WO2010091610A1

WO2010091610A1 - Link detection method, apparatus and communications system thereof

Info

Publication number: WO2010091610A1
Application number: PCT/CN2010/070399
Authority: WO
Inventors: 何贵洲
Original assignee: 华为技术有限公司
Priority date: 2009-02-13
Filing date: 2010-01-28
Publication date: 2010-08-19
Also published as: CN101488835A

Abstract

A link detection method, apparatus and communications system thereof are provided in the present invention. The method comprises: at least one user data stream is identified by a multi-element set; When the preset statistical period is started, the sequence number of the downlink direction start message in the user data stream is received and obtained, a counter is started, the number of the actually received message is recorded as 1, and then the counter is increased by 1 when receiving a message every time; when the statistical period ends, the sequence number of the end message is received and obtained, the number of the messages which should be received is calculated according to the sequence number of the start message and the sequence number of the end message; the packet loss number and/or packet loss ratio are calculated according to the number of messages which should be received and the number of the actually received messages. The system includes: a first unit, for identifying at least one user data stream by a multi-element set; a second unit, for receiving and obtaining the sequence numbers of the downlink direction messages in the user data stream; a third unit, for counting the actually received messages; a forth unit, for calculating the number of messages which should be received in the statistical period; a fifth unit, for calculating the packet loss number and/or the package loss ratio.

Description

Link detection method, device and communication system

The present application claims priority to Chinese Patent Application No. 2009-10105436.6, entitled "A Link Detection Method, Apparatus, and Communication System", filed on February 13, 2009. Technical field

The embodiments of the present invention relate to the field of communications, and in particular, to a link detection method, apparatus, and communication system. Background technique

The GTP (General Packet Radio Service Tunnel Protocol) protocol is applied between the GSN (GPRS Support Node, GPRS service node, including the GGSN and the SGSN) to establish a GTP channel for each mobile station (MS). The GTP channel is a secure channel between the GSNs. Two hosts can exchange data through this channel. The GTP is carried over the UDP (User Datagram Protocol) or the Transmission Control Protocol (TCP). It is divided into a signaling plane and a transmission plane. The signaling plane defines various messages and transmission planes. The GTP protocol is divided into GTP-C, GTP-U and GTP, and the GTP-C is a signaling control protocol, and GTP-U is a packaged user data protocol, GTP, It is a billing related agreement. There is a TEID (Tunnel End Point Identifier) in the GTP protocol, which identifies the tunnel endpoint on the data receiver GTP-U or GTP-C protocol entity unambiguously. The two nodes of a GTP tunnel each allocate a TEID that the party is going to use to receive specific data on the tunnel, and negotiates between the nodes through the GTP-C message. In addition, GRE (Generic Routing Encapsulation) is also a commonly used tunneling protocol. GRE defines a protocol for encapsulating any other network layer protocol on any network layer protocol. The GRE protocol can encapsulate multiple IP tunnels. A protocol package that creates a virtual point-to-point link over an IP network. In the prior art, the base station and the controller side usually adopt IPPM (IP Performance). Monitoring, performance monitoring) completes the detection of intermediate network transmission links. The existing detection methods have high requirements for compatibility between the detection devices. If the detection device and the peer device are provided by different manufacturers, the detection may be difficult to complete; in addition, the existing detection method requires the control surface to do More complex calculations and configurations consume huge amounts of system resources. Summary of the invention

A technical problem to be solved by embodiments of the present invention is to provide a link detection method, apparatus, and communication system.

To solve the above technical problem, an embodiment of the present invention provides a link detection method, including the following steps:

Identifying, by the multi-group, at least one user data stream; receiving, at the beginning of the preset statistical period, receiving and acquiring a sequence number of the downlink start message in the user data stream, starting a counter; in the statistical period, each received a message, the number of the received message in the counter is incremented by one; when the statistical period is over, the sequence number of the end message in the downlink direction of the user data stream is received and obtained; and the sequence number of the message is started according to the downlink direction. And ending the packet sequence number to calculate the number of receivable packets; calculating the number of lost packets and/or the packet loss rate according to the number of receivables and the number of received packets.

