WO2013000342A1 - 拥塞控制方法、装置和系统 - Google Patents

拥塞控制方法、装置和系统 Download PDF

Info

Publication number
WO2013000342A1
WO2013000342A1 PCT/CN2012/075281 CN2012075281W WO2013000342A1 WO 2013000342 A1 WO2013000342 A1 WO 2013000342A1 CN 2012075281 W CN2012075281 W CN 2012075281W WO 2013000342 A1 WO2013000342 A1 WO 2013000342A1
Authority
WO
WIPO (PCT)
Prior art keywords
cps
frame
user data
cell
data frame
Prior art date
Application number
PCT/CN2012/075281
Other languages
English (en)
French (fr)
Inventor
张美峰
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2013000342A1 publication Critical patent/WO2013000342A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/32Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames

Definitions

  • the present invention relates to the field of communications, and in particular, to a congestion control method and system for an Asynchronous Transfer Mode Adaptation Layer Type 2 (AAL2, ATM Adaptation Layer 2) bearer type.
  • AAL2 Asynchronous Transfer Mode Adaptation Layer Type 2
  • AAL2 type transmission networks With the continuous development of communication technologies, the length of service data carried by AAL2 type transmission networks has grown from an early ten or tens of bytes to a few thousand bytes today.
  • ATM asynchronous transfer mode
  • a large amount of retransmission data may appear in the network, and these retransmission data may occur. Occupy transmission bandwidth, causing congestion, and even network congestion.
  • ATM switching is a kind of packet switching.
  • a more efficient way is to discard in units of user frames, because when performing cell-level discarding Naturally, bad frames are generated, which causes retransmission of the upper layer protocol. Therefore, frame dropping can avoid or reduce network congestion caused by retransmission.
  • Figure 1 shows that during a transmission, when a cell needs to be discarded, the subsequent cell belonging to the same user data will also be discarded, leaving only the last cell for the upper layer.
  • AAL2 includes a Common Part Sublayer (CPS) and a Service Specific Segmentation and Reassembly Sublayer.
  • CPS Common Part Sublayer
  • Service Specific Segmentation and Reassembly Sublayer the relationship between the data units of each layer of the transport network is shown in Figure 2.
  • ATM cells belonging to the same piece of user data may be discontinuously transmitted.
  • CPS-PKT CPS packet
  • the CPS-PKT of the user data may be CPS multiplexed by inter-interleaving intervals, so that packet-by-packet processing cannot be performed during the ATM transmission process (one upper layer service data packet may include multiple CPS-PKTs).
  • the segmentation identifier of the user data is also terminated in the CPS-PKT, and the ATM layer cannot distinguish the data in the current cell. No is the data for the last segment. Therefore, according to the standard AAL2 processing, the conditions for ATM switching by frame dropping cannot be satisfied, and can only be discarded by cells, which can still cause a large amount of data retransmission, causing network congestion and even network congestion. Summary of the invention
  • the present invention provides a congestion control method and system that solves the problem of network congestion in the AAL2 bearer mode.
  • a congestion control method in which AAL2 type data is transmitted between network nodes, the method includes: performing CPS multiplexing on a user data frame to be transmitted, and continuously multiplexing multiple CPS-PKTs in the same user data frame;
  • the cell of the user data frame is sent to the network node.
  • performing CPS multiplexing on the user data frame to be sent includes:
  • the SSSAR-PDU After the segmentation processing of the user data frame to be sent in the SSSAR sub-layer, the SSSAR-PDU is obtained, and the last SSSAR-PDU of the user data frame carries a frame tail flag; at the CPS layer, the user data is The SSSAR-PDU obtained after the frame segmentation is encapsulated into a CPS-PKT, and the encapsulated CPS-PKT is continuously multiplexed, and the CPS-PKT corresponding to the last SSSAR-PDU carries a frame tail flag.
  • the continuously multiplexed plurality of CPS-PKTs are encapsulated to obtain a cell, and the end of the CPS-PKT obtained by encapsulation of the CPS multiplexed queue is filled with a frame tail flag.
  • the steps include:
  • the consecutively multiplexed multiple CPS-PKTs are encapsulated to obtain one or more CPS-PDUs, and a frame tail flag is added to a CPS-PDU carrying a CPS-PKT carrying a frame tail flag;
  • the mode (ATM) layer receives the CPS-PDU from the CPS layer, performs cell encapsulation on the received CPS-PDU, and adds a foot-end flag to the cell carrying the CPS-PDU carrying the end-of-frame flag.
  • adding the end of frame flag to a cell loaded with a CPS-PDU carrying a frame tail flag includes: setting an ATM user-to-user indication (AUU) bit of a payload type (PTI) field of the cell, To indicate that the cell contains the last CPS-PKT of the user data frame.
  • AUU ATM user-to-user indication
  • PTI payload type
  • the method further includes: the network node receiving the continuously transmitted cells belonging to the same user data frame;
  • the frame of the user data frame corresponding to the cell is discarded according to the indication of the cell carrying the frame tail flag.
  • the present invention also provides a congestion control apparatus, which is suitable for a network node that uses AAL2 type to transmit data, including:
  • a multiplexing module configured to: perform CPS multiplexing on user data frames to be sent, and continuously multiplex multiple CPS-PKTs of the same user data frame;
  • Encapsulating module configured to: encapsulate the plurality of consecutively multiplexed CPS-PKTs to obtain a cell, and fill in a frame tail flag in the cell including the last CPS-PKT, indicating to receive the user data frame
  • the network node discards the user data frame during congestion control
  • a sending module configured to: send the cell of the user data frame to a network node.
  • the multiplexing module includes:
  • the SSSAR sub-layer segmentation unit is configured to: after the segmentation processing of the user data frame to be sent in the SSSAR sub-layer, obtain the SSSAR-PDU, and carry the end-of-frame flag in the last SSSAR-PDU of the user data frame. ;
  • the CPS multiplexing unit is configured to: in the CPS layer, encapsulate the SSSAR-PDU obtained by segmenting the user data frame into a CPS-PKT, and continuously multiplex the encapsulated CPS-PKT, the last SSSAR- The PDU corresponding to the CPS-PKT carries a frame tail flag.
  • the package module comprises:
  • the CPS-PKT encapsulation unit is configured to: at the CPS layer, encapsulate the continuously multiplexed CPS-PKT to obtain one or more CPS-PDUs, and add in a CPS-PDU loaded with a CPS-PKT carrying a frame tail flag. End of frame flag;
  • the ATM encapsulation unit is configured to: at the ATM layer, receive the CPS-PDU from the CPS, perform cell encapsulation on the received CPS-PDU, and add a frame tail flag to the cell carrying the CPS-PDU carrying the end-of-frame flag. .
  • the present invention also provides a congestion control system including a first network node and a second network node, wherein the first network node and the second network node use AAL2 type to transmit data;
  • the first network node is configured to: perform CPS multiplexing on the user data frame to be sent, and continuously multiplex multiple CPS-PKTs of the same user data frame, and encapsulate the multiple multiplexed CPS-PKTs. a cell, and filling in a frame tail flag in a cell containing the last CPS-PKT, indicating that the network node receiving the user data frame performs frame discarding on the user data frame during congestion control, and A cell of the user data frame is transmitted to the second network node.
  • the second network node is configured to: receive a cell that is sent by the first network node and that belongs to the same user data frame, and when it needs to perform congestion control, search for a cell that carries a frame tail flag, and When the cell carrying the frame tail flag is detected, the frame of the user data frame corresponding to the cell is discarded according to the indication of the cell carrying the frame tail flag.
  • An embodiment of the present invention provides a congestion control method and system, which performs CPS multiplexing on a user data frame to be transmitted, and multiple CPS-PKTs of the same user data frame are continuously multiplexed, and the multiplexed queues in the CPS are multiplexed.
  • the CPS-PKT performs encapsulation to obtain a cell, and fills a frame tail flag in the cell containing the last CPS-PKT obtained after encapsulation of the CPS multiplexing queue, indicating that the network node that receives the user data frame is During the congestion control, the frame of the user data frame is discarded, and the cell of the user data frame is sent to the network node, so that other cells belonging to one data frame are lost in the case of the lost cell. The frame is discarded, which solves the problem of network congestion in the AAL2 bearer mode.
