分组格式,图3示出了由压缩方法使用的发送器报告分组的分组格式。 Figure 1 shows a session to initialize compression method used by the packet format initialization packet, Figure 2 shows a packet format used by the compression method to update the dynamic context parameters refresh packet Figure 3 illustrates the use of the compression method The sender report packet format packet.

图4示出了由压缩方法使用的接收器报告分组的分组格式。 Figure 4 illustrates a packet format used by the compression method of the receiver report packet.

图5示出了由压缩方法使用的应用定义的分组的分组格式。 Figure 5 illustrates a packet format used by the compression method of application-defined packet.

具体实施方式 DETAILED DESCRIPTION

为了降低报告间隔T,减小控制协议的数据分组的平均分组大小。 In order to reduce the reporting interval T, to reduce the average packet size of the packet data control protocol. 在RFC1889中定义的标准RTCP协议使用下面的分组以控制在实时或近乎实时的环境中的媒体数据传输:发送器报告,用于在媒体数据传输会话中发送和接收来自作为有效发送器的参与者的静态内容;接收器报告,用于从不是有效发送器的参与者接收静态内容;源描述项,用于说明发送源;BYE分组,用于指示前一个参与者的参与的结束;应用限定的(APP)分组,用于发送应用特定的数据。 Use the following grouping criteria defined in RFC1889 RTCP protocol to control the real-time or near real-time data transmission media environment: report transmitter for transmitting and receiving data at the media session from the participants as an effective transmitter static content; a receiver report for receiving static content from transmitter is not a valid participant; source description entry for explaining the transmission source; BYE packet for instructions before the end of the participation of a participant; Application defined (APP) packet for sending application-specific data.

为了降低上述的数据分组的大小,首先分析在分组结构的字段。 In order to reduce the size of said data packet, the first packet structure analysis in the field. 一般,在RTCP分组中的所有字段可以被分类为:静态场境参数,它们是被期望在分组流(会话)的使用期限内恒定的字段;动态场境参数,它们是被期望在限定值组或范围内或以一些方式、例如随机地或一些其他的方式变化的字段。 In general, all the fields in RTCP packets can be classified as: a static context parameter, which is expected within a packet stream (session) the period of use constant field; dynamic context parameters, which are to be expected in the limit value set or within the range, or in some way, e.g., random changes in some other way or fields.

动态场境参数(动态RTCP分组字段)可以进一步被分类如下:偶尔改变的场境参数、随机字符的场境参数、类似计数器(counter-like)的场境参数、经常改变的场境参数和规则地改变固定的增量的场境参数。 Dynamic context parameters (dynamic RTCP packet fields) can be further classified as follows: occasionally changed context parameters, context parameters of random characters, similar to the counter (counter-like) in the context parameters, frequently changing context parameters and rules changing context parameters fixed increment.

偶尔改变的场境参数是那些偶尔改变但在有限个数量的分组之后恢复它们的原始值的字段。 Occasionally changing context parameters that change occasionally but recover their original value after a finite number of packets field. 对于在标准RTCP分组内的场境参数和字段,那些值或字段是:接收报告计数(RC),它指示在分组中报告块的数量;源计数(SC)字段,它们指示在源描述分组中的同步源或贡献源的数量,或者识别在BYE分组(by packet)中的同步源或贡献源的数量;有效载荷类型(PT)字段,它识别独立的分组类型;源描述(SDES)项目,包括描述分组源属性的信息;子类型字段,在应用限定的(APP)分组内,允许在唯一名称下定义一组应用限定的(APP)分组。 For context parameters and fields in the standard RTCP packets, those values or field is: reception report count (RC), which indicates the number of reports in the packet block; source count (SC) fields, which indicates the source description packet The synchronization source or amount of contribution sources, or to identify the BYE packet (by packet) synchronization source or sources of the contribution amount; payload type (PT) field that identifies a separate packet type; source description (SDES) project, information packet including a description of the source property; subtype field in the application defined (APP) packets, allows you to define a set of application defined (APP) grouped under a unique name.