In addition, the embodiment of the present invention further provides a device for detecting a link, including:

a first unit, configured to identify at least one user data stream by using a multi-group; and a second unit, configured to receive and acquire a downlink message sequence number in the user data stream in a preset statistical period; Counting the number of packets actually received in the statistical period; the fourth unit is configured to calculate the number of receivables in the statistical period according to the sequence number of the packet received and obtained by the second unit; And configured to calculate a number of lost packets and/or a packet loss rate in the statistical period according to calculation results of the third unit and the fourth unit.

In addition, an embodiment of the present invention further provides a communication system, including the foregoing link detecting apparatus. Compared with the prior art, the method, the device, and the communication system provided by the embodiments of the present invention can detect the link quality without the participation of the peer device, even if the device docking of different vendors does not affect the detection. It is implemented and simple to implement, and consumes less system resources. DRAWINGS

1 is a schematic diagram of a method according to an embodiment of the present invention;

2 is a schematic diagram of a system according to an embodiment of the present invention. detailed description

The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments. The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

This embodiment provides a link detection method, which is a link detection for a flow ID (flow ID), that is, a detection and statistics of a user data stream by using a data stream ID as a statistical object. The method of this embodiment includes: first identifying a user data stream, where the meaning of the user data stream is not used to be defined as a data stream between the user equipment and the communication device, but is used to distinguish from the control flow of the control plane. Concept, the user data stream includes end-to-end various service flows generated in the communication, for example, an end-to-end data stream generated by a mobile phone call, or an end-to-end data stream generated by a user using a BT (BitTorrent) service when surfing the Internet. It can also be an end-to-end data stream generated by the user using the FTP service. At the same time, there are many ways to identify the user data stream, for example, it can be identified by a triple or a quintuple. In addition, a single user may generate one or more user data streams during the communication process, that is, one or more flow IDs may be corresponding, and only one of the user data streams may be counted in one statistical period, or Statistics are performed on multiple user data streams.

Preset a statistical period T. The statistical period can be adjusted according to the requirements of statistical accuracy. For example, it can be set to 5 seconds. When the statistical period starts, the serial number of the start message is received and obtained, and the first one will be received. The message is recorded as a received message, and the counter is started at the same time. The number of received messages is counted as 1. Of course, the initial count of the counter here may not be 1, when the initial count of the counter is not 1, The initial count value of the counter can be read once at the beginning of the statistical period. When the statistical period ends, the initial count value read at the beginning of the statistical period is subtracted from the counter count; in the next statistical period, the single stream The number of received messages for each message sequence number counter is incremented by one; at the end of the statistical period, the user data stream based on the flow ID is received and the sequence number of the end message in the downlink direction is obtained; according to the start message The sequence number of the sequence number and the end number of the end message can be used to calculate the number of packets that should be received, that is, the number of receivable packets, and calculate the number of lost packets based on the calculated number of receivables and the number of received packets counted in the counter. And / or packet loss rate. In the detection process, the data stream received by the detecting end can be defined as a downlink data stream. The statistics performed on the user data stream in this embodiment are statistics for the downlink data packet in the user data stream, that is, the detecting end. Statistics on the received packets in the received data stream. The statistics can be completed on the detection end.

The solution provided in this embodiment can be implemented by using the intrinsic characteristics of the device, and the end-to-end implementation is not required. The detection process can be completed at the detecting end, and the peer device can be used without participation. In practice, there are often devices that are connected by different vendors. With the solution provided by the embodiment, even if the devices of different vendors are connected, the implementation of the detection is not affected, and the compatibility is good. Moreover, the method of the embodiment is simpler because the participation of the peer device is not required. It does not require the control plane to do more complicated calculations and configurations, consumes less system resources, and can perform real-time detection and closed-loop regulation of link quality.

This embodiment provides another method for link detection. First, the structure of the GTP message frame involved in this embodiment is introduced. As shown in Table 1, the message header of the GTP VI version consists of a fixed part and an optional part. The fixed part consists of 8 bytes, describing the version number, flag bit information, GTP message type, message length, and TEID field used by the GTP message. The optional part includes the serial number, the N-PDU number, and the extension part. The presence or absence of these optional parts is determined by the value of the identification bit.