  • Figure 1 is a schematic diagram of packet loss by frame
  • FIG. 2 is a relationship diagram of data units of each layer in an AAL2 type transmission network
  • FIG. 5 is a schematic structural diagram of a congestion control apparatus according to Embodiment 1 of the present invention;
  • FIG. 6 is a schematic diagram of Embodiment 1 of the present invention.
  • FIG. 7 is a schematic structural diagram of a congestion control system according to Embodiment 2 of the present invention
  • FIG. 8 is a flowchart of a congestion control method according to Embodiment 2 of the present invention.
  • FIG. 9 is a schematic diagram of AAL2 layer processing after implementation according to frame discarding data configuration according to Embodiment 2 of the present invention.
  • step 601 in FIG. 6 is a specific flowchart of step 601 in FIG. 6;
  • FIG 11 is a detailed flow chart of step 602 of Figure 6.
  • Preferred embodiment of the invention
  • the conditions for ATM switch-by-frame discarding cannot be met, and can only be discarded by cells, which can still cause a large amount of data retransmission, causing network congestion and even network congestion.
  • the embodiment of the present invention provides a congestion control system.
  • the structure of the system is as shown in FIG. 4, including a first network node 401 and a second network node 402, and the first network node 401 and the second network node 402. Transfer data using the AAL2 type.
  • the second network node 402 is an ATM switching device, and the system further includes a third network node 403 for performing data exchange with the first network node 401.
  • the first network node 401 and the third network node 403 each include an ATM layer and an AAL2 layer, and the AAL2 layer further includes an SSSAR sublayer and a CPS.
  • the first network node 401 is configured to perform CPS multiplexing on the user data frame to be sent, and multiple CPS-PKTs of the same user data frame are continuously multiplexed, and the multiple multiplexed CPS-PKTs are encapsulated. Get the cell, and fill in the end of the frame in the cell containing the last CPS-PKT, The network node receiving the user data frame discards the user data frame during congestion control, and transmits the cell of the user data frame to the second network node 402.
  • the second network node 402 is configured to receive the cells that are sent by the first network node 401 and belong to the same user data frame, and when the congestion control is needed, find the cell that carries the frame tail flag. And when the cell carrying the frame tail flag is detected, the frame of the user data frame corresponding to the cell is discarded according to the indication of the cell carrying the frame tail flag.
  • the second network node 402 is further configured to forward the cell that the first network node 401 sends to the third network node 403 to the third network node 403 when congestion control is not required.
  • the cell involved in the embodiment of the present invention is specifically an ATM cell.
  • the embodiment of the invention further provides a congestion control device, which can be integrated in the first network node 401 of the congestion control system shown in FIG.
  • the structure of the device is as shown in FIG. 5, including: a multiplexing module 501, configured to perform CPS multiplexing on user data frames to be sent, and multiple CPS-PKTs of the same user data frame are continuously multiplexed;
  • the encapsulating module 502 is configured to encapsulate the continuously multiplexed multiple CPS-PKTs to obtain a cell, and fill in a frame tail flag in the last CPS-PKT cell, and indicate a network that receives the user data frame.
  • the node discards the user data frame during congestion control;
  • the sending module 503 is configured to send the cell of the user data frame to a network node.
  • the multiplexing module 501 is specifically configured to perform SSSAR segmentation and CPS multiplexing encapsulation on the user data frame to be sent, and multiple CPS-PKTs of the same user data frame are continuously multiplexed, and the internal structure of the multiplexing module 501 is Includes:
  • the SSSAR sub-layer segmentation unit 5011 is configured to obtain the SSSAR-PDU after the segmentation processing of the user data frame to be sent in the SSSAR sub-layer, and carry the end-of-frame flag in the last SSSAR-PDU of the user data frame. ;
  • a CPS multiplexing unit 5012 configured to segment the user data frame at the CPS layer
  • the SSSAR-PDU is encapsulated into CPS-PKT, and the encapsulated CPS-PKT is multiplexed into the same CPS multiplexing queue.
  • the last SSSAR-PDU corresponding to the CPS-PKT carries the frame tail flag.
  • the encapsulating module 502 is specifically configured to obtain a CPS-PDU from the multiplexing module 501 and perform cell encapsulation, and encapsulate the end-of-slot tag delivered by the multiplexing module into the UI, indicating the network that receives the user data frame.
  • the frame discarding of the user data frame by the node during congestion control includes:
  • the CPS-PKT encapsulating unit 5021 is configured to, in the CPS layer, encapsulate one or more CPS-PDUs in the CPS-PKT in the CPS multiplexing queue, and load the CPS-PDU of the CPS-PKT carrying the end-of-frame flag. Add a frame tail flag;
  • the ATM encapsulating unit 5022 is configured to receive, at the ATM layer, a CPS-PDU from the CPS, perform cell encapsulation on the received CPS-PDU, and add a frame tail flag to a cell carrying a CPS-PDU carrying a frame tail flag.
  • the CPS-PKT encapsulation unit 5021 can also be added to the multiplexing module 501.
  • the embodiment of the present invention further provides a congestion control method, and the process for completing the congestion control using the method is as shown in FIG. 6, including:
  • Step 601 The first network node performs CPS multiplexing on the user data to be sent, and multiple CPS-PKTs of the same user data frame are continuously multiplexed;
  • the step is specifically performed on the AAL2 layer of the first network node.
  • the user data to be sent is segmented and processed in the SSSAR sublayer to obtain an SSSAR-PDU.
  • the data to be sent after the segmentation process is performed. Packed, sent to the ATM layer after CPS multiplexing. In order to ensure that the CPS-PKT of the same user data frame must be continuously subjected to CPS multiplexing processing.
  • Step 602 The first network node encapsulates a plurality of CPS-PKTs in the CPS multiplexing queue to obtain a cell, and fills a frame tail flag in a cell that includes the last CPS-PKT.
  • the end of frame flag is used to indicate that the network node that receives the user data frame discards the user data frame during congestion control.
  • one or more CPS-PDUs are obtained by successively multiplexing multiple CPS-PKTs, and a frame tail flag is added to a CPS-PDU carrying a CPS-PKT carrying a frame tail flag;
  • Receiving a CPS-PDU from the CPS performing cell encapsulation on the received CPS-PDU, and adding a frame tail flag to the cell carrying the CPS-PDU carrying the end-of-frame flag, where the first network device is in the PTI field,
  • Filling in the end of frame flag specifically, an ATM user-to-user indication (AUU) bit of the payload type ( ⁇ ) field of the cell may be set to indicate that the cell includes the The last CPS-PKT of the user data frame.
  • AUU ATM user-to-user indication
  • the CPS-PKT carrying the frame tail flag is at the end of the CPS multiplexing queue, and the CPS multiplexing implements the original CPS multiplexing function, and also implements the end-of-frame flag to the ATM layer.
  • the function of filling the end-of-frame flag in the cell header of the ATM cell is added to be used when determining the last ATM cell in a user data frame.
  • Step 603 The first network node sends the cell of the user data frame to the second network node.
  • Step 604 The second network node receives the cell that is continuously sent by the first network node and belongs to the same user data frame.
  • Step 605 The second network node searches for a cell carrying a frame tail flag when congestion control is required.
  • the second network node determines that congestion control is required, and frame discarding is performed. Specifically, the cell header of the cell from the lost cell is checked, and it is confirmed whether the PTI of the cell header contains a frame tail flag indicating that the cell payload has the last CPS-PKT.
  • Step 606 When detecting a cell carrying a frame tail flag, perform frame discarding on the user data frame corresponding to the cell according to the indication of the cell carrying the frame tail flag.
  • the second network node determines, according to the end-of-frame flag in the cell header, the cell in which the last CPS-PKT corresponding to the user data frame is located, and then performs frame discarding.