那些偶尔改变的场境参数可以被初始发送来用于初始化,但是如果它们改变也应当存在一种发送或更新那些字段的方式。 Occasionally changing context parameters that may be used to initialize the initial transmission, but if they should be changed or there is a transmission mode update those fields. 因此,对于压缩的数据分组的所建议的控制协议引入了刷新分组,它被用于发送场境参数来用于更新的目的。 Therefore, the proposed compressed data packet introduces a refresh control protocol packet, which is used to transmit context parameters for update purposes. 下面进一步说明分组的使用和结构。 Further described below use and structure of the packet.

经常改变的场境参数包括那些通常或者恒定或者具有可以从一些其他的字段导出的值、但是经常脱离这种行为的参数。 Frequently changing context parameters, including those usually either constant or can be derived from having some other field values, but often out of the parameters of this behavior. 因此,必须有一种有效的方式来更新在接收器或发送器端更新所述经常改变的场境参数。 Therefore, there must be an effective way to update the receiver or transmitter end updating the frequently changing context parameters. 所述的刷新分组可以用于这样的情况下,或者在新定义的控制分组中各个字段原样被发送。 The refresh packet may be used under such circumstances, or for each field in the control packet is transmitted as a new definition.

必须经常更新的字段包括:RTP时间标记,它们指示自最后接收到的发送器报告起的延迟(自最后的发送器报告起的延迟)的字段;最后的发送器报告的时间标记;到达之间的抖动字段,它们指示RTP数据分组到达之间的时间的这些统计变化的估计值;RTCP分组的长度字段。 The fields must be updated frequently: RTP timestamp, they received instructions from the final report from the transmitter (delay since last sender report) field; time of the last report transmitter tag; reach between Jitter field, which indicates the estimated values of these RTP data packets between the arrival time of the statistical variations of; RTCP packet length field.

另一个类别的动态场境参数是随机字符的分组。 Dynamic context parameters for another class is a grouping of random characters. 那些参数的示例是在由J.Ott et al.in“Extended RTP Profile for RTCP-based Feedback(RTP/AVPF)”,Internet Draft,Oct.2002(J.Ott等在“用于基于RTCP的反馈(RTP/AVPF)的扩展RTP规范概要”,因特网草案,2002年10月)所指定的RTCPAPP分组的比特映象屏蔽(mask)(BLP)中的RTCP部分损耗。 Examples of those parameters is in the "for based on feedback from the J.Ott et al.in" Extended RTP Profile for RTCP-based Feedback (RTP / AVPF) ", Internet Draft, Oct.2002 (J.Ott et RTCP (the RTP / AVPF) extended RTP specification summary "Internet Draft, October 2002) bit maps designated RTCPAPP grouping shield (mask) (BLP) in the RTCP part of the loss. 由于那些字段完全随机,因此它们被原样包括在所有的压缩分组首标中。 Since those fields completely random, so they are as included in all compressed packet header.

下一个类别的动态场境参数是类似计数器的场境参数。 Dynamic context parameters is a category similar context parameter counter. 那些参数是像计数器一样作为的字段,并且在所有的RTCP分组的不同计数值之间具有固定的增量。 Those arguments are the same as the counter as a field, and a fixed increment between different count values of all RTCP packets. 对于那些字段的传输编码的唯一要求是在发送器端的压缩器和在接收器端的解压器之间的分组损耗必须是可以容许的。 The only requirement for the transfer of those fields are coded at the transmitter end of the compressor and the packet loss between the receiver end of the decompression device must be permissible. 如果那些字段的几个存在,则所有那些字段也可以一起被通信。 If several of those fields exist, all of those fields may also be communicated together. 这样的参数也可以用于解释经常改变的场境参数的值。 Such parameters may also be used to explain the frequently changing context parameters.

