WO2011144061A2 - 移动网络小区的拥塞检测方法和装置 - Google Patents

移动网络小区的拥塞检测方法和装置 Download PDF

Info

Publication number
WO2011144061A2
WO2011144061A2 PCT/CN2011/074451 CN2011074451W WO2011144061A2 WO 2011144061 A2 WO2011144061 A2 WO 2011144061A2 CN 2011074451 W CN2011074451 W CN 2011074451W WO 2011144061 A2 WO2011144061 A2 WO 2011144061A2
Authority
WO
WIPO (PCT)
Prior art keywords
cell
trip time
round
congestion
loss rate
Prior art date
Application number
PCT/CN2011/074451
Other languages
English (en)
French (fr)
Other versions
WO2011144061A3 (zh
WO2011144061A9 (zh
Inventor
蒋铭
杨文进
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2011/074451 priority Critical patent/WO2011144061A2/zh
Priority to CN201180000704.3A priority patent/CN102893654B/zh
Publication of WO2011144061A2 publication Critical patent/WO2011144061A2/zh
Publication of WO2011144061A3 publication Critical patent/WO2011144061A3/zh
Publication of WO2011144061A9 publication Critical patent/WO2011144061A9/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control

Definitions

  • the present invention relates to the field of mobile communication technologies, and in particular, to a congestion detection method and apparatus for a mobile network cell. Background technique
  • Bandwidth management includes admission control, traffic classification, traffic policing, congestion avoidance, traffic shaping, and queue scheduling.
  • the congestion detection and control of the mobile network cell is through a Base Station Controller (BSC)/Radio Network Controller (RNC) network element or a Gateway GPRS Support Node (Gateway GPRS Support Node, referred to as GGSN)
  • BSC Base Station Controller
  • RNC Radio Network Controller
  • GGSN Gateway GPRS Support Node
  • the BSC/RNC network element is responsible for congestion detection, control, and service data flow differentiation.
  • the BTS/NodeB (3G base station) network element reports the use of the cell resource to the RNC network element, and the BSC/RNC network element determines the congestion degree of the cell according to the actual use of the cell resource, and the BSC/RNC network according to different congestion levels.
  • the meta implements the corresponding countermeasures.
  • the BSC/RNC network element is only responsible for congestion detection, and then reports the cell congestion information to the GGSN network element.
  • the GGSN network element is responsible for congestion control and service data flow differentiation.
  • the BSC/RNC network element can be directly reported to the GGSN network element through the Iu-Ps interface, or it can be reported to the Operation Support System (OSS) first. Then, the OSS system collects and analyzes the congestion feature, and finally the GGSN network element. Congestion control is performed on it.
  • OSS Operation Support System
  • the BSC/RNC network element is responsible for congestion detection and control, and the GGSN network element is responsible for the service data flow area.
  • the packets are then tagged for different service data packets, and the BSC/RNC network elements implement differentiated services based on these tags.
  • the technical problem to be solved by the present invention is to provide a congestion detection method and apparatus for a mobile network cell, which can implement congestion detection and data offload through a core network element.
  • the embodiment of the present invention adopts the following technical solutions:
  • a congestion detection method for a mobile network cell includes:
  • a core network element including:
  • An acquiring unit configured to acquire a round-trip time of a cell data packet in real time
  • a comparison determining unit configured to compare the round-trip time with a preset congestion round-trip time threshold, and if the round-trip time is greater than the congestion round-trip time threshold, determine the cell overload congestion, if If the round-trip time is less than the congestion round-trip time threshold, it is determined that the cell is not congested.
  • the congestion detection of the mobile network cell is implemented by using the RTT detection of the data packet and comparing the RTT with the preset congestion round-trip time threshold, the method making the core network
  • the network element can not only complete the congestion detection and control, but also fully utilize the core network to distinguish the service data flow; and, unlike the prior art, the method does not involve the change of the existing mobile network architecture and interface, so the BSC is not needed.
  • FIG. 1 is a flowchart of a congestion detection method of a mobile network cell according to Embodiment 1 of the present invention
  • FIG. 2 is a schematic diagram of a method for acquiring a cell RTT according to an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for detecting congestion of a mobile network cell according to Embodiment 2 of the present invention
  • FIG. 4 is a schematic diagram of hierarchical sampling of data services according to an embodiment of the present invention
  • FIG. 5 is a signaling diagram of an Intra SGSN process according to an embodiment of the present invention.
  • FIG. 6 is a signaling diagram of an Inter SGSN process according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of congestion detection by RTT according to Embodiment 2 of the present invention.
  • FIG. 8 is a flowchart of a method for detecting congestion of a mobile network cell according to Embodiment 3 of the present invention
  • FIG. 9 is a schematic diagram of congestion detection by a packet loss rate according to Embodiment 3 of the present invention.
  • FIG. 10 is a flowchart of a congestion detection method for a mobile network cell according to Embodiment 4 of the present invention
  • FIG. 11 is a schematic diagram of congestion detection by RTT and bandwidth joint according to Embodiment 4 of the present invention
  • FIG. 12 is a core network according to Embodiment 5 of the present invention; Schematic diagram of the structure of the network element;
  • FIG. 13 is a schematic structural diagram of an acquiring unit according to Embodiment 5 of the present invention.
  • FIG. 14 is a schematic structural diagram 2 of a network element of a core network according to Embodiment 5 of the present invention. detailed description
  • the embodiment of the invention provides a method and a device for detecting congestion of a mobile network cell, which can implement congestion detection and data offloading through a core network element.
  • Embodiment 1
  • the embodiment of the invention provides a congestion detection method for a mobile network cell. As shown in FIG. 1, the method includes:
  • the core network element acquires a round trip time (RTT) of the cell data packet in real time;
  • the core network element may be an SGSN, a GGSN, a SAE Gateway, or the like.
  • the method for obtaining the RTT of the data packet is:
  • the core network element After receiving the downlink TCP data packet of the Packet Data Network (PDN), the core network element performs four layers of parsing, for example, retransmission or out-of-order judgment, and statistics retransmission packets and chaos. The number of sequence packets is used to obtain the packet loss rate of the TCP data packet. Meanwhile, the core network element records the time at which the TCP data is received.
  • PDN Packet Data Network
  • the user equipment After receiving the data packet of the TCP data packet, the user equipment (User Equipment, UE for short) sends an acknowledgement packet (ack) to the core network element, and the core network element records the acknowledgement packet (ack) At time Ta, the RTT value of the TCP data message is equal to Ta-Tr.
  • the core network element compares the RTT with a preset congestion round-trip time threshold, and if the RTT is greater than the congestion round-trip time threshold, determining that the cell is overloaded, if the RTT is smaller than The congestion round-trip time threshold determines that the cell is not congested.