  • the frame discarding is specifically: discarding the lost cell to the cell between the cells including the end of the user data frame end tag, and only retaining the cell carrying the end of frame tag.
  • the embodiment of the present invention provides a congestion control system, and its structure is as shown in FIG. 7.
  • the method includes: a congestion control attribute configuration module 701, an AAL2 adaptation layer data processing module 702 based on congestion control requirements, and an AAL2 letter based on congestion control requirements.
  • the congestion control attribute configuration module 701 configured according to the frame-based congestion control policy (the policy indicates whether the function of dropping by frame) is configured to indicate whether the AAL2 layer satisfies the CPS multiplexing based on frame dropping. Whether to implement frame discarding with the ATM switching device;
  • the AAL2 adaptation layer data processing module 702 required by the plug control configures the configuration information of the module according to the congestion control attribute. The module needs to ensure that the CPS-PKT of the same user data frame must be continuously CPS complex while completing the standard AAL2 layer function.
  • CPS multiplexing passes the end-of-frame flag to the ATM layer (which carries the end-of-frame flag in the CPS-PDU transmitted to the ATM layer) while completing the original CPS multiplexing function;
  • AAL2 based on congestion control needs
  • the cell processing module 403 adds a function of filling the end-of-frame flag in the ATM cell header based on the encapsulation of the original ATM cell, so as to be used by the frame discard-based congestion control module 704;
  • the congestion control module 704 is configured to implement a function of dropping by frame according to a frame tail flag in the cell header.
  • the congestion control attribute configuration module 701 is in the network controller for completing the configuration of the entire congestion control; the network nodes for completing the service include the AAL2 adaptation layer data processing module 702 based on the congestion control needs and the AAL2 based on the congestion control requirement.
  • the cell processing module 703, specifically, the AAL2 adaptation layer data processing module 702 based on the congestion control needs is in the AAL2 layer, including the SSSAR sublayer module and the CPS module, and the AAL2 cell processing module 703 based on the congestion control needs is in the ATM. Layer;
  • the frame drop based congestion control module 704 is at the ATM layer and is typically integrated into the ATM switching device.
  • the embodiment of the present invention further provides a congestion control method, and the congestion control system shown in FIG. 7 completes the congestion control process as shown in FIG. 8 , and includes:
  • Step 801 The AAL2 layer SSSAR sublayer module performs segmentation processing according to the configuration attribute. If the configuration requires frame transmission, the SSSAR-PDUs of different SSCS-PDUs of the same ATM virtual link are not crossed into the receiving queue of the CPS module. To the SSSAR-PDU queue);
  • Step 802 The CPS module performs CPS-PKT encapsulation on the SSSAR-PDU obtained by the SSSAR sub-layer segment, performs CPS multiplexing, and carries the end-of-sale flag of the SSSAR-PDU to the AAL2 cell processing module required for congestion control;
  • Step 803 The AAL2 cell processing module that is required for congestion control fills the end of the frame in the ATM-SDU into the PTI field.
  • Step 804 The ATM switching device performs congestion control according to the configuration attribute. If the configuration requires congestion control according to the frame, the frame discarding function is implemented according to the AUU flag of the PTI field.
  • the frame discard-based congestion control module in the ATM switching device selects whether to perform the frame discarding congestion control policy according to the ATM virtual link configuration attribute.
  • the configuration attribute is frame-based discarding
  • the cell discarding process is performed by identifying the tail packet flag in the cell header.
  • This processing method can use the EPD/PPD processing principle.
  • the CPS-PKT corresponding to the same user data frame will be continuously multiplexed and continuously transmitted in the ATM network, as shown in FIG. 9, thus having the frame-based disc dropping cell Under the condition that the CPS-PKT end-of-frame flag is encapsulated into the cell, the congestion control method based on the frame discarding cell can be realized.
  • Step 801 is specifically shown in FIG. 10, and includes:
  • Step 8011 After receiving the service packet, if it is required to implement the frame flow control policy, go to step 8012; otherwise, go to step 8013;
  • Step 8012 Obtain the enqueue permission of the CPS module receiving queue corresponding to the ATM virtual link;
  • Step 8013 Perform segmentation processing on the packet;
  • Step 8014 Send the current segment to a corresponding CPS receiving queue.
  • Step 8015 if it is the last segment, proceeds to step 8016, if it is not the last segment, proceeds to step 8013;
  • Step 8016 if the CPS module enrollment authority has been obtained, go to step 8017, if not, go to step 8018;
  • Step 8017 Release the CPS module to receive the queue enrollment authority
  • Step 8018 the end.
  • Step 802 is specifically as shown in FIG. 11, and includes:
  • Step 8201 Receive a CPS-SDU, encapsulate the CPS-PKT header, if it is the last SSSAR-PDU, set the current CPS-PKT tail packet flag to 1, otherwise set to 0, go to step 8202;
  • Step 8202 performing text multiplexing processing, and proceeding to step 8203;
  • Step 8203 according to the currently received CPS-PKT end of frame flag and the previous end of frame flag
  • the end-of-frame flag of the new current CPS-PDU recording whether the current CPS-PDU includes a frame tail flag, and if so, the flag value is 1;
  • Step 8204 Whether the current CPS-PDU meets the completion multiplexing condition, if yes, go to step 8205; otherwise, go to step 8212;
  • Step 8205 whether the current CPS-PDU includes a frame tail flag, if it is included, proceeds to step 8206, if not, proceeds to step 8207;
  • Step 8206 set the AUU flag to 0, and proceeds to step 8208;
  • Step 8207 set the AUU flag to 1, and proceed to step 8208;
  • Step 8208 Send the current CPS-PDU to the ATM queue, and go to step 8209; Step 8209, whether the current CPS-PKT data has been completely reused, if yes, go to step 8210, if not, go to step 8211. ;
  • Step 8210 Clear the end of the current CPS-PDU, and proceed to step 8213.
  • Step 8211 whether the current CPS-PKT is a tail packet, if yes, the current CPS-PDU tail packet flag is set to 1, if not, then set to 0, and proceeds to step 8212;
  • Step 8212 waiting to continue multiplexing
  • Step 8213 the end.
  • the congestion control system provided by the embodiment of the present invention can be combined with the congestion control method provided by the embodiment of the present invention.
  • the AAL2 type bearer data is used between the network nodes, and the user data to be sent is CPS multiplexed, and the same Multiple CPS-PKTs of user data are continuously multiplexed, and in the PTI field of the ATM cell including the last CPS-PKT corresponding to the user data, the end of frame flag is filled in, and the ATM cell of the user data is used.
  • all or part of the steps of the foregoing embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve.
  • the invention is not limited to any particular combination of hardware and software.
  • the various devices/function modules/functional units in the above embodiments may be implemented using a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.
  • Each device/function module/functional unit in the above embodiments can be stored in a computer readable storage medium when implemented in the form of a software function module and sold or used as a standalone product.
  • the above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.
  • the present invention provides a congestion control method and system, which performs CPS multiplexing on user data frames to be transmitted, and multiple CPS-PKTs of the same user data frame are continuously multiplexed, and the CPS multiplexing queue is multiplexed.
  • the plurality of CPS-PKTs are encapsulated to obtain a cell, and a cell tail flag is filled in the cell containing the last CPS-PKT obtained after encapsulation of the CPS multiplexing queue, indicating that the user data frame is received.
  • the network node discards the user data frame during the congestion control, and sends the cell of the user data frame to the network node, so that in the case that the sending cell is lost, the same belongs to a data frame.
  • the frame discarding of other cells solves the problem of network congestion in the AAL2 bearer mode.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