那些字段的示例是:RTP序号;扩展的所接收的最高序号字段;发送器的分组计数,指示在会话开始和包括发送器分组计数的分组的产生之间的时间帧中、发送器已经发送的RTP分组的总数;分组发送器的八比特组计数,它指示在RTP分组中由发送器在会话开始和包括发送器分组计数的分组的产生之间的时间帧中发送的有效载荷八比特组的总数;损耗分组的累计数量,它指示在发送期间损耗的分组的累计数量。 Examples of those fields is: RTP sequence number; extended highest sequence number received field; sender packet count, indicating the session start and the time frame including a packet sender packet count generated between the transmitter has been sent The total number of RTP packets; octet of the packet transmission count, which indicates the RTP packet by the sender in the payload octet session start and time frame includes a packet sender packet count generated between the transmitted group Total; total number of packet loss, which indicates the cumulative number of packet loss during transmission.

最后一种动态场境参数包括那些以固定的增量规则变化的场境参数。 Finally, a dynamic context parameters including those context parameters in fixed increments rule change. 在连续的分组中,那些字段通常提高固定的增量。 In successive packets, those fields are generally improving in fixed increments. 因此,那些字段彼此相关联。 Therefore, those fields associated with each other. 在这种场境中,可取的是使用初始化分组来启动字段的值,然后通过发送它们的增量来更新字段。 In this context, and it is preferable to use an initialization packet to start value of the field, and then send them to update the field by increments.

以固定的增量规则变化的场境参数的一个示例是RTP时间标记。 An example of context parameters in fixed increments rule change is RTP timestamp.

而且,可以确定在静态和动态场境参数之外的第三种类型。 Moreover, it can determine the static and dynamic outside the context of the third type parameter. 在标准RTCP分组中的所谓的公知或先验已知字段可以或者在初始化期间被发送或者被省略。 In a so-called standard RTCP packet or known a priori known field may be omitted or be sent or during initialization. 先验已知场境参数的一个示例是RTCP版本字段。 An example of a priori known context parameter is RTCP version field.

可以例如通过用于分组和首标压缩的引用表来执行场境参数的所述分类。 Can be performed, for example the context parameters for packet and header by compressing reference table classification. 也有可能在所建议的新RTCP压缩方法中动态地分类场境参数。 But also in the proposed new RTCP compression method to dynamically classify context parameters possible.

在已经分类了场境参数之后,编码动态场境参数的一部分以降低它们的大小。 After having classified the context parameter coding part of the dynamic context parameters in order to reduce their size. 具体上,类似计数器的场境参数、经常改变的场境参数和以固定的增量规则变化的场境参数被最低有效位(LSB)编码,以便可以充分地减少原始字段的大小(以比特为单位)。 Specifically, the context parameters similar to the counter, constantly changing context parameters and rules of changes in fixed increments of context parameter is the least significant bit (LSB) encoded so that it can substantially reduce the size of the original field (in bits units).

在分类场境参数之后,形成控制协议的分组。 After classification context parameter, forming a packet control protocol. 为了初始化一个会话,形成和发送一个初始化分组。 To initiate a conversation, form and send an initial packet. 图1示出了一个初始化分组的分组格式。 Figure 1 shows a packet format of an initialization packet. 所述分组包括:静态场境参数,诸如填充标志、发送器和源的同步源;在分组的“静态链”字段中的贡献源字段。 The packet includes: a static context parameters, such as stuffing flags, the transmitter and the synchronization source source; contributing source field in the "static chain" field in the packet. 所述“静态链”字段因此是象分组的“动态链”字段一样在长度上可变的。 The "static chain" field is thus as "dynamic chain" field in the same grouping variable length. 而且,在初始化分组中并入了源计数和接收报告计数、有效载荷类型标识、一个或多个SDES项和应用限定(APP)的分组的子类型字段。 Also, incorporated in the initialization packet source and receive reports count count, payload type identification, subtype field SDES items and one or more limited application (APP) packet.