  • the congestion detection of the mobile network cell is implemented by detecting and judging the data packet, and the method may be implemented on the core network element such as the SGSN, the GGSN, and the SAE Gateway, where the core network element is It can complete congestion detection and control, and can fully utilize the core network's ability to differentiate service data streams. Moreover, unlike existing technologies, this method does not involve changes to existing mobile network architectures and interfaces, so BSC/RNC and BTS/NodeB upgrade, easy and fast implementation.
  • TCP data packets exceeding the rate limit will be buffered. Queues, due to the increase of cache time, the RTT time becomes longer, and with the continuous increase of traffic, the depth of the cache queue is deeper and deeper, and the RTT value will continue to increase. If the traffic continues to increase, the buffer queue will overflow. The subsequent TCP packets will be discarded and packet loss will occur. Therefore, it can be determined whether the cell is congested by checking the change of the cell RTT value.
  • the embodiment of the present invention provides a congestion detection method for a mobile network cell.
  • the method performs congestion detection on a cell by detecting and judging the RTT. As shown in FIG. 2, the method includes:
  • the core network element acquires the TCP data of the cell in real time.
  • the method for obtaining the RTT of the TCP data packet of the cell is described in detail in step 101 of the first embodiment and FIG. 2, and is not described in this embodiment.
  • the embodiment of the present invention detects whether a cell is congested according to an RTT change of a TCP data packet, and each mobile cell necessarily has multiple TCP connections. Therefore, the RTT of the cell needs to be acquired. Moreover, it is necessary to minimize the time required for the core network element device to extract and analyze data, so as to reduce the impact on the processing capability of the core network element device due to detecting cell congestion.
  • a statistical sampling method is introduced to perform measurement and statistics of RTT. That is, the sampled value of the round-trip time and/or the sampled value of the packet loss rate may be extracted according to the preset sampling interval according to the number of service types of the TCP data packet of the cell and the service usage ratio.
  • sampling interval is determined as:
  • the sampling method can be divided into many types according to the distribution of the sampling time interval.
  • periodic sampling random additional sampling and Poisson sampling can be used.
  • Periodic sampling is a sampling every fixed time.
  • the sampling intervals of random oversampling are independent of each other and subject to a distribution function.
  • Poisson sampling whose time interval is consistent with the Poisson distribution.
  • the sample is extracted by the above method, and the RTT of the cell is obtained according to the sampled value of the RTT described above, that is, the RTT average of the samples is taken as the RTT of the cell.
  • the packet loss rate of these samples can also be counted as the packet loss rate of the cell.
  • the sampling strategy is:
  • the sample value of the RTT may be extracted according to a preset sampling interval according to the service usage ratio of each service type in the data packet, for example, layering.
  • Sampling method During sampling, the TCP data packets in the cell are divided into layers that do not intersect each other according to the service type, and then a certain number of observation units are randomly selected from each layer according to a certain ratio, and combined to form a sample. With this method, the representativeness of the sample is better, and the sampling error is smaller.
  • TCP data packets are classified into m-type services, such as common web browsing, streaming media, point-to-point (P2P), and email mail.
  • the proportion of each service is not Similarly, a certain number of TCP data packets are extracted from each service according to the proportion of each service. Assuming that n TCP data packet samples are extracted from m services, the RTT value of the cell is considered to be equal to this! ! One! ⁇ RTT mean of data packet samples; Similarly, the packet loss rate of the cell is equal to the packet loss rate of the n TCP data packet samples.
  • Each operator's mobile service distribution is different.
  • the service ratio can be configured according to the actual application. This configuration is an average statistical distribution.
  • DPI Deep Packet Inspection
  • the core network element can use the DPI function to parse the packet. According to the analysis result, the number of service types of the cell and the usage ratio of each service can be obtained.
  • DPI can reflect the real situation of service distribution under the cell in real time.
  • DPI is an application layer-based traffic detection and control technology.
  • IP packet TCP or UDP data stream passes through a system with DPI function, the system reads the contents of the packet payload deeply into the OSI seven-layer protocol.
  • the application layer information is identified and parsed to obtain the entire service information.
  • the core network element such as the GGSN, already has the DPI function. Therefore, the DPI function of the core network element can be fully utilized to parse the message, thereby obtaining the service type and service distribution of the cell in real time.
  • the DPI function is better, it can get the business proportion more accurately, so the sample is better represented.
  • the SGSN should perform a process of whether the cell has changed.
  • the following embodiments use only the Routing (Routing Area Update, RAU) process for illustration, and the processes such as Serving RNS Relocation have similar processing.
  • the SGSN is an intra-SGSN (intra SGSN) process, that is, the user equipment UE switches between routing areas belonging to the same SGSN, and the SGSN determines whether the user equipment UE leaves the cell that performs congestion detection, if If the user equipment UE is a cell that performs congestion detection, and the user equipment UE is a sample user for congestion detection, the user equipment UE should be deleted from the sample, and the traffic generated by the user equipment UE is not calculated in the cell usage bandwidth.
  • intra SGSN intra-SGSN
  • Inter SGSN inter-SGSN
  • the process of inter-SGSN is performed, that is, the user equipment UE switches between routing areas belonging to different SGSNs, and the old SGSN determines whether the user equipment UE leaves or not to perform congestion.
  • the detected cell if the user equipment UE leaves the cell that performs congestion detection, and the user is a sample user of congestion detection, the user equipment UE is deleted from the sample, and the traffic generated by the user equipment UE is not calculated in the cell. Use bandwidth.
  • the core network element compares the RTT with a preset congestion round-trip time threshold RTTa. If the round-trip time is greater than the congestion round-trip time threshold RTTa, the cell is determined to be overloaded. If the round-trip time is less than the congestion round-trip time threshold RTTa, the cell is determined not to be Congested.
  • the congestion round-trip time threshold RTTa is a preset value, and the core network element detects the RTT of the cell in real time, and compares with the RTTa, if the RTT of the cell is greater than the RTTa, it determines that the cell is overloaded.
  • the core network element When determining that the cell is overloaded, the core network element sends a cell overload signal.