Offset 8 7 6 5 4 3 2 1

1 Version PT ( * ) E S PN

2 Message Type

3 Length ( 1 Octet)

4 Length (2 Octet)

5 TEID ( 1 Octet) 6 TEID (2 Octet)

7 TEID (3 Octet)

8 TEID (4 Octet)

9 Sequence Number ( 1 Octet)

10 Sequence Number ( 2 Octet )

11 N-PDU Number ² ) ⁴ )

12 Next Extension Header Type

Table 1

The Version field defines the version number used by GTP, and is set to "001" in the VI version. The PT field is a protocol specifier for distinguishing between GTP and GTP, and the message; the E header (Extension Header flag) is used to indicate whether it exists. The extension number header is used to indicate whether there is an identifier bit of the SN domain; the PN field is used to indicate whether the GTP message header has an N-PDU domain; the Message Type field is used to indicate the GTP message type; the Length field indicates The length of the message except the fixed length message part; the TEID field indicates that the receiver allocates the TEID of the receiving end, and negotiates with the sender through the GTP-C, RANAP, etc. TEID negotiation; in the GTP VI version, the Sequence Number The domain is an optional domain part; the N-PDU Number field is also an optional domain part, which is used for the routing area update process and the SRNS relocation process between the SGSNs. The Next Extension Header Type is an extension of the GTP header that maintains compatibility with future versions.

Next, the GRE frame structure is introduced. As shown in Table 2, the GRE frame structure is as follows:

0 1 2 3

01234567890123456789012345678901

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

|C|R|K|S|s|Recur| Flags | Ver | Protocol Type |

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

I Checksum ( optional ) | Offset (optional) |

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

I Key ( optional ) |

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

I Sequence Number ( optional ) |

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

I Routing ( optional )

+-+ -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+-+-+-+-+-+.+

Table 2

The GRE frame structure is compared with the GTP frame structure, wherein the Key field is equivalent to the TEID in the GTP frame, and the sequence number Sequence Number parameter is the same as the Sequence Number in the GTP frame structure, and is optional. As shown in FIG. 1, the method in this embodiment includes:

Step 101: Identify at least one user data stream by using a multi-group;

Here, there are various methods for identifying a user data stream, for example, a triplet or a quintuple may be used. If a triplet method is used, at least one user data may be identified by a source IP, a destination IP, and a protocol type. If the method of the quintuple is used, the source IP, the destination IP, the source port, the destination port, and the protocol type may be used to identify at least one user data stream. Of course, those skilled in the art may also have other extension manners to identify At least one user data stream. In addition, in the process of detecting the identified user data stream, the packet may be counted and detected by the stream ID. The flow ID (flow ID) may be different according to the packet type. For example, for a GTP packet, the stream ID can be a TEID. In the actual application, the Key field in the GRE packet is usually processed as mandatory. Therefore, for the GRE "3⁄4 text, the stream ID can be Key.

Step 102: At the beginning of the preset statistical period, receive and acquire the sequence number of the start message in the downlink direction of the user data stream, start a counter, and record the number of the received message as 1;

There are various methods for obtaining the sequence number of the start message in the downlink direction of the user data stream. For example, according to the description of the structure of the GTP frame, some fields in the GTP message frame structure are optional, and therefore, the use may be considered. An optional field in the frame structure of the GTP packet carries the sequence number of the message. For example, the sequence number field in the GTP message frame structure can be used to carry the sequence number of the start message. In addition, the sequence number of the start message can also be passed. The GRE packet in the downlink direction of the user data stream is carried. Considering that a part of the field in the frame structure of the GRE packet is optional, the sequence number of the start message can be carried, for example, by using the Sequence Number field in the GRE packet, for example. The sequence number of the downstream data stream can be expressed in hexadecimal, increasing from 0 to OXffffffff by 4 bytes and allowing flipping. In addition, in order to count the number of actually received packets, a counter can be started to record the number of packets actually received. Specifically, a variable BeginSeqNo can be set, the serial number of the received message is taken out and assigned to the variable BeginSeqNo, and then a variable RealRcvPktNum is set to represent the number of received messages, and the number of received messages is set to 1, that is, RealRcvPktNum =L Step 103: In the statistical period, the number of received messages in the counter is incremented by one for each received message;

In this embodiment, the sequence number of each message sent on a single stream is increased by 1. Specifically, when a GTP or GRE message is received, the received message variable is incremented by 1, so that RealRcvPktNum= RealRcvPktNum + 1.