偶尔变化的场境参数也可以被集成到使用它们的初始值而形成的初始化分组中。 Occasionally changing context parameters may also be integrated into the initialized packet using their initial values formed. 这些偶尔变化的场境参数的初始化值位于初始化分组的“动态链”字段内。 The initial value of occasionally changing context parameters is available in initialization packet "dynamic chain" field inside.

一旦初始化了偶尔变化的场境参数,则可以通过刷新分组在下面更新它们。 Once initialized occasionally changing context parameters, you can update them in the following refresh packets.

详细而言,被压缩的初始化分组包括:场境ID(CID,“Add-CID octet”),它识别在分组接收端要使用的解压器的状态,以便在分组的开头解压初始化分组;分组标识符(“1111110D”),用于使能分组接收器来识别分组类型;发送器的规范概要的规范概要信息(“Profile”);循环冗余查看(“CRC”)字段,用于查看初始化分组的数据完整性;静态信息链(“Static Chain”),包括静态场境参数;最后,动态信息链(“动态链”),包括必须被初始化一次的动态场境参数。 Specifically, the compressed initialization packet includes: context ID (CID, "Add-CID octet"), which identifies the receiving end of decompression in the packet to be used in the state, so that at the beginning of the packet decompression initialization packet; packet identification character ("1111110D"), is used to enable the receiver to identify the packet type of packet; specification summary transmitter specification profiles ("Profile"); Cyclic Redundancy Check ("CRC") field, used to view the initialization packet data integrity; static information chain ("Static Chain"), including static context parameters; Finally, dynamic information chain ("Dynamic Chain"), including the need to be initialized once dynamic context parameters. 后者对应于上述的偶尔改变的场境参数,诸如源计数、接收报告计数、RTCP有效载荷类型、SDES项和应用限定(APP)的分组的子类型字段。 The latter corresponding to the occasional change of context parameters, such as source count, receiving reports count, RTCP payload subtype field type, SDES and applications of defined (APP) packet.

图2示出了刷新分组的分组格式。 Figure 2 shows a packet format of refresh packets. 由于初始化分组的后面所述的字段是动态的,因此引入新的刷新分组来更新那些字段。 Since the back of the initialization packet fields are dynamic, so the introduction of the new refresh packets to update those fields. 详细而言,刷新分组包括场境ID(CID,“Add-CID octet”),它识别在分组接收端要使用的首标解压器的状态,以便解压刷新分组;分组标识符(“11111000”);分组发送器的规范概要信息(“Profile”);循环冗余查看(“CRC”)字段,用于查看刷新分组的数据完整性;动态信息链(“动态链”),包括必须被更新的动态场境参数。 Specifically, the refresh packet includes context ID (CID, "Add-CID octet"), which identifies the packet header decompressor receiving end of the state to be used in order to extract the refresh packets; packet identifier ("11111000") ; packet transmitter specification profiles ("Profile"); Cyclic Redundancy Check ("CRC") field, for viewing data integrity refresh packet; dynamic information chain ("Dynamic Chain"), including the need to be updated Dynamic context parameters.

另外,在使用大场境标识符(CID)的情况下,初始化分组和刷新分组可以包括跟随分组标识符的多达两个附加字节。 Further, in the case of using a large context identifier (CID), the initialization and refresh packet follow up packet may comprise two additional bytes of the packet identifier.

图3和4示出了发送器和接收器报告分组的压缩版本,图5示出了新的压缩应用限定(APP)的分组。 Figures 3 and 4 shows a compressed version of the transmitter and the receiver report packet, Figure 5 shows a new compression applications defined (APP) packets.