一种拥塞控制方法、装置和系统。涉及通信领域;解决了异步传输模式适配层2(AAL2)承载模式下网络拥塞的问题。该方法包括:将待发送的用户数据帧进行公共部分子层(CPS)复用,同一用户数据帧的多个CPS-PKT连续复用;对所述连续复用的多个CPS-PKT进行封装得到信元,并在包含最后一个CPS-PKT的信元中填入帧尾标志,指示接收所述用户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丢弃;将所述用户数据帧的信元向网络节点发送。本申请提供的技术方案适用于AAL2承载类型下的传输网络,实现了高效的拥塞控制。

Description

拥塞控制方法、 装置和系统
技术领域
本发明涉及通信领域, 尤其涉及一种于异步转移模式适配层类型 2 ( AAL2, ATM Adaptation Layer 2 )承载类型下的拥塞控制方法和系统。
背景技术
随着通信技术的不断发展, AAL2 类型传输网络承载的业务数据的长度 由早期的十几、 几十字节, 发展到了如今的几千字节。 在这种传输网络中, 当异步转移模式(ATM )传输过程中出现丟包时, 如果不能按照上层业务数 据包进行丟弃的话, 就可能在网络中出现大量的重传数据, 这些重传数据占 用传输带宽, 引起拥塞, 甚至网络瘫痪。
ATM交换作为一种分组交换, 当因交换网络出现拥塞或其它原因需要丟 弃信元时, 一种更有效的方式是以用户帧为单位进行丟弃, 因为当进行信元 一级丟弃时自然就产生了坏帧, 从而引起上层协议的重传, 因此, 帧丟弃可 以避免或者减少因重传带了的网络拥塞。 如图 1所示, 在传输过程中, 当某 信元需要被丟弃时, 属于同一个用户数据的之后的信元也将被丟弃, 只保留 最后一个信元用于上层进行。
AAL2包括公共部分子层(CPS, Common Part Sublayer )和特定业务分 段和重装子层 ( SSSAR子层, Service Specific Segmentation and Reassembly Sublayer )。在 AAL2承载模式下,传输网络各层数据单元的关系如图 2所示。
在传输网络工作在 AAL2承载模式时, 归属于同一条用户数据(或同一 帧)的 ATM信元可能是非连续传输的。 如图 3所示, 同一条 ATM虚链接的 用户数据经过 SSSAR子层分段, 并封装成 CPS包( CPS-PKT )进入 CPS复 用队列时, 是不保证被连续复用的, 也就是多条用户数据的 CPS-PKT可能被 互相穿插间隔进行 CPS复用, 这样, 在 ATM传输过程中就无法进行按包丟 弃的处理(一个上层业务数据包可能包括多个 CPS-PKT ) 。 另外, 用户数据 的分段标识也终结在 CPS-PKT中, ATM层也无法区分当前信元中的数据是 否是最后一个分段的数据。 因此, 按照标准的 AAL2处理, 无法满足 ATM交 换按帧丟弃的条件, 只能按信元丟弃, 仍然可以造成大量数据重传的局面, 引发网络拥塞, 甚至网络瘫痪。 发明内容
本发明提供了一种拥塞控制方法和系统, 解决了 AAL2承载模式下网络 拥塞的问题。
一种拥塞控制方法, 网络节点间使用 AAL2类型传输数据, 该方法包括: 将待发送的用户数据帧进行 CPS复用, 同一用户数据帧的多个 CPS-PKT 连续复用;
对所述连续复用的多个 CPS-PKT进行封装得到信元,并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示接收所述用户数据帧的网络节点在拥 塞控制时对该用户数据帧的进行帧丟弃;
将所述用户数据帧的信元向网络节点发送。
优选的, 所述将待发送的用户数据帧进行 CPS复用包括:
在 SSSAR 子层将所述待发送的用户数据帧进行分段处理后得到 SSSAR-PDU, 在该用户数据帧的最后一个 SSSAR-PDU中携带有帧尾标志; 在 CPS 层, 将所述用户数据帧分段后得到的 SSSAR-PDU 封装成 CPS-PKT, 并将封装得到的 CPS-PKT连续复用, 所述最后一个 SSSAR-PDU 对应的 CPS-PKT均携带有帧尾标志。
优选的, 对所述连续复用的多个 CPS-PKT进行封装得到信元, 并在所述 CPS复用队列的封装后得到的包含最后一个 CPS-PKT的信元中填入帧尾标志 的步骤包括:
在 CPS 层, 将所述连续复用的多个 CPS-PKT 封装得到一个或多个 CPS-PDU,在荷载有携带帧尾标志的 CPS-PKT的 CPS-PDU中添加帧尾标志; 在异步传输模式(ATM )层, 接收来自 CPS层的 CPS-PDU, 对接收到 的 CPS-PDU进行信元封装, 在荷载有携带帧尾标志的 CPS-PDU的信元中添 力口†贞尾标志。 优选的,所述在荷载有携带帧尾标志的 CPS-PDU的信元中添加帧尾标志 包括: 设置所述信元的净荷类型 ( PTI )字段的 ATM用户至用户指示 ( AUU ) 位, 以表明该信元包含所述用户数据帧的最后一个 CPS-PKT。
优选的, 所述将所述用户数据帧的信元向网络节点发送的步骤之后, 还 包括: 所述网络节点接收连续发送的归属于同一用户数据帧的信元;
在需要进行拥塞控制时, 查找携带有帧尾标志的信元;
在检测到携带有帧尾标志的信元时, 根据该携带有帧尾标志的信元的指 示, 对该信元对应的用户数据帧进行帧丟弃。
本发明还提供了一种拥塞控制装置, 适用于使用 AAL2类型传输数据的 网络节点, 包括:
复用模块, 其设置为: 将待发送的用户数据帧进行 CPS复用, 同一用户 数据帧的多个 CPS-PKT连续复用;
封装模块, 其设置为: 对所述连续复用的多个 CPS-PKT进行封装得到信 元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示接收所述用户 数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃;
发送模块, 其设置为: 将所述用户数据帧的信元向网络节点发送。
优选的, 所述复用模块包括:
SSSAR子层分段单元, 设置为: 在 SSSAR子层将所述待发送的用户数 据帧进行分段处理后得到 SSSAR-PDU , 在该用户数据帧的最后一个 SSSAR-PDU中携带有帧尾标志;
CPS 复用单元, 设置为: 在 CPS层, 将所述用户数据帧分段后得到的 SSSAR-PDU封装成 CPS-PKT , 并将封装得到的 CPS-PKT连续复用, 所述最 后一个 SSSAR-PDU对应 CPS-PKT均携带有帧尾标志。