Step 104: At the end of the statistical period, receive a TEID-based user data stream and obtain a sequence number in which the downlink direction ends >3⁄4;

Similar to the start message sequence number, the sequence number of the end message can also be carried by an optional field in the GTP or GRE message frame structure, such as the Sequence Number field. Specifically, a variable EndSeqNo can be set, at the end of the statistical period. , the serial number of the received message is taken out and assigned to the variable EndSeqNo ₀

Step 105: Calculate the number of receivables according to the start sequence sequence number and the end message sequence number in the downlink direction. Specifically, a variable ShouldRcvPktNum may be set to indicate the number of receivables, and it is noted that the report is considered. The sequence number of the text may be reversed. When calculating the number of receivable packets, the sequence number of the start message and the sequence number of the end message may be reversed. Specifically: if the sequence number of the end message is greater than the sequence number of the start message, The difference between the species is the number of packets received, that is, ShouldRcvPktNum = EndSeqNo - BeginSeqNo + 1; otherwise, the serial number has been inverted, then the number of receivables is ShouldRcvPktNum = OxFFFFFFFF + EndSeqNo - BeginSeqNo + 1;

Step 106: Calculate the number of lost packets and/or the packet loss rate according to the number of receivable packets and the number of received packets. Specifically, if LostPktNum is set to indicate the number of lost packets, then LostPktNum=ShouldRcvPktNum-RealRcvPktNum, and LostPktRatio is used to indicate the packet loss rate. , then LostPktRatio= LostPktNum/ ShouldRcvPktNum, the calculation result of the number of receivables is

In hexadecimal notation, for the convenience of calculation and statistics, the hexadecimal number can be converted to a decimal number when calculating the packet loss rate or the number of lost packets, and the number of received packets calculated in decimal Perform mathematical calculations. In addition, in order to facilitate the statistics of the next period, at the end of the statistical period, the sequence number of the end message is incremented by 1 and then preset to the start message sequence number of the next statistical period, that is, the start sequence of the next period. No. NextBeginSeqNo = EndSeqNo + 1.

It should be pointed out that the sequence number of the last end message of the last statistical period is very important. If the message is out of order or incorrect, it will have a greater impact on the statistics of the next cycle. Therefore, when an abnormal situation occurs, you can consider The start sequence serial number actually received in the next cycle is corrected. Specifically, for the next cycle, if the first sequence number of the first message received in the cycle is not much different from the sequence number of the last message in the previous cycle, for example, the difference is about 2-3, It is considered that the out-of-order message in this period is still used as the statistical message of this period; if the sequence number of the first message actually received in this cycle is very different from the previous period, such as 10000 or more, it is beyond the period. The statistical category can be regarded as the message that is transmitted incorrectly in this period, and is also recorded as the current period statistics. If the first sequence number actually received in this period is smaller than the sequence number of the last period ending message, it is considered to be the upper period message. It can be discarded. In addition, the continuity of several message serial numbers received continuously at the beginning of this cycle should be considered. If the continuity is good and the sequence number of the previous cycle is quite different, it can be considered that there is a previous cycle. Problem, do the assignment correction processing. In addition, communication networks usually use private networks. The private network is well planned. Usually, congestion and out-of-order situations do not occur. In this scenario, packets can be considered as ordered, but if The communication link goes to the public network. The public network usually has no quality guarantee. Generally, different routes will be sent to different routes, causing packets to be out of order. To avoid the impact of packet disorder on detection, statistics can be obtained. The messages in the user data stream are sorted before the start of the cycle to further improve the accuracy and efficiency of the detection.

After the number of lost packets and the packet loss rate are known, real-time monitoring and statistics of link quality can also be performed according to these parameters. Specifically, by setting the statistics period, the detecting end performs the statistics on the packet loss rate data to the second level, and the accumulated time, for example, 1 minute, is accumulated according to the statistical result, and then the minute is calculated. The packet loss rate is reported to the upper-layer operation and maintenance center. For example, the base station OM, the base station OM periodically reports the packet loss rate data of the minute level to the network management system. The network management system can form the packet loss rate curve according to the baseline. Hour or 1 day, the network management can draw time, The packet loss rate curve of the day, using the packet loss rate curve, the user can monitor the quality of the transmission network to adjust the network operation parameters in real time.