源描述分组和BYE分组对应于在RFC 1889中所提出的标准分组格式。 Source description packets and BYE packet corresponds to a standard packet format in RFC 1889 as proposed. 这是因为那些分组在一个会话中是很少出现的,因此它们的压缩将不会大大降低RTCP分组的平均分组大小。 This is because those packets in a session is rare, so they will not be compressed significantly reduce the average packet size RTCP packet.

如图3所示的发送器报告分组包括一个分组首标和至少一个报告块。 As shown in Figure 3 reports sent packet includes a packet header and at least one report block. 所述报告块可以后随规范概要特定的扩展。 It can block with the specifications after the report outline specific extensions. 在规范概要特定的扩展中,落入上述的类别之一中的所有字段也可以使用最小有效位编码来被压缩。 In the specification outline specific extension, all fields fall into one of the above categories may also be used in the least significant bit encoding is compressed. 因此,也可以减小扩展字段的大小,导致较小的平均分组大小。 Therefore, the extension field can be reduced in size, resulting in a smaller average packet size.

在附图中的缩写“LSB”表示“最小有效位”,并且指示各个字段被最小有效位编码。 In the drawings, the abbreviations "LSB" represents "least significant bit", and indicates the least significant bit of each field are coded.

发送器报告分组的首标包括分组标识符(“111”),用于识别发送器报告分组类型。 Sender report packet header includes a packet identifier ("111"), it is used to identify the sender report packet type. 而且,接收报告计数字段(“RC”)指示在压缩的发送器报告分组中包括的报告块的数量。 Further, receiving report count field ("RC") indicates the number of blocks in the compressed report sender report packet included. 有效发送器标志(“S”)指示形成报告块的参与者是有效的发送器[Gu1]与否[FH2]。 Transmitter valid flag ("S") indicates the formation of the participants report block is valid transmitter [Gu1] or not [FH2]. 循环冗余查看(“CRC”)字段用于查看被压缩的发送器报告分组的数据完整性。 Cyclic Redundancy Check ("CRC") field is used to view the compressed sender report packet data integrity. 填充标志或比特“P”指示是否发送器报告分组包括在分组尾部的附加的填充字段。 Stuffing flags or bits "P" indicates whether the sender report packet includes additional padding field at the tail of the packet. 附加的填充字段不是实际的场境参数的一部分。 Additional padding field is not part of the actual context parameters.

最小有效位(LSB)编码的RTP时间标记(“LSB Scaled RTP Timestamp”)还被包括在首标内。 Least significant bit (LSB) encoded RTP timestamp ("LSB Scaled RTP Timestamp") is also included in the header. 一个扩展标志(“X”)指示是否分组在分组尾部的特殊扩展字段中包括规范概要特定扩展。 An extension flag ("X") indicates whether the packet includes specification summary specific extensions in a special extension field in the packet tail.

为了进一步减少发送器报告分组大小,发送器的分组计数字段也被最小有效位(LSB)编码。 To further reduce the sender report packet size, packet count field transmitter is also the least significant bit (LSB) encoding. 在发送器报告分组的首标中的发送器的分组计数字段(“LSB Sender's Packet Count”)指示在媒体数据传输会话的开始和相应的发送器报告分组的产生之间的时间帧中发送器已经发送的RTP分组的总数。 Sender report packet in the header of the packet count field transmitter ("LSB Sender's Packet Count") indicates the start and the corresponding media data transmission session sender report packet generation time frame between the transmitter has The total number of RTP packet sent.

而且,发送器报告分组的首标包括用于发送器八比特组计数的字段,它指示在RTP数据分组中由发送器在会话的开始和发送器报告分组的产生之间的时间帧中发送的有效载荷八比特组的总数。 Moreover, the sender report packet header comprises a transmitter octet count field, which indicates the transmitter at the beginning of the session and report time frame between packets generated by the transmitter to send data packets in RTP total payload octet group. 再一次,所述字段被最小有效位(LSB)编码以降低分组首标的大小。 Again, the field is the least significant bit (LSB) encoded to reduce the size of the packet header. 在附图中,这个字段被分为两个部分(“LSB Sender Octet Count Partl”和“LSB SOC P2”),分组的第五个字节和分组的第六个字节的前面五个比特包含发送器的八比特组计数。 In the drawings, the field is divided into two parts ("LSB Sender Octet Count Partl" and "LSB SOC P2"), in front of the sixth byte of the packet and the packet fifth byte contains five bits octet count of the transmitter.