优选的, 所述封装模块包括:
CPS-PKT封装单元, 设置为: 在 CPS层, 将所述连续复用的 CPS-PKT 封装得到一个或多个 CPS-PDU, 在荷载有携带帧尾标志的 CPS-PKT 的 CPS-PDU中添加帧尾标志; ATM封装单元, 设置为: 在 ATM层, 接收来自 CPS的 CPS-PDU, 对 接收到的 CPS-PDU进行信元封装, 在荷载有携带帧尾标志的 CPS-PDU的信 元中添加帧尾标志。
本发明还提供了一种拥塞控制系统,包括第一网络节点和第二网络节点, 所述第一网络节点和所述第二网络节点间使用 AAL2类型传输数据;
所述第一网络节点设置为: 将待发送的用户数据帧进行 CPS复用, 同一 用户数据帧的多个 CPS-PKT连续复用, 对所述连续复用的多个 CPS-PKT进 行封装得到信元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示 接收所述用户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟 弃, 并将所述用户数据帧的信元向所述第二网络节点发送。
优选的, 所述第二网络节点设置为: 接收所述第一网络节点连续发送的 归属于同一用户数据帧的信元, 在需要进行拥塞控制时, 查找携带有帧尾标 志的信元, 并在检测到携带有帧尾标志的信元时, 根据该携带有帧尾标志的 信元的指示, 对该信元对应的用户数据帧进行帧丟弃。
本发明实施例提供了一种拥塞控制方法和系统, 将待发送的用户数据帧 进行 CPS复用, 同一用户数据帧的多个 CPS-PKT连续复用, 对所述 CPS复 用队列中的多个 CPS-PKT进行封装得到信元, 并在所述 CPS复用队列的封 装后得到的包含最后一个 CPS-PKT的信元中填入帧尾标志,指示接收所述用 户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃, 并将所述 用户数据帧的信元向网络节点发送, 实现了在发信元丟失的情况下, 对同属 于一个数据帧的其他信元的帧丟弃, 解决了 AAL2承载模式下网络拥塞的问 题。 附图概述
图 1 为按帧丟包的原理图;
图 2为 AAL2类型传输网络中各层数据单元关系图;
图 3为传输网络工作在 AAL2承载模式时标准的 AAL2层处理方式示意 图; 图 4为本发明的实施例一提供的一种拥塞控制系统的结构示意图; 图 5为本发明的实施例一提供的一种拥塞控制装置的结构示意图; 图 6为本发明的实施例一提供的一种拥塞控制方法的流程图;
图 7为本发明的实施例二提供的一种拥塞控制系统的结构示意图; 图 8为本发明的实施例二提供的一种拥塞控制方法的流程图;
图 9为本发明的实施例二中按照基于帧丟弃数据配置实现后的 AAL2层 处理原理图;
图 10为图 6中步骤 601的具体流程图;
图 11为图 6中步骤 602的具体流程图。 本发明的较佳实施方式
按照标准的 AAL2处理,无法满足 ATM交换按帧丟弃的条件,只能按信 元丟弃, 仍然可以造成大量数据重传的局面, 引发网络拥塞, 甚至网络瘫痪。
为了解决上述问题, 本发明的实施例提供了一种拥塞控制方法和系统。 下文中将结合附图对本发明的实施例进行详细说明。 需要说明的是, 在不冲 突的情况下, 本申请中的实施例及实施例中的特征可以相互任意组合。
首先结合附图, 对本发明的实施例一进行说明。
本发明实施例提供了一种拥塞控制系统, 该系统的结构如图 4所示, 包 括第一网络节点 401和第二网络节点 402, 所述第一网络节点 401和所述第 二网络节点 402间使用 AAL2类型传输数据。
优选的,所述第二网络节点 402为 ATM交换设备,该系统还包括与所述 第一网络节点 401进行数据交换的第三网络节点 403。
第一网络节点 401和第三网络节点 403均包括 ATM层和 AAL2层, AAL2 层又包括 SSSAR子层和 CPS。
所述第一网络节点 401 , 用于将待发送的用户数据帧进行 CPS复用, 同 一用户数据帧的多个 CPS-PKT连续复用, 对所述连续复用的多个 CPS-PKT 进行封装得到信元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指 示接收所述用户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟 弃, 并将所述用户数据帧的信元向所述第二网络节点 402发送。
优选的, 所述第二网络节点 402, 用于接收所述第一网络节点 401连续 发送的归属于同一用户数据帧的信元, 在需要进行拥塞控制时, 查找携带有 帧尾标志的信元, 并在检测到携带有帧尾标志的信元时, 根据该携带有帧尾 标志的信元的指示, 对该信元对应的用户数据帧进行帧丟弃。
优选的, 所述第二网络节点 402, 还用于在不需要进行拥塞控制时, 向 所述第三网络节点 403转发所述第一网络节点 401向第三网络节点 403发送 的信元。
本发明实施例中所涉及的信元具体为 ATM信元。
本发明实施例还提供了一种拥塞控制装置, 该装置可集成于图 4所示拥 塞控制系统的所述第一网络节点 401中。 该装置的结构如图 5所示, 包括: 复用模块 501 , 用于将待发送的用户数据帧进行 CPS复用, 同一用户数 据帧的多个 CPS - PKT连续复用;
封装模块 502, 用于对所述连续复用的多个 CPS - PKT进行封装得到信 元, 并在包含最后一个 CPS-PKT信元中填入帧尾标志, 指示接收所述用户数 据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃;
发送模块 503 , 用于将所述用户数据帧的信元向网络节点发送。
优选的, 所述复用模块 501具体用于将待发送的用户数据帧进行 SSSAR 分段和 CPS复用封装, 同一用户数据帧的多个 CPS-PKT连续复用, 复用模 块 501的内部结构包括:
SSSAR子层分段单元 5011 , 用于在 SSSAR子层将所述待发送的用户数 据帧进行分段处理后得到 SSSAR - PDU,在该用户数据帧的最后一个 SSSAR - PDU中携带有帧尾标志;
CPS复用单元 5012, 用于在 CPS层, 将所述用户数据帧分段后得到的