  • the congestion detection of the mobile network cell is implemented by detecting and judging the RTT, and the method can be implemented on the core network element such as the SGSN, the GGSN, and the SAE Gateway, and the core network element can Completing congestion detection and control, and fully utilizing the core network's ability to differentiate service data streams; and, unlike existing technologies, this method does not involve changes to existing mobile network architectures and interfaces, and thus does not require BSC/RNC and BTS. /NodeB upgrade, easy to implement and fast.
  • An embodiment of the present invention provides a congestion detection method for a mobile network cell.
  • the method implements congestion detection on a cell by detecting and determining a packet loss rate in real time.
  • the method includes: 301.
  • a core network The real-time acquisition of the packet loss rate of the TCP data packet of the cell;
  • step 101 For the method for obtaining the packet loss rate of the TCP data packet of the cell, refer to step 101 in the first embodiment.
  • the detailed description of FIG. 2 and the embodiment shown in FIG. 2 are not described in detail.
  • the packet loss rate of the sample is used as the packet loss rate of the cell. If the packet rate is greater than the packet loss threshold, the cell is determined to be overloaded. If the packet loss rate is less than the packet loss threshold, the cell is not congested.
  • the packet loss rate threshold is a preset value, and the core network element detects the packet loss rate of the cell in real time, and compares with the packet loss rate threshold. If the RTT of the cell is greater than RTTa, the cell is determined to be overloaded. congestion.
  • the core network element When determining that the cell is overloaded, the core network element sends a cell overload signal.
  • the specific process of the above steps 302 and 303, as shown in FIG. 9, illustrates the process of performing cell congestion determination according to the change of the packet loss rate.
  • the cell service traffic is small in the uncongested state, and the cell does not have packet loss or the packet loss rate is small;
  • the congestion detection of the mobile network cell is implemented by real-time detection and judgment of the packet loss rate, and the method can be implemented on the core network element such as the SGSN, the GGSN, and the SAE Gateway, and the core network is implemented.
  • the element can not only complete the congestion detection and control, but also fully utilize the core network's ability to distinguish the service data flow; and, unlike the existing technology, the method does not involve the change of the existing mobile network architecture and interface, so the BSC/ is not needed.
  • RNC and BTS/NodeB upgrades are easy and fast to implement.
  • Embodiment 4 An embodiment of the present invention provides a congestion detection method for a mobile network cell.
  • the method implements congestion detection on a cell by performing real-time detection and judgment on the RTT and the bandwidth. As shown in FIG. 10, the method includes:
  • the network element of the core network obtains the RTT of the TCP data packet and the bandwidth of the cell in real time.
  • the method for obtaining the RTT of the TCP data packet is referred to the foregoing embodiment, and details are not described herein.
  • the total bandwidth of the cell may be static. Configuration can also be obtained in real time in real time. When the total traffic of the user in the cell is greater than the total bandwidth of the cell, the RTT of the TCP connection will increase.
  • the core network element sends a cell overload signal.
  • Figure 11 illustrates the process of RTT and bandwidth joint cell congestion judgment.
  • the congestion detection of the mobile network cell is implemented by jointly detecting the RTT and the bandwidth of the cell, and the method may be implemented on the core network element such as the SGSN, the GGSN, and the SAE Gateway, where the core network element is It can complete congestion detection and control, and can fully utilize the core network's ability to distinguish service data streams; and, unlike existing technologies, this method does not involve The existing mobile network architecture and interface changes, so no BSC/RNC and BTS/NodeB upgrades are required, and the implementation is convenient and fast.
  • the embodiment of the present invention further provides a core network element that applies the congestion detection method of the mobile network cell according to the foregoing embodiment.
  • the core network element includes:
  • the obtaining unit 11 is configured to acquire a round-trip time of the cell data packet in real time
  • the comparison judging unit 12 is configured to compare the round-trip time with a preset congestion round-trip time threshold, and if the round-trip time is greater than a preset congestion round-trip time threshold, determine that the cell is overloaded with congestion, if If the round-trip time is less than the preset congestion round-trip time threshold, it is determined that the cell is not congested.
  • the obtaining unit 11 includes: a sampling module 111 and a computing module 112.
  • the sampling module 111 is configured to: according to the service usage ratio of each service type in the data packet, extract a sample value of the round-trip time according to a preset sampling interval; and the calculating module 112 is configured to obtain the sampling value according to the round-trip time Round trip time.
  • the sampling method of the preset sampling interval is: periodic sampling, random additional sampling and Poisson sampling. Further, as shown in FIG. 14, the acquiring unit 11 further includes:
  • the service ratio configuration module 113 is configured to obtain a service usage ratio of each service type in the data packet by using a pre-configuration or a deep packet parsing function to parse the packet.
  • the acquiring unit 11 is further configured to acquire the bandwidth of the cell in real time; when the smooth fluctuation is maintained, determine that the cell is overloaded.
  • the acquiring unit 11 is further configured to acquire a packet loss rate of the cell data packet in real time; and if the packet loss rate is greater than the packet loss rate threshold, determine that the cell is overloaded, if Place If the packet loss rate is less than the packet loss rate threshold, it is determined that the cell is not congested.
  • the core network element further includes:
  • the overload prompting unit 13 is configured to send a cell overload signal when determining that the cell is overloaded. It should be noted that the functions of the core network element may be implemented by a core network element such as an SGSN, a GGSN, or a SAE Gateway.
  • the congestion detection of the mobile network cell is implemented by detecting and judging the RTT of the data packet, and the method may be implemented on the core network element such as the SGSN, the GGSN, and the SAE Gateway, and the core network is implemented.
  • the network element can not only perform congestion detection and control, but also fully utilize the core network's ability to distinguish service data streams.
  • the method does not involve changes to the existing mobile network architecture and interfaces, so the BSC is not needed. /RNC and BTS/NodeB upgrades are easy and fast to implement.
  • the present invention can be implemented by means of software plus necessary general hardware, and of course, by hardware, but in many cases, the former is a better implementation. .
  • the technical solution of the present invention which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a readable storage medium, such as a floppy disk of a computer.
  • a hard disk or optical disk or the like includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments of the present invention.