分组首标还包括一个指示发送器的报告长度的字段(“LSB Len SR”)。 Packet header also includes a field indicating the length of reports of a transmitter ("LSB Len SR"). 这个字段也被最小有效位(LSB)编码。 This field is also the least significant bit (LSB) encoding. 在分组的第六个字节之后,首标的尾部被标注了“+=+=+=”行。 After the sixth byte of the packet, header tail is marked with "+ = + = + =" line.

在发送器报告分组中的所述至少一个报告块包括下列字段:部分损耗字段(“fraction lost”)指示被容许接收的分组的数量除损耗的分组的数量,累计损耗字段(“cummu.loss”)指示在传输期间损耗的分组的累计数量。 The sender report packet in at least one report block contains the following fields: Part loss field ("fraction lost") indicates the number of the received packet is allowed in addition to the number of packet loss, the cumulative loss field ("cummu.loss" ) indicates that during the transmission loss of the cumulative number of packets. 为了降低大小,累计损耗字段与报告块的剩余字段一样被最小有效位(LSB)编码。 In order to reduce the size of the accumulated loss of the remaining fields are the same report block is the least significant bit (LSB) encoding. 序号循环字段(“LSB SN Cycle”)指示所接收的分组的扩展最高序号的序号周期。 Numbering field ("LSB SN Cycle") indicates that the received packet extended highest sequence number cycle. 最高序号字段(”LSB Highest SN”)指示由发送器报告分组的发送器接收的最高序号。 Highest number field ("LSB Highest SN") indicates the highest number by the sender report packet transmitter to receive. 到达间抖动字段(“intera.jitter”)包括RTP数据分组到达之间的时间的统计变化的估计值。 Estimates between arrival jitter field ("intera.jitter") include RTP packet statistics vary between the time of arrival. 在报告块中还包括RTP时间标记字段(“LSB TS last SR”),它指示自已经发送了最后一个发送器报告之后的时间。 Also included in the report block RTP timestamp field ("LSB TS last SR"), which indicates that the time has been transmitted from a transmitter final report later. 也包括用于指示自最后一个压缩RTCP发送器报告起的延迟的字段。 Also includes means for indicating a delay since the last RTCP sender report field from the compression.

在报告块中除了部分损耗字段之外的所有字段使用最小有效位(LSB)编码被编码。 All fields except section loss field using the least significant bit (LSB) encoding block is encoded in the report.

一个单个报告块是四个字节长(附图中示出的字节七-十)。 A single report block is four bytes long (shown in the drawings byte seven - X). 如图所示,单个发送器报告可以包括多个报告块。 As shown, a single sender report blocks may include a plurality of reports.

除了发送器报告分组之外,还在下面提出RTCP接收器报告分组的压缩版本。 In addition to sending a packet outside reports, also put forward RTCP packets received reports of a compressed version below. 图4所示的接收器报告分组包括首标(字节一到三)和与上述的报告块类似的至少一个报告块。 The receiver shown in FIG. 4 report packet includes a header (byte one to three) and the above-mentioned report blocks similar to at least one report block. 压缩的接收器报告分组以及压缩的发送器报告分组也可以包括在它们的尾部的规范概要特定扩展,由在分组首标中的扩展标志(“X”)指示。 Reports received compressed packets and compressed packets sent reports may also include in their tails specification outline specific extensions by the packet header extension flag ("X") indication.