SSSAR - PDU封装成 CPS - PKT , 并将封装得到的 CPS - PKT连续复用至同 一 CPS复用队列, 所述最后一个 SSSAR - PDU对应 CPS - PKT均携带有帧 尾标志。 优选的,所述封装模块 502具体用于从复用模块 501得到 CPS-PDU并进 行信元封装,将复用模块传递下来的帧尾标记封装到 ΡΤΙ中, 指示接收所述用 户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃包括:
CPS-PKT封装单元 5021 ,用于在 CPS层,将所述 CPS复用队列中的 CPS - PKT封装得到一个或多个 CPS - PDU,在荷载有携带帧尾标志的 CPS - PKT 的 CPS - PDU中添加帧尾标志;
ATM封装单元 5022, 用于在 ATM层,接收来自 CPS的 CPS - PDU, 对 接收到的 CPS - PDU进行信元封装,在荷载有携带帧尾标志的 CPS - PDU的 信元中添加帧尾标志。
可选的, 亦可将 CPS-PKT封装单元 5021加入复用模块 501。
结合上述拥塞控制系统和拥塞控制装置, 本发明的实施例还提供了一种 拥塞控制方法, 使用该方法完成拥塞控制的流程如图 6所示, 包括:
步骤 601、 第一网络节点将待发送的用户数据进行 CPS复用, 同一用户 数据帧的多个 CPS-PKT连续复用;
本步骤具体在第一网络节点的 AAL2层完成, 具体的, 在 SSSAR子层将 所述待发送的用户数据进行分段处理, 得到 SSSAR-PDU; 在 CPS, 将分段处 理后的待发送数据打包, 经过 CPS复用后发送至 ATM层。 以保证同一用户 数据帧的 CPS-PKT必须连续进行 CPS复用处理。
步骤 602、第一网络节点对所述 CPS复用队列中的多个 CPS-PKT进行封 装得到信元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志;
本发明实施例中, 所述帧尾标志用于指示接收所述用户数据帧的网络节 点在拥塞控制时对该用户数据帧的进行帧丟弃。
具体的, 在 CPS , 将连续复用的多个 CPS-PKT封装得到一个或多个 CPS-PDU,在荷载有携带帧尾标志的 CPS-PKT的 CPS-PDU中添加帧尾标志; 在 ATM层, 接收来自 CPS的 CPS-PDU, 对接收到的 CPS-PDU进行信 元封装,在荷载有携带帧尾标志的 CPS-PDU的信元中添加帧尾标志, 第一网 络设备在 PTI字段中, 填入帧尾标志, 具体的, 可以设置所述信元的净荷类 型 (ΡΉ )字段的 ATM用户至用户指示(AUU )位, 以表明该信元包含所述 用户数据帧的最后一个 CPS-PKT。
本步骤中, 携带有帧尾标志的 CPS-PKT在所述 CPS复用队列的最后, CPS 复用在完成原有的 CPS 复用功能的同时, 还实现了将帧尾标志传递给 ATM层。
在原有的 ATM信元的封装的基础上,增加将帧尾标志填充在 ATM信元 的信元头中的功能, 以备确定一个用户数据帧中最后一个 ATM信元时使用。
步骤 603、 第一网络节点将所述用户数据帧的信元向第二网络节点发送; 步骤 604、 第二网络节点接收第一网络节点连续发送的归属于同一用户 数据帧的信元;
步骤 605、 第二网络节点在需要进行拥塞控制时, 查找携带有帧尾标志 的信元;
在发生信元丟失时, 第二网络节点判断需要进行拥塞控制, 进行帧丟弃。 具体的, 查看自丟失的信元起的信元的信元头, 确认信元头的 PTI内是否包 含指示该信元荷载有最后一个 CPS-PKT的帧尾标志。
步骤 606、 在检测到携带有帧尾标志的信元时, 根据该携带有帧尾标志 的信元的指示, 对该信元对应的用户数据帧进行帧丟弃;
本步骤中, 第二网络节点根据信元头中的帧尾标志确定用户数据帧对应 的最后一个 CPS-PKT所在的信元, 然后进行帧丟弃。 帧丟弃具体为, 将丟失 的信元至包含该用户数据帧帧尾标记的信元之间的信元全部丟弃, 仅保留携 带帧尾标记的信元。
下面结合附图, 对本发明的实施例二进行说明。
本发明实施例提供了一种拥塞控制系统, 其结构如图 7所示, 包括: 拥塞控制属性配置模块 701、 基于拥塞控制需要的 AAL2适配层数据处 理模块 702、 基于拥塞控制需要的 AAL2信元处理模块 703和基于帧丟弃的 拥塞控制模块 704。
其中, 拥塞控制属性配置模块 701 , 根据本发明实施例中基于帧的拥塞 控制策略(该策略指示是否具有按帧丟弃的功能)进行配置, 指示 AAL2层 是否满足基于帧丟弃的 CPS复用和 ATM交换设备是否实现帧丟弃; 基于拥 塞控制需要的 AAL2适配层数据处理模块 702, 根据拥塞控制属性配置模块 的配置信息, 此模块在完成标准的 AAL2层功能的同时, 需要保证同一用户 数据帧的 CPS-PKT必须连续进行 CPS复用处理, CPS复用在完成原有的 CPS 复用功能的同时,将帧尾标志传递给 ATM层(在向 ATM层传递的 CPS-PDU 中携带该帧尾标志) ; 基于拥塞控制需要的 AAL2信元处理模块 403 , 在原 有的 ATM信元的封装的基础上,增加将帧尾标志填充在 ATM信元头中的功 能, 以备基于帧丟弃的拥塞控制模块 704使用; 基于帧丟弃的拥塞控制模块 704 , 用于根据信元头中的帧尾标志实现按帧丟弃的功能。
拥塞控制属性配置模块 701处于网络控制器, 用于完成对整个拥塞控制 的配置; 用于完成业务的网络节点均包含基于拥塞控制需要的 AAL2适配层 数据处理模块 702和基于拥塞控制需要的 AAL2信元处理模块 703, 具体的, 基于拥塞控制需要的 AAL2适配层数据处理模块 702处于 AAL2层, 包括 SSSAR子层模块和 CPS模块两部分,基于拥塞控制需要的 AAL2信元处理模 块 703处于 ATM层; 基于帧丟弃的拥塞控制模块 704处于 ATM层, 一般集 成于 ATM交换设备中。
本发明实施例还提供了一种拥塞控制方法,结合图 7所示拥塞控制系统, 完成拥塞控制的流程如图 8所示, 包括:
步骤 801、 AAL2层 SSSAR子层模块根据配置属性进行分段处理, 如果 配置要求按帧发送, 则保证同一 ATM 虚链接的不同 SSCS-PDU 的 SSSAR-PDU不会交叉进入 CPS模块的接收队列 (接收到的 SSSAR-PDU的 队列) ;
步骤 802、 CPS模块对 SSSAR子层分段得到的 SSSAR-PDU进行 CPS-PKT 封装后进行 CPS复用, 并将 SSSAR-PDU的帧尾标记携带给基于拥塞控制需 要的 AAL2信元处理模块;