Landscapes

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

Description

移动网络小区的拥塞检测方法和装置 技术领域
本发明涉及移动通信技术领域， 尤其涉及一种移动网络小区的拥塞检测 方法和装置。 背景技术
随着无线宽带业务的广泛应用以及智能手机用户量的激增， 移动网络的 小区的拥塞状况日益严重， 这主要体现在用户对带宽的需求与网络可提供的 带宽之间的矛盾。
为了緩解这一矛盾， 我们应该对带宽资源进行有效管理， 提高带宽资源 的利用效率。 带宽管理包括准入控制、 流分类、 流量监管、 拥塞避免、 流量 整形、 队列调度等。
目前， 移动网络小区的拥塞检测和控制是通过基站控制器 (Base Station Controller, 简称 BSC)/无线网络控制器（ Radio Network Controller, 简称 RNC ) 网元或者网关 GPRS支持节点 （ Gateway GPRS Support Node , 简称 GGSN ) 网元来实现的， 主要有以下三种方式：
一、 BSC/RNC 网元负责拥塞检测、 控制和业务数据流区分。 其中 BTS/NodeB ( 3G基站 )网元向 RNC网元上报小区资源的使用情况， BSC/RNC 网元依据小区资源的实际使用情况来判断小区的拥塞程度， 根据不同的拥塞 程度， BSC/RNC网元执行相应的对策。
二、 BSC/RNC网元只负责拥塞检测， 然后把小区拥塞信息上报给 GGSN 网元， 由 GGSN网元负责拥塞控制和业务数据流区分。 BSC/RNC网元可以通 过 Iu-Ps接口直接上报给 GGSN网元，也可以先上报给运营支撑系统（ Operating Support System, 简称 OSS ), 然后由 OSS系统统计和分析拥塞特征， 最终在 GGSN网元上进行拥塞控制。
三、 BSC/RNC网元负责拥塞检测和控制， GGSN网元负责业务数据流区 分， 然后针对不同业务数据报文进行标记， BSC/RNC网元依据这些标记来实 施差异化服务。
现有技术的技术方案至少存在以下问题： 采用第一种实现方式时， BSC/RNC网元本身并不适合进行报文识别和解析， 实现业务数据流区分的能 力比较有限； 采用第二种实现方式和第三种实现方式都涉及到接口更改， 需 要对相关接口进行标准化工作， 在多厂商设备组网的情况下实施困难。 发明内容
本发明所要解决的技术问题在于提供一种移动网络小区的拥塞检测方法 和装置， 能够通过核心网网元实现拥塞检测和数据分流。
为解决上述技术问题， 本发明实施例采用如下技术方案：
一种移动网络小区的拥塞检测方法， 包括：
实时地获取小区数据报文的往返时间；
将所述往返时间与预设的拥塞往返时间门限值相比较， 若所述往返时间 大于所述拥塞往返时间门限值， 则判断所述小区过载拥塞， 若所述往返时间 小于所述拥塞往返时间门限值， 则判断所述小区未拥塞。
一种核心网网元， 包括：
获取单元， 用于实时地获取小区数据报文的往返时间；
比较判断单元， 用于将所述往返时间与预设的拥塞往返时间门限值相比 较， 若所述往返时间大于所述拥塞往返时间门限值， 则判断所述小区过载拥 塞， 若所述往返时间小于所述拥塞往返时间门限值， 则判断所述小区未拥塞。
在本实施例的技术方案中， 通过对数据报文的 RTT检测， 并根据 RTT与 预设的拥塞往返时间门限值相比较的结果， 实现对移动网络小区的拥塞检测， 该方法使得核心网网元既能够完成拥塞检测和控制， 又能充分发挥核心网进 行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉及现有移动 网络架构和接口的更改，因此不需要 BSC/RNC和基站收发台（ Base Transceiver Station, 简称 BTS ) /NodeB升级， 实施方便、 快捷。 附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案， 下面将对实 施例描述中所需要使用的附图作简单地介绍， 显而易见地， 下面描述中的附 图仅仅是本发明的一些实施例， 对于本领域普通技术人员来讲， 在不付出创 造性劳动的前提下， 还可以根据这些附图获得其他的附图。
图 1为本发明实施例一中移动网络小区的拥塞检测方法的流程图； 图 2为本发明实施例中获取小区 RTT的方法示意图；
图 3为本发明实施例二中移动网络小区的拥塞检测方法的流程图； 图 4为本发明实施例中对数据业务进行分层抽样的示意图；
图 5为本发明实施例中 Intra SGSN流程的信令图；
图 6为本发明实施例中 Inter SGSN流程的信令图；
图 7为本发明实施例二中通过 RTT进行拥塞检测的示意图；
图 8为本发明实施例三中移动网络小区的拥塞检测方法的流程图； 图 9为本发明实施例三中通过丟包率进行拥塞检测的示意图；
图 10为本发明实施例四中移动网络小区的拥塞检测方法的流程图； 图 11为本发明实施例四中通过 RTT和带宽联合进行拥塞检测的示意图； 图 12为本发明实施例五中核心网网元的结构示意图一；
图 13为本发明实施例五中获取单元的结构示意图；
图 14为本发明实施例五中核心网网元的结构示意图二。 具体实施方式
下面将结合本发明实施例中的附图， 对本发明实施例中的技术方案进行 清楚、 完整地描述， 显然， 所描述的实施例是本发明一部分实施例， 而不是 全部的实施例。 基于本发明中的实施例， 本领域普通技术人员在没有做出创 造性劳动前提下所获得的所有其他实施例， 都属于本发明保护的范围。
本发明实施例提供一种移动网络小区的拥塞检测方法和装置， 能够通过 核心网网元实现拥塞检测和数据分流。 实施例一
本发明实施例提供一种移动网络小区的拥塞检测方法， 如图 1 所示， 该 方法包括：
101、 核心网网元实时地获取小区数据报文的往返时间 ( Round Trip Time 简称 RTT );
在本实施例中， 核心网网元可以为 SGSN、 GGSN以及 SAE Gateway等。 具体地， 如图 2所示， 以数据报文为 TCP数据报文为例， 则获得数据报 文的 RTT的方法为：
1011、 核心网网元收到分组数据网 ( Packet Data Network , 简称 PDN )下 行 TCP数据报文后， 对该报文进行四层解析， 例如重传或者乱序判断， 并且 统计重传包以及乱序包次数， 以便于获得该 TCP数据报文的丟包率； 同时， 核心网网元记录收到该 TCP数据 4艮文时刻 Tr。
1012、 用户设备（ User Equipment , 简称 UE )收到该 TCP数据报文的数 据包后， 会发送一个确认数据包（ack )到核心网网元， 核心网网元记录收到 确认数据包（ ack ) 的时刻 Ta, 则该 TCP数据报文的 RTT值等于 Ta-Tr。
102、核心网网元将所述 RTT与预设的拥塞往返时间门限值相比较， 若所 述 RTT大于所述拥塞往返时间门限值，则判断所述小区过载拥塞，若所述 RTT 小于所述拥塞往返时间门限值， 则判断所述小区不拥塞。
在本实施例的技术方案中， 通过对数据报文的检测和判断实现对移动网 络小区的拥塞检测， 该方法可以在 SGSN、 GGSN以及 SAE Gateway等核心 网网元上实施， 核心网网元既能够完成拥塞检测和控制， 又能充分发挥核心 网进行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉及现有 移动网络架构和接口的更改， 因此不需要 BSC/RNC和 BTS/NodeB升级， 实 施方便、 快捷。