接收器报告分组的首标包括一个分组标识符(“111”),用于识别接收器报告分组类型,因此接收端可以识别接收器报告的压缩版本。 The receiver report packet header includes a packet identifier ("111"), used to identify the type of the receiver report packet, and therefore the receiving side can recognize the compressed version of the receiver report. 接收计数字段(“RC”)指示在接收器报告分组中包括的报告块的数量。 Reception count field ("RC") indicates the number of reports in the receiver report blocks included in the packets. 作为发送器报告分组,[Gu3]接收器报告分组可以包括跟随各个分组首标的几个报告块。 As a sender report packet, [Gu3] reports received packet may include follow several reports each packet header block.

有效发送器指示标志(“S”)指示产生各个报告块的会话参与者的状态。 Effective transmitter indicator ("S"), indicate that the state of each session participants report blocks. 而且,循环冗余查看字段被包括以验证数据完整性。 Further, a cyclic redundancy check for the field are included to verify data integrity. 填充标志(“P”)指示是否接收器报告分组在接收器报告分组的尾部包含附加的填充字段。 Fill mark ("P") indicating whether a receiver report packet at the end of the receiver report packet contains an additional padding field. 所述附加填充字段不是实际场境参数的一部分。 The additional padding field is not part of the actual context parameters. 最后,在接收器报告分组的首标中包括长度字段(“LSB Length RR”)以指示最小有效位(LSB)编码格式的压缩RTCP接收器报告的长度。 Finally, including the length field ("LSB Length RR") receiver report packet header to indicate that the least significant bit (LSB) length coding compression format of RTCP receiver reports.

最后,图5示出了应用限定(APP)分组格式。 Finally, Figure 5 shows the application is defined (APP) packet format. 所提出的压缩的应用限定(APP)分组格式的用户仅仅适用于使用由Ott等提出的增进的RTCP反馈的情况下。 Compression of the proposed application defined (APP) packet format of the user applies only to the case of using RTCP feedback from the promotion proposed by Ott. 因此,这个分组规格比一般的方式更为应用特殊——如其名称所示。 Thus, the packet size is more than apply a special way - as the name implies.

压缩的应用限定(APP)分组格式包括分组标识符(“111”),用于标识压缩的应用限定(APP)分组。 Limited compression application (APP) packet format includes a packet identifier ("111") for identifying compression applications defined (APP) packets. 反馈类型字段(“FMT”)指示在这个分组中提供的反馈类型。 Feedback type field ("FMT") indicates the type of feedback provided in this packet. 而且,在反馈长度字段(“LSB Feedback Length”)中指示了分组的长度。 Moreover, in the feedback length field ("LSB Feedback Length") indicates the length of the packet. 这个字段被最小有效位(LSB)编码。 This field is the least significant bit (LSB) encoding. 比特屏蔽字段(“BLP”)指示最后的分组。 Bit mask field ("BLP") indicates the last packet. 第一比特是BLP字段(比特屏蔽字段),允许对在由分组标识符指示的RTP分组后面紧随的任何RTP分组报告损耗。 BLP field is the first bit (bit mask field), allowing for any later by the RTP packet identifier indicating the RTP packet immediately report packet loss. 在反馈类型字段(FMT)指示[Gu4]一般确认的情况下,BLP字段的第一比特是(所谓的R比特)1。 In the feedback type field (FMT) indicates [Gu4] case of a general acknowledgment, BLP is the first bit of the field (the so-called R-bit) 1. 在这种情况下,BLP字段用于识别由所述压缩的应用限定(APP)分组确认的附加分组的数量。 In this case, BLP field is used to identify the number of application defined by said compressed (APP) packet acknowledgment packet added. 否则,如果R=0,则BLP字段带有指示损耗分组的比特屏蔽。 Otherwise, if the bit R = 0, then the packet loss indication with BLP field shielding.