步骤 803、 基于拥塞控制需要的 AAL2信元处理模块将 ATM-SDU中携 带的帧尾标志填入 PTI字段中;
本步骤中, 基于拥塞控制需要的 AAL2信元处理模块根据 CPS复用报文 携带的信息进行信元头的构造, 将 AUU字段封装入 PTI字段的最低位。 步骤 804、 ATM交换设备根据配置属性进行拥塞控制, 如果配置要求按 帧进行拥塞控制, 则根据 PTI字段的 AUU标志实现帧丟弃功能;
本步骤中, ATM交换设备内的基于帧丟弃的拥塞控制模块根据 ATM虚 链接配置属性选择是否执行帧丟弃的拥塞控制策略。 当配置属性为基于帧丟 弃时, 通过识别信元头中的尾包标志进行信元的丟弃处理。 此处理方式可以 釆用 EPD/PPD处理原则。
按照上述方法进行 AAL2 层协议处理后, 同一用户数据帧对应的 CPS-PKT将被连续复用, 并在 ATM网络中被连续传输, 如图 9所示, 这样 就具备了基于帧丟弃信元的条件,另外再将 CPS-PKT的帧尾标志传入信元封 装后, 就能实现基于帧丟弃信元的拥塞控制方法了。
其中, 步骤 801具体如图 10所示, 包括:
步骤 8011、 接收到业务报文, 如果需要实现按帧流控策略, 转入步骤 8012, 否则, 转入步骤 8013;
步骤 8012、 获取对应 ATM虚链接的 CPS模块接收队列的入队权限; 步骤 8013、 对报文进行分段处理;
步骤 8014、 将当前分段送入对应的 CPS接收队列;
步骤 8015、 如果是最后一个分段, 转入步骤 8016, 如果不是最后一个分 段, 转入步骤 8013;
步骤 8016、 如果已获取了 CPS模块入队权限, 转入步骤 8017 , 如果没 有获取, 转入步骤 8018;
步骤 8017、 释放 CPS模块接收队列入队权限;
步骤 8018、 结束。
步骤 802具体如图 11所示, 包括:
步骤 8201、 收到 CPS-SDU , 封装 CPS-PKT 头部, 如果是最后一个 SSSAR-PDU, 则置当前 CPS-PKT尾包标志为 1 , 否则置 0, 转步骤 8202;
步骤 8202、 进行 文复用处理, 转入步骤 8203;
步骤 8203、 根据当前收到的 CPS-PKT的帧尾标志和之前的帧尾标志更 新当前 CPS-PDU的帧尾标志, 记录当前 CPS-PDU是否包含帧尾标志, 如果 包含, 则标志值 1 ;
步骤 8204、 当前 CPS-PDU是否满足完成复用条件, 如果满足, 转入步 骤 8205; 否则, 执行步骤 8212;
步骤 8205、当前 CPS-PDU是否包含帧尾标志,如果包含,转入步骤 8206, 如果不包含, 转入步骤 8207;
步骤 8206、 将 AUU标志置 0, 转入步骤 8208;
步骤 8207、 将 AUU标志置 1 , 转入步骤 8208;
步骤 8208、 将当前 CPS-PDU发送给入 ATM队列, 转入步骤 8209; 步骤 8209、 当前 CPS-PKT的数据是否已全部复用完成, 如果是, 转入 步骤 8210, 如果不是, 转入步骤 8211 ;
步骤 8210、 清空当前 CPS-PDU的帧尾标志, 转入步骤 8213;
步骤 8211、 当前 CPS-PKT是否是尾包, 如果是, 将当前 CPS-PDU的尾 包标志置 1 , 如果不是, 则置 0, 转入步骤 8212;
步骤 8212、 等待继续复用;
步骤 8213、 结束。
本发明的实施例提供的拥塞控制系统, 能够与本发明的实施例所提供的 拥塞控制方法相结合, 网络节点间使用 AAL2类型承载数据, 将待发送的用 户数据进行 CPS复用, 对同一条用户数据的多个 CPS-PKT进行连续复用, 在包含所述用户数据对应的最后一个 CPS-PKT的 ATM信元的 PTI字段中, 填入帧尾标志,将所述用户数据的 ATM信元依序向网络节点发送,所述用户 数据的所有 ATM信元在网络中连续按序传输,所述 PTI字段携带有帧尾标志 的 ATM信元在所述用户数据所有 ATM信元的最后,指示接收所述用户数据 的所述网络节点在拥塞控制时对该用户数据的携带有帧尾标记的 ATM信元 之前的 ATM信元执行包丟弃, 实现了在发生丟包待需要拥塞控制的情况下, 对一条用户数据对应的全部信元进行的帧丟弃, 解决了 AAL2承载模式下网 络拥塞的问题。 对提高网络传输利用率、 降低网络拥塞风险有明显的作用, 是一种全新的解决方案。 本领域普通技术人员可以理解上述实施例的全部或部分步骤可以使用计 算机程序流程来实现,所述计算机程序可以存储于一计算机可读存储介质中, 所述计算机程序在相应的硬件平台上(如系统、 设备、 装置、 器件等)执行, 在执行时, 包括方法实施例的步骤之一或其组合。
可选地, 上述实施例的全部或部分步骤也可以使用集成电路来实现, 这 些步骤可以被分别制作成一个个集成电路模块, 或者将它们中的多个模块或 步骤制作成单个集成电路模块来实现。 这样, 本发明不限制于任何特定的硬 件和软件结合。
上述实施例中的各装置 /功能模块 /功能单元可以釆用通用的计算装置来 实现, 它们可以集中在单个的计算装置上, 也可以分布在多个计算装置所组 成的网络上。
上述实施例中的各装置 /功能模块 /功能单元以软件功能模块的形式实现 并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。 上述提到的计算机可读取存储介质可以是只读存储器, 磁盘或光盘等。
任何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可轻易想 到变化或替换, 都应涵盖在本发明的保护范围之内。 因此, 本发明的保护范 围应以权利要求所述的保护范围为准。
工业实用性 本发明实施例提供了一种拥塞控制方法和系统, 将待发送的用户数据帧 进行 CPS复用, 同一用户数据帧的多个 CPS-PKT连续复用, 对所述 CPS复 用队列中的多个 CPS-PKT进行封装得到信元, 并在所述 CPS复用队列的封 装后得到的包含最后一个 CPS-PKT的信元中填入帧尾标志,指示接收所述用 户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃, 并将所述 用户数据帧的信元向网络节点发送, 实现了在发信元丟失的情况下, 对同属 于一个数据帧的其他信元的帧丟弃, 解决了 AAL2承载模式下网络拥塞的问 题。