实施例二
当移动网络小区发生拥塞时， 超过速率限制的 TCP数据报文将被緩存入 队列， 由于緩存时间的增加， 导致 RTT时间变长， 而且伴随着流量的不断增 加， 緩存队列的深度也越来越深， RTT值也将不断变大。 如果业务流量仍然 持续增加， 则会造成緩存队列溢出， 这时后续到达的 TCP数据报文将会被丟 弃进而出现丟包现象。 因此， 可以通过检查小区 RTT值的变化来确定小区是 否发生拥塞。
本发明实施例提供一种移动网络小区的拥塞检测方法， 该方法通过对 RTT的检测和判断， 实现对小区的拥塞检测， 如图 2所示， 该方法包括：
201、 核心网网元实时地获取小区 TCP数据 4艮文的 RTT;
其中，获取小区 TCP数据报文的 RTT的方法参照实施例一步骤 101中的 具体描述和图 2所示， 在本实施例中不加以贅述。
由于本发明实施例是根据 TCP数据报文的 RTT变化来检测小区是否产生 拥塞， 而每个移动小区必然存在多个 TCP连接， 因此， 需要获取小区的 RTT。 并且， 需要尽量减少核心网网元设备提取和分析数据时所需的时间， 以减少 由于检测小区拥塞而对核心网网元设备处理能力的影响。 为了满足上述要求， 在本实施例中， 引入统计抽样的方法来进行 RTT的测量和统计。 即可以根据 小区 TCP数据报文的业务类型的数量和业务使用比例， 按照预设抽样间隔抽 取往返时间的抽样值和 /或丟包率的抽样值。
其中， 上述抽样间隔的确定为：
依据抽样时间间隔所服从的分布， 抽样方法可分为很多种， 在本实施例 中， 可以采用周期抽样、 随机附加抽样和泊松抽样。
周期抽样是每隔固定时间产生一次抽样。 随机附加抽样的抽样间隔的产 生是相互独立的， 并服从某种分布函数。 泊松抽样， 它的时间间隔符合泊松 分布。
利用上述方法来抽取样本， 根据上述的 RTT的抽样值获得小区的 RTT, 即， 取这些样本的 RTT平均值作为小区的 RTT。
Figure imgf000008_0001
同样地， 也可以统计这些样本的丟包率作为小区的丟包率。
其中， 抽样策略为：
由于移动终端涉及的 TCP业务众多， 为了更准确地度量小区的 RTT, 可 以根据所述数据报文中对应各业务类型的业务使用比例， 按照预设抽样间隔 抽取 RTT的抽样值， 例如采用分层抽样方法。 抽样时， 将小区下的 TCP数据 报文按照业务类型分为互不交叉的层， 然后按一定的比例， 再从每一层内随 机抽取一定数量的观察单位， 合起来组成样本。 采用这种方法， 样本的代表 性比较好， 而且抽样的误差也比较小。
如图 4所示， 将 TCP数据报文划分成 m类业务， 例如常见的网页浏览、 流媒体、 点到点业务（ Point to point简称 P2P ) 以及 Email邮件等业务， 每种 业务的比例是不相同的， 按照每种业务的比例从中抽取一定数量的 TCP数据 报文， 假设从 m个业务抽取了 n个 TCP数据报文样本， 则认为小区的 RTT 值就等于这！！个！^卩数据报文样本的 RTT均值； 同样地， 小区的丟包率等于 这 n个 TCP数据报文样本的丟包率。
采用分层抽样时需要按照对应各业务类型的业务使用比例抽取样本， 确 定业务使用比例可以采用两种途径：
1 )通过预先配置来实现；
每个运营商的移动业务分布都不相同， 可以根据实际应用情况来配置业 务比例， 该配置是一种平均统计分布。
2 )利用深度报文解析 (Deep Packet Inspection, 简称 DPI)功能解析报文来 实现。
核心网网元可以利用 DPI功能解析报文， 根据解析结果可以获得小区的 业务类型的数量以及每种业务的使用比例。
与预先配置相比，采用 DPI能够实时地体现小区下业务分布的真实情形。 DPI是一种基于应用层的流量检测和控制技术，当 IP数据包、 TCP或 UDP 数据流通过具有 DPI功能的系统时， 该系统通过深入读取报文载荷的内容来 对 OSI七层协议中的应用层信息进行识别和解析， 从而得到整个业务信息。
目前， 核心网网元， 例如 GGSN等， 其本身已经具备 DPI功能， 因此， 可以充分利用核心网网元的 DPI功能来解析报文， 从而实时地获得小区的业 务类型和业务分布。
相比于静态配置方法， 采用 DPI功能效果更佳， 它可以更加准确地获取 业务比例， 因此样本的代表性更好。
需要说明的是， 由于用户设备 UE会不时在小区与小区或路由区之间切 换， 因此， 当用户设备 UE移动时进入新的小区 /路由区时将会产生路由区更 新 /切换（ Serving RNS Relocation )等流程， 此时 SGSN应该进行小区是否发 生更改的处理。 以下实施例仅采用更新 (Routeing Area Update, 简称 RAU) 流 程进行举例说明， 切换（ Serving RNS Relocation )等流程具有类似的处理。
如图 5所示， 为 SGSN内路由区更新 ( Intra SGSN ) 流程， 即用户设备 UE在隶属于同一 SGSN下的路由区之间切换， 则 SGSN判断用户设备 UE是 否离开进行拥塞检测的小区，如果用户设备 UE离开进行拥塞检测的小区， 并 且该用户设备 UE为拥塞检测的样本用户，则应该将该用户设备 UE从样本中 删除， 并且将该用户设备 UE产生的流量不计算在小区使用带宽中。
如图 6所示， 为 SGSN间路由区更新 ( Inter SGSN ) 流程， 即用户设备 UE在隶属于不同 SGSN下的路由区之间切换， 则原 SGSN ( Old SGSN )判断 用户设备 UE是否离开进行拥塞检测的小区，如果用户设备 UE离开进行拥塞 检测的小区， 并且该用户为拥塞检测的样本用户， 则将该用户设备 UE从样本 中删除， 并且将该该用户设备 UE产生的流量不计算在小区使用带宽中。
202、核心网网元将所述 RTT与预设的拥塞往返时间门限值 RTTa相比较， 若所述往返时间大于所述拥塞往返时间门限值 RTTa, 则判断所述小区过载拥 塞， 若所述往返时间小于所述拥塞往返时间门限值 RTTa, 则判断所述小区不 拥塞。
其中， 拥塞往返时间门限值 RTTa为预设值， 核心网网元实时地检测小区 的 RTT, 并与 RTTa相比较， 若小区的 RTT大于 RTTa, 则判断所述小区过载 拥塞。
203、 在判断所述小区过载拥塞时， 核心网网元发出小区过载信号。
上述步骤 202和 203的具体过程， 如图 7所示，
(1)在 T1时间之前， 小区业务流量较小处于不拥塞状态中， 小区 RTT值 在一定的区间内平稳波动；
(2)从 T1时间开始，小区业务流量持续增加，报文緩存时间增加导致 RTT 值不断地增加；
(3)在 T2时间， 当小区 RTT值大于 RTTa时， 系统认为小区发生过载， 并且发出小区过载信号；
(4)从 T3时间开始， 小区的业务流量开始减少；
(5)在 T4时间， 当小区 RTT值小于 RTTa时， 系统认为小区恢复到不拥 塞状态。
在本实施例的技术方案中， 通过对 RTT的检测和判断， 实现对移动网络 小区的拥塞检测， 该方法可以在 SGSN、 GGSN以及 SAE Gateway等核心网 网元上实施， 核心网网元既能够完成拥塞检测和控制， 又能充分发挥核心网 进行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉及现有移 动网络架构和接口的更改， 因此不需要 BSC/RNC和 BTS/NodeB升级， 实施 方便、 快捷。