总之,上面提出的压缩控制分组以及两个新引入的分组格式(初始化分组和刷新分组)意欲降低控制协议的分组的总体平均分组大小,使得可以降低报告间隔T。 In summary, the overall average packet size of the packet compression control set forth above, and the introduction of two new packet format (initialization and refresh packet packet) intended to reduce the control protocol packets, making it possible to reduce the reporting interval T.

一方面[Gu5],通过“提前”发送在初始化分组内的静态场境参数以及初始化偶尔改变的场境参数来降低数据量。 On the one hand [Gu5], by the "Advance" initialization packet sent within a static context initialization parameters and context parameters occasionally changed to reduce the amount of data. 为了能够在偶尔改变的场境参数改变时更新它,使用刷新分组来做到这一点。 To be able to update it occasionally changed context parameters change, use the refresh group to do this. 另一方面,编码多数控制分组字段,以便进一步降低它们的大小。 On the other hand, most of the control packet coding field, in order to further reduce their size.

因此,与标准RTCP协议相比较,有可能降低压缩的控制数据分组的平均大小。 Therefore, compared with the standard RTCP protocol, it is possible to reduce the compressed data packet control average size. 因此,利用所提出的分组格式,有可能大大降低报告间隔T而不扩展控制通信量[Gu5][FH7]的所分配带宽。 Therefore, using the proposed packet format, it is possible to greatly reduce the reporting interval T without expansion control traffic [Gu5] [FH7] of the allocated bandwidth. 因此,通过能够在较短的时间间隔中提供反馈,媒体数据传输会话的参与者可以比在使用标准RTP/RTCP协议组合的会话中更快地适应于变化的传输环境。 Therefore, feedback can be provided by participants in the media data transmission session may be shorter than the time interval using the standard RTP / RTCP session protocol combination to adapt more quickly to changes in the transmission environment. 因此,可以大大地改进诸如MPEG-4编码视频数据的被发送(或被广播)的应用数据的整体质量。 Thus, as can greatly improve the overall MPEG-4 encoding quality is transmitted (or broadcast) of the application data of the video data.

所提出的首标和数据压缩机制仅仅处理RTCP首标和数据部分,而不是低层UDP/IP首标。 Header and data compression scheme proposed only deal RTCP header and data section, rather than the lower UDP / IP header. 因此,与一般适用于最后一个跳跃或第一个跳跃点到点链路的、诸如与在RPF 3095中所提出的首标压缩方案相比较,在此所述的手段可以被最后跳跃或第一跳跃的端到端适用。 Thus, generally applicable to the last jump or skip the first point of the link, such as a comparison with header compression scheme RPF 3095 as proposed, means herein may be the last jump or first jumping end apply. 沿着去分组的路线的中间跳跃不必关心被压缩的控制分组,因为中间跳跃将它们看作或者第三层IP分组数据或第四层传送层分组数据。 Along the center line of the jump to the grouping does not have to be concerned about the compressed control packets, because the middle will be seen as they jump or third layer or fourth layer IP packet data transport layer packet data. 不引入任何附加的中间主机的开销。 Without introducing any additional overhead intermediate host. 但是如果与用于在第一/最后跳跃无线链路中的低层首标的鲁棒首标压缩一起使用,则可以节省更多的带宽。 However, if used with for the first / last jump wireless link in the lower header robust header compression, you can save more bandwidth.

Referenced by
Citing PatentFiling datePublication dateApplicantTitle
CN1315307C *5 Aug 20049 May 2007北京航空航天大学Method for transmission using Mbus communication middle ware
Classifications
International ClassificationH04L29/06, H04L12/56
Cooperative ClassificationH04L65/608, H04L29/06027, H04L69/22, H04L69/04
European ClassificationH04L29/06M6P, H04L29/06C5, H04L29/06C2
Legal Events
DateCodeEventDescription
7 Jul 2004C06Publication
15 Sep 2004C10Request of examination as to substance
23 Aug 2006C02Deemed withdrawal of patent application after publication (patent law 2001)