Claims

权 利 要 求 书
1、 一种拥塞控制方法, 网络节点间使用异步传输模式适配层 2 ( AAL2 ) 类型传输数据, 包括:
将待发送的用户数据帧进行公共部分子层 ( CPS ) 复用, 同一用户数据 帧的多个 CPS包( CPS-PKT )连续复用;
对所述连续复用的多个 CPS-PKT进行封装得到信元,并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示接收所述用户数据帧的网络节点在拥 塞控制时对该用户数据帧的进行帧丟弃;
将所述用户数据帧的信元向网络节点发送。
2、 根据权利要求 1所述的拥塞控制方法, 其中, 所述将待发送的用户数 据帧进行 CPS复用包括:
在特定业务分段和重装子层(SSSAR子层)将所述待发送的用户数据帧 进行分段处理后得到 SSSAR子层协议数据单元(SSSAR-PDU ) , 在该用户 数据帧的最后一个 SSSAR-PDU中携带有帧尾标志;
在 CPS 层, 将所述用户数据帧分段后得到的 SSSAR-PDU 封装成
CPS-PKT, 并将封装得到的 CPS-PKT连续复用, 所述最后一个 SSSAR-PDU 对应的 CPS-PKT均携带有帧尾标志。
3、 根据权利要求 1所述的拥塞控制方法, 其中, 对所述连续复用的多个 CPS-PKT进行封装得到信元,并在所述 CPS复用队列的封装后得到的包含最 后一个 CPS-PKT的信元中填入帧尾标志的步骤包括:
在 CPS 层, 将所述连续复用的多个 CPS-PKT 封装得到一个或多个 CPS-PDU,在荷载有携带帧尾标志的 CPS-PKT的 CPS-PDU中添加帧尾标志; 在异步传输模式(ATM )层, 接收来自 CPS层的 CPS-PDU, 对接收到 的 CPS-PDU进行信元封装, 在荷载有携带帧尾标志的 CPS-PDU的信元中添 加帧尾标志。
4、 根据权利要求 3所述的拥塞控制方法, 其中, 所述在荷载有携带帧尾 标志的 CPS-PDU的信元中添加帧尾标志包括:
设置所述信元的净荷类型 (PTI )字段的 ATM用户至用户指示 (AUU ) 位, 以表明该信元包含所述用户数据帧的最后一个 CPS-PKT。
5、 根据权利要求 1所述的拥塞控制方法, 其中, 所述将所述用户数据帧 的信元向网络节点发送的步骤之后, 还包括:
所述网络节点接收连续发送的归属于同一用户数据帧的信元;
在需要进行拥塞控制时, 查找携带有帧尾标志的信元;
在检测到携带有帧尾标志的信元时, 根据该携带有帧尾标志的信元的指 示, 对该信元对应的用户数据帧进行帧丟弃。
6、 一种拥塞控制装置, 适用于使用异步传输模式适配层 2 ( AAL2 )类 型传输数据的网络节点, 包括:
复用模块, 其设置为: 将待发送的用户数据帧进行 CPS复用, 同一用户 数据帧的多个 CPS-PKT连续复用;
封装模块, 其设置为: 对所述连续复用的多个 CPS-PKT进行封装得到信 元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示接收所述用户 数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟弃;
发送模块, 其设置为: 将所述用户数据帧的信元向网络节点发送。
7、 根据权利要求 6所述的拥塞控制装置, 其中, 所述复用模块包括:
SSSAR子层分段单元, 设置为: 在 SSSAR子层将所述待发送的用户数 据帧进行分段处理后得到 SSSAR-PDU , 在该用户数据帧的最后一个 SSSAR-PDU中携带有帧尾标志;
CPS 复用单元, 设置为: 在 CPS层, 将所述用户数据帧分段后得到的 SSSAR-PDU封装成 CPS-PKT , 并将封装得到的 CPS-PKT连续复用, 所述最 后一个 SSSAR-PDU对应 CPS-PKT均携带有帧尾标志。
8、 根据权利要求 6所述的拥塞控制装置, 其中, 所述封装模块包括:
CPS-PKT封装单元, 设置为: 在 CPS层, 将所述连续复用的 CPS-PKT 封装得到一个或多个 CPS-PDU, 在荷载有携带帧尾标志的 CPS-PKT 的 CPS-PDU中添加帧尾标志;
ATM封装单元, 设置为: 在 ATM层, 接收来自 CPS的 CPS-PDU, 对 接收到的 CPS-PDU进行信元封装, 在荷载有携带帧尾标志的 CPS-PDU的信 元中添加帧尾标志。
9、 一种拥塞控制系统, 包括第一网络节点和第二网络节点, 所述第一网 络节点和所述第二网络节点间使用 AAL2类型传输数据;
所述第一网络节点设置为: 将待发送的用户数据帧进行 CPS复用, 同一 用户数据帧的多个 CPS-PKT连续复用, 对所述连续复用的多个 CPS-PKT进 行封装得到信元, 并在包含最后一个 CPS-PKT的信元中填入帧尾标志, 指示 接收所述用户数据帧的网络节点在拥塞控制时对该用户数据帧的进行帧丟 弃, 并将所述用户数据帧的信元向所述第二网络节点发送。
10、 根据权利要求 9所述的拥塞控制系统, 其中,
所述第二网络节点设置为: 接收所述第一网络节点连续发送的归属于同 一用户数据帧的信元, 在需要进行拥塞控制时, 查找携带有帧尾标志的信元, 并在检测到携带有帧尾标志的信元时,根据该携带有帧尾标志的信元的指示, 对该信元对应的用户数据帧进行帧丟弃。
PCT/CN2012/075281 2011-06-30 2012-05-10 拥塞控制方法、装置和系统 WO2013000342A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110182377.X 2011-06-30
CN201110182377.XA CN102361473B (zh) 2011-06-30 2011-06-30 拥塞控制方法、装置和系统

Publications (1)

Publication Number Publication Date
WO2013000342A1 true WO2013000342A1 (zh) 2013-01-03

Family

ID=45586728

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/075281 WO2013000342A1 (zh) 2011-06-30 2012-05-10 拥塞控制方法、装置和系统

Country Status (2)

Country Link
CN (1) CN102361473B (zh)
WO (1) WO2013000342A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102361473B (zh) * 2011-06-30 2016-12-21 中兴通讯股份有限公司 拥塞控制方法、装置和系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1252911A (zh) * 1997-04-16 2000-05-10 Ntt移动通信网株式会社 码分多址通信方法
CN1305293A (zh) * 1999-11-12 2001-07-25 阿尔卡塔尔公司 异步传输模式适应层2接续的阻塞控制
CN102361473A (zh) * 2011-06-30 2012-02-22 中兴通讯股份有限公司 拥塞控制方法、装置和系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100298356B1 (ko) * 1998-11-13 2001-11-22 윤종용 에이티엠(atm)망의에이티엠적응계층(all)2처리장치및방법
KR100446234B1 (ko) * 2000-12-21 2004-08-30 엘지전자 주식회사 차세대 통신 시스템에서 멀티 캐스트용 aal2 스위치

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1252911A (zh) * 1997-04-16 2000-05-10 Ntt移动通信网株式会社 码分多址通信方法
CN1305293A (zh) * 1999-11-12 2001-07-25 阿尔卡塔尔公司 异步传输模式适应层2接续的阻塞控制
CN102361473A (zh) * 2011-06-30 2012-02-22 中兴通讯股份有限公司 拥塞控制方法、装置和系统

Also Published As

Publication number Publication date
CN102361473A (zh) 2012-02-22
CN102361473B (zh) 2016-12-21

Similar Documents

Publication Publication Date Title
US5917828A (en) ATM reassembly controller and method
KR100823785B1 (ko) 시스템 패브릭에서의 개방 루프 정체 제어를 위한 방법,장치, 제품 및 시스템
US7609702B2 (en) Interworking between cell and packet based networks
EP3573297B1 (en) Packet processing method and apparatus
US7602809B2 (en) Reducing transmission time for data packets controlled by a link layer protocol comprising a fragmenting/defragmenting capability
US20070058543A1 (en) ATM over ethernet scheduler
US9065764B2 (en) Method, apparatus and system for maintaining quality of service QoS
EP1766876A2 (en) Technique for transferring data over a packet switched network
JP2001308928A (ja) 通信システムのアクセスネットワークにおける基地局とアクセスネットワークコントローラとの間の送信方法
WO2009129721A1 (zh) 一种业务数据的发送、接收方法、装置和系统
WO2011143968A1 (zh) 拥塞控制信息的通知方法和系统
WO2011023072A1 (zh) 一种微波链路的业务调度方法、装置及系统
WO2013000342A1 (zh) 拥塞控制方法、装置和系统
US11902172B2 (en) Device and method for transferring identification and/or data flow control information between devices
WO2010102453A1 (zh) 提高中继网络系统性能的方法、通信系统及协议实体
US20220407742A1 (en) Time-sensitive transmission of ethernet traffic between endpoint network nodes
US20100329245A1 (en) Transparent Mapping of Cell Streams to Packet Services
CN101019406B (zh) 一种在通信设备内部二次封装信息的方法及装置
WO2009132542A1 (zh) 报文传输的方法及装置
US20130028264A1 (en) Packet reassembly processing
JP5666420B2 (ja) パケット多重伝送装置及び方法
CN101478498B (zh) 一种基于fpga的atm信元交换装置及方法
Han et al. Simulation study of AAL type 2
CN116781611A (zh) 一种基于灵活以太网和确定性网络的数据交换设备和方法
JP2005303516A (ja) 通信装置および通信方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12804941

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12804941

Country of ref document: EP

Kind code of ref document: A1