实施例三
本发明实施例提供一种移动网络小区的拥塞检测方法， 该方法通过对丟 包率的实时检测和判断， 实现对小区的拥塞检测， 如图 8所示， 该方法包括： 301、 核心网网元实时地获取小区 TCP数据报文的丟包率；
其中， 获取小区 TCP数据报文的丟包率的方法参照实施例一步骤 101中 的具体描述和图 2所示， 本实施例不加以贅述。
进一步地， 可以参照实施例二中 RTT样本的统计方法， 统计样本的丟包 率作为小区的丟包率。 包率大于所述丟包率门限值， 则判断所述小区过载拥塞， 若所述丟包率小于 所述丟包率门限值， 则判断所述小区不拥塞；
其中， 丟包率门限值为预设值， 核心网网元实时地检测小区的丟包率， 并与丟包率门限值相比较， 若小区的 RTT大于 RTTa, 则判断所述小区过载拥 塞。
303、 在判断所述小区过载拥塞时， 核心网网元发出小区过载信号。 上述步骤 302和 303的具体过程， 如图 9所示， 说明了根据丟包率变化 来进行小区拥塞判断的过程。
(1)在 T1时间之前， 小区业务流量较小处于不拥塞状态中， 小区没有发 生丟包或者丟包率很小；
(2) 随着小区业务流量的持续增加， 小区开始进入拥塞状态， 此时丟包率 会呈现上升的情况， 在 T1时间当丟包率大于丟包率门限值时， 系统认为小区 发生过载， 并且发出小区过载信号；
(3)在 T2时间， 当丟包率小于丟包率门限值时， 系统认为小区恢复到不 拥塞状态。
在本实施例的技术方案中， 通过对丟包率的实时检测和判断， 实现对移 动网络小区的拥塞检测， 该方法可以在 SGSN、 GGSN以及 SAE Gateway等 核心网网元上实施， 核心网网元既能够完成拥塞检测和控制， 又能充分发挥 核心网进行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉及 现有移动网络架构和接口的更改， 因此不需要 BSC/RNC和 BTS/NodeB的升 级， 实施方便和快捷。
实施例四 本发明实施例提供一种移动网络小区的拥塞检测方法，该方法通过对 RTT 和带宽的实时检测和判断， 实现对小区的拥塞检测， 如图 10所示， 该方法包 括：
401、 核心网网元实时地获取小区 TCP数据报文的 RTT和小区的带宽； 其中， TCP数据报文的 RTT的获取方式参照前述实施例， 在此不再贅述， 小区的总带宽可以通过静态配置， 也可以动态实时地获取。 当小区下用户的 业务总流量大于小区总带宽时， TCP连接的 RTT会呈现增大趋势。
402、 当所述往返时间的值逐渐变大， 而所述带宽保持平稳波动时， 核心 网网元判断所述小区过载拥塞；
403、 在判断所述小区过载拥塞时， 核心网网元发出小区过载信号。
其中， 图 11举例说明了 RTT和带宽联合进行小区拥塞判断的过程， 如图
11所示，
(1)在 T1时间之前， 尽管小区业务流量持续地增加， 但是实际使用带宽 小于小区总带宽， 此时小区 RTT值在一定的区间内平稳波动；
(2)从 T1时间开始， 小区业务流量持续增加， 当报文速率超过最大带宽 时， 超过速率限制的报文将首先被緩存入队列， 由于緩存时间的增加， RTT 值也将逐步变大， TCP 源将降低数据发送速度以响应拥塞， 减小对带宽的使 用。 因此， 在 T1〜T2时间内， 小区 RTT值逐渐地变大， 但是小区的使用带宽 保持平稳波动， 如果检测到这种变化， 系统认为小区发生过载， 并且发出小 区过载信号；
(3)从 Τ3时间开始， 伴随着小区使用带宽的逐步减少， RTT值也逐步下 降， 直至平稳波动。
在本实施例的技术方案中， 通过对小区 RTT和带宽联合检测， 实现对移 动网络小区的拥塞检测， 该方法可以在 SGSN、 GGSN以及 SAE Gateway等 核心网网元上实施， 核心网网元既能够完成拥塞检测和控制， 又能充分发挥 核心网进行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉及 现有移动网络架构和接口的更改， 因此不需要 BSC/RNC和 BTS/NodeB升级， 实施方便和快捷。
实施例五
本发明实施例还提供一种应用上述实施例所述的移动网络小区的拥塞检 测方法的核心网网元， 如图 12所示， 该核心网网元包括：
获取单元 11和比较判断单元 12, 其中，
获取单元 11 , 用于实时地获取小区数据报文的往返时间；
比较判断单元 12, 用于将所述往返时间与预设拥塞往返时间门限值相比 较， 若所述往返时间大于预设拥塞往返时间门限值， 则判断所述小区过载拥 塞， 若所述往返时间小于所述预设拥塞往返时间门限值， 则判断所述小区不 拥塞。
进一步地， 如图 13所示， 所述获取单元 11包括： 抽样模块 111和计算模 块 112.
抽样模块 111 , 用于根据所述数据报文中对应各业务类型的业务使用比 例， 按照预设抽样间隔抽取往返时间的抽样值； 计算模块 112, 用于根据所述 往返时间的抽样值获得所述往返时间。
所述预设抽样间隔的抽样方法为： 周期抽样、 随机附加抽样和泊松抽样。 进一步地， 如图 14所示， 所述获取单元 11还包括：
业务比例配置模块 113 ,用于通过预先配置或通过深度报文解析功能解析 报文获得所述数据报文中对应各业务类型的业务使用比例。
进一步地， 所述获取单元 11还用于实时地获取所述小区的带宽； 保持平稳波动时， 判断所述小区过载拥塞。
进一步地， 所述获取单元 11还用于实时地获取小区数据报文的丟包率； 较， 若所述丟包率大于所述丟包率门限值， 则判断所述小区过载拥塞， 若所 述丟包率小于所述丟包率门限值， 则判断所述小区未拥塞。
进一步地， 该核心网网元还包括：
过载提示单元 13 , 用于在判断所述小区过载拥塞时，发出小区过载信号。 需要说明的是， 上述核心网网元的功能可以由 SGSN、 GGSN以及 SAE Gateway等核心网网元实现。
在本实施例的技术方案中， 通过对数据报文的 RTT的检测和判断， 实现 对移动网络小区的拥塞检测， 该方法可以在 SGSN、 GGSN以及 SAE Gateway 等核心网网元上实施， 核心网网元既能够完成拥塞检测和控制， 又能充分发 挥核心网进行业务数据流区分的能力； 并且， 区别于现有技术， 该方法不涉 及现有移动网络架构和接口的更改， 因此不需要 BSC/RNC和 BTS/NodeB升 级， 实施方便、 快捷。
通过以上的实施方式的描述， 所属领域的技术人员可以清楚地了解到本 发明可借助软件加必需的通用硬件的方式来实现， 当然也可以通过硬件， 但 很多情况下前者是更佳的实施方式。 基于这样的理解， 本发明的技术方案本 质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来， 该 计算机软件产品存储在可读取的存储介质中， 如计算机的软盘， 硬盘或光盘 等， 包括若干指令用以使得一台计算机设备（可以是个人计算机， 服务器， 或者网络设备等）执行本发明各个实施例所述的方法。
以上所述， 仅为本发明的具体实施方式， 但本发明的保护范围并不局限 于此， 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 范围应以所述权利要求的保护范围为准。

Claims

权利 要求 书
1、 一种移动网络小区的拥塞检测方法， 其特征在于， 包括：
实时地获取小区数据报文的往返时间；
将所述往返时间与预设的拥塞往返时间门限值相比较， 若所述往返时间大 于所述拥塞往返时间门限值， 则判断所述小区过载拥塞， 若所述往返时间小于 所述拥塞往返时间门限值， 则判断所述小区未拥塞。
2、 根据权利要求 1所述的方法， 其特征在于， 所述实时地获取小区数据报 文的往返时间包括：
根据所述数据报文中对应各业务类型的业务使用比例， 按照预设抽样间隔 抽取往返时间的抽样值；
根据所述往返时间的抽样值获得所述往返时间。
3、 根据权利要求 2所述的方法， 其特征在于， 在所述根据所述数据报文中 对应各业务类型的业务使用比例， 按照预设抽样间隔抽取往返时间的抽样值之 前包括：
通过预先配置或通过深度报文解析功能解析报文获得所述数据报文中对应 各业务类型的业务使用比例。
4、 根据权利要求 2所述的方法， 其特征在于， 所述预设抽样间隔的抽样方 法为： 周期抽样、 随机附加抽样和泊松抽样。
5、 根据权利要求 1-4任一权利要求所述的方法， 其特征在于， 在所述实时 地获取小区数据报文的往返时间之后， 还包括：
实时地获取所述小区的带宽；
当所述往返时间的值逐渐变大， 而所述带宽保持平稳波动时， 判断所述小 区过载拥塞。
6、 根据权利要求 1所述的方法， 其特征在于， 还包括：
实时地获取小区数据报文的丟包率；
将所述丟包率与预设的丟包率门限值相比较， 若所述丟包率大于所述丟包 率门限值， 则判断所述小区过载拥塞， 若所述丟包率小于所述丟包率门限值， 则判断所述小区未拥塞。
7、 根据权利要求 1、 2、 3、 4或 6任一权利要求所述的方法， 其特征在于， 还包括：
在判断所述小区过载拥塞时， 发出小区过载信号。
8、 一种核心网网元， 其特征在于， 包括：
获取单元， 用于实时地获取小区数据报文的往返时间；
比较判断单元， 用于将所述往返时间与预设的拥塞往返时间门限值相比较， 若所述往返时间大于所述拥塞往返时间门限值， 则判断所述小区过载拥塞， 若 所述往返时间小于所述拥塞往返时间门限值， 则判断所述小区未拥塞。
9、 根据权利要求 7所述的核心网网元， 其特征在于， 所述获取单元包括： 抽样模块， 用于根据所述数据报文中对应各业务类型的业务使用比例， 按 照预设抽样间隔抽取往返时间的抽样值；
计算模块， 用于根据所述往返时间的抽样值获得所述往返时间。
10、 根据权利要求 9所述的核心网网元， 其特征在于， 所述预设抽样间隔 的抽样方法为： 周期抽样、 随机附加抽样和泊松抽样。
11、 根据权利要求 9 所述的核心网网元， 其特征在于， 所述获取单元还包 括：
业务比例配置模块， 用于通过预先配置或通过深度报文解析功能解析报文 获得所述数据报文中对应各业务类型的业务使用比例。
12、 根据权利要求 8-11任一权利要求所述的核心网网元， 其特征在于， 所述获取单元还用于实时地获取所述小区的带宽； 平稳波动时， 判断所述小区过载拥塞。
13、 根据权利要求 8所述的核心网网元， 其特征在于，
所述获取单元还用于实时地获取小区数据报文的丟包率； 所述比较判断单元还用于将所述丟包率与预设的丟包率门限值相比较， 若 所述丟包率大于所述丟包率门限值， 则判断所述小区过载拥塞， 若所述丟包率 小于所述丟包率门限值， 则判断所述小区未拥塞。
14、 根据权利要求 8、 9、 10、 11或 13任一权利要求所述的核心网网元， 其特征在于， 还包括：
过载提示单元， 用于在判断所述小区过载拥塞时， 发出小区过载信号。
PCT/CN2011/074451 2011-05-20 2011-05-20 移动网络小区的拥塞检测方法和装置 WO2011144061A2 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2011/074451 WO2011144061A2 (zh) 2011-05-20 2011-05-20 移动网络小区的拥塞检测方法和装置
CN201180000704.3A CN102893654B (zh) 2011-05-20 2011-05-20 移动网络小区的拥塞检测方法和装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/074451 WO2011144061A2 (zh) 2011-05-20 2011-05-20 移动网络小区的拥塞检测方法和装置

Publications (3)

Publication Number Publication Date
WO2011144061A2 true WO2011144061A2 (zh) 2011-11-24
WO2011144061A3 WO2011144061A3 (zh) 2012-04-19
WO2011144061A9 WO2011144061A9 (zh) 2012-07-12

Family

ID=44992103

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/074451 WO2011144061A2 (zh) 2011-05-20 2011-05-20 移动网络小区的拥塞检测方法和装置

Country Status (2)

Country Link
CN (1) CN102893654B (zh)
WO (1) WO2011144061A2 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103782554A (zh) * 2012-07-25 2014-05-07 华为技术有限公司 数据分流方法，数据发送装置以及分流节点装置
US9584419B1 (en) 2015-03-18 2017-02-28 Google Inc. Systems and methods for detecting consistent fabric congestion in a datacenter based on historical traffic data

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106912072A (zh) * 2015-12-22 2017-06-30 上海自足网络科技有限公司 一种基于虚拟传输协议的无线Ad Hoc网络拥塞控制机制

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090022055A1 (en) * 2007-07-16 2009-01-22 Microsoft Corporation Systems and methods for improving tcp-friendliness of delay-based congestion control
CN101645765A (zh) * 2009-08-03 2010-02-10 四川大学 面向高误码率、长时延特性网络的可靠传输加速方法
CN101917342A (zh) * 2010-08-25 2010-12-15 天津理工大学 一种面向普适服务的自适应网络拥塞控制方法
CN101984601A (zh) * 2010-11-04 2011-03-09 中国人民解放军空军工程大学 混合型多目标优化组播拥塞控制方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101222296B (zh) * 2008-01-31 2010-06-09 上海交通大学 上行蜂窝视频通信中自适应的传输方法及系统
CN101420419B (zh) * 2008-10-27 2011-05-18 吉林大学 一种自适应高速网络流量分层抽样采集方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090022055A1 (en) * 2007-07-16 2009-01-22 Microsoft Corporation Systems and methods for improving tcp-friendliness of delay-based congestion control
CN101645765A (zh) * 2009-08-03 2010-02-10 四川大学 面向高误码率、长时延特性网络的可靠传输加速方法
CN101917342A (zh) * 2010-08-25 2010-12-15 天津理工大学 一种面向普适服务的自适应网络拥塞控制方法
CN101984601A (zh) * 2010-11-04 2011-03-09 中国人民解放军空军工程大学 混合型多目标优化组播拥塞控制方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103782554A (zh) * 2012-07-25 2014-05-07 华为技术有限公司 数据分流方法，数据发送装置以及分流节点装置
CN103782554B (zh) * 2012-07-25 2016-10-26 华为技术有限公司 数据分流方法，数据发送装置以及分流节点装置
US10630801B2 (en) 2012-07-25 2020-04-21 Huawei Technologies Co., Ltd. Data shunting method, data transmission device, and shunting node device
US9584419B1 (en) 2015-03-18 2017-02-28 Google Inc. Systems and methods for detecting consistent fabric congestion in a datacenter based on historical traffic data

Also Published As

Publication number Publication date
WO2011144061A3 (zh) 2012-04-19
CN102893654B (zh) 2016-06-08
CN102893654A (zh) 2013-01-23
WO2011144061A9 (zh) 2012-07-12

Similar Documents

Publication Publication Date Title
US9326182B2 (en) Method and system for radio service optimization using active probing over transport networks
JP6553196B2 (ja) トラフィックフローの監視
Wang et al. An active-passive measurement study of TCP performance over LTE on high-speed rails
US7286474B2 (en) Method and apparatus for performing admission control in a communication network
JP4819063B2 (ja) サービス区別型無線ネットワークにおけるＱｏＳの測定及び監視
US8085710B2 (en) Minimizing packet loss during fast roaming
US11778491B2 (en) 5G user plane packet delay performance measurements
JP2008533900A (ja) サービス差別型無線ネットワークでのＱｏＳの測定及び監視
KR20150085405A (ko) 통신 시스템에서 무선 네트워크의 혼잡 검출 장치 및 방법
Song et al. Leveraging frame aggregation for estimating WiFi available bandwidth
US8976689B2 (en) Methods, systems, and computer program products for monitoring network performance
WO2011074116A1 (ja) 優先制御装置、車載装置および優先制御方法
CN111066272B (zh) 移动无线电接入网络中的分组延迟减少
Dolivo Signal processing for high-density digital magnetic recording
WO2011144061A2 (zh) 移动网络小区的拥塞检测方法和装置
US11785442B2 (en) Data transport network protocol based on real time transport network congestion conditions
Wu et al. TCP fairness issues in IEEE 802.11 wireless LANs
WO2020063661A1 (zh) 一种流量拥塞监测方法及装置
Szilágyi et al. LTE user plane congestion detection and analysis
Michelinakis et al. Passive mobile bandwidth classification using short lived TCP connections
Qiu et al. Hierarchy virtual queue based flow control in LTE/SAE
CN113950099A (zh) 一种网络拥塞控制方法及设备
JP2003037630A (ja) ディジタルコンテンツの伝送におけるサービス品質を測定するための方法及び装置、並びにサービス品質を制御するための方法及び装置
Bergfeldt et al. Real‐time bandwidth measurements over mobile connections
JP5528372B2 (ja) フロー品質劣化特定装置及び方法

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201180000704.3

Country of ref document: CN

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11782978

Country of ref document: EP

Kind code of ref document: A2