In addition, after calculating the packet loss rate, the detection link may be closed-loop adjusted according to the result. Specifically, a packet loss rate threshold may be preset, and the calculated packet loss rate and the packet loss rate threshold may be performed. If the calculated packet loss rate exceeds the preset packet loss rate threshold, the peer can send a message limiting the flow rate. After receiving the message, the peer performs the current limiting process. After that, the detection link can be monitored and compared in real time. If the packet loss rate is lower than the preset packet loss threshold for a certain period of time, the peer can be notified to release the restriction on the flow rate.

The embodiment provides a link detecting apparatus. As shown in FIG. 2, the system includes: a first unit 11 configured to identify at least one user data stream by using a multi-group; wherein the method for identifying a user data stream may have multiple For example, using a triplet (source IP, destination IP, protocol type) or a quintuple (source IP, destination IP, source port, destination port, protocol type) to identify the user data stream, a single user can generate during the communication process. One or more user data streams, that is, one or more TEIDs. In one statistical period, statistics can be performed only for one user data stream, or multiple user data streams can be performed in units of TEIDs. Statistics

The second unit 12 is configured to receive and obtain a downlink message sequence number in the user data stream in a preset statistical period, where the sequence number of the downlink packet may be carried in an optional field in the GTP/GRE frame. , for example, in the Sequence Number field;

The third unit 13 is configured to count the actually received packets in the statistics period, and the counter can be used for statistics, and the number of received ones is incremented by one on a single stream; The method is used to calculate the number of receivables in the statistical period according to the sequence number of the packet received and obtained by the second unit 12, specifically: if the ending packet sequence number is greater than the start message sequence number, the difference between the two That is, the number of receivables is obtained, that is, the number of receivables ShouldRcvPktNum is equal to the end 4艮 serial number EndSeqNo minus the start 4艮 serial number BeginSeqNo plus 1; otherwise, the serial number has been inverted, then the receivable message The number is ShouldRcvPktNum equal to OxFFFFFFFF plus EndSeqNo minus BeginSeqNo force 1; The fifth unit 15 is configured to calculate the number of lost packets and/or the packet loss rate in the statistics period according to the calculation result of the third unit 13 and the fourth unit 14, specifically, setting LostPktNum to indicate the number of lost packets, then LostPktNum =| ShouldRcvPktNum-RealRcvPktNum|, where "| |" means the absolute value, LostPktRatio is used to indicate the packet loss rate, and 'j LostPktRatio= LostPktNum /

ShouldRcvPktNum„

In addition, the device in this embodiment may further include a sixth unit for sorting the messages, and then starting the detection after sorting, in consideration of the fact that the packets may already be out of order before the detection starts.

In addition, the apparatus of this embodiment may further include a seventh unit, configured to compare the calculated actual packet loss rate with a preset packet loss rate threshold, if the packet loss rate exceeds the preset packet loss rate threshold A value can be used to send a traffic limiting message to the peer.

The present embodiment provides a communication system including the link detection device mentioned in the embodiment shown in FIG. 2, wherein the link detection device may include the first unit mentioned in the embodiment shown in FIG. 11 to fifth unit 15, preferably, may also include a sixth unit and/or a seventh unit.

It should be noted that the methods and systems provided by the foregoing embodiments can be used not only in an LTE (Long Term Evolution) system but also in WiMaX (World Interoperability for Microwave Access, WCDMA, etc.). In other communication systems.

In addition, all or part of the processes in the foregoing embodiment may be completed by a computer program to instruct related hardware, and the program may be stored in a computer readable storage medium, and the program may include, when executed, The flow of an embodiment of each of the above methods. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

It should be noted that the above embodiments are only for explaining the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: Modifications to the technical solutions described in the foregoing embodiments, or equivalent replacement of some of the technical features; and these modifications or replacements do not make corresponding techniques The essence of the technical solution lies in the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

Claim

A link detection method, comprising:

Identifying at least one user data stream with a multi-group;

Receiving and acquiring the sequence number of the downlink start message in the user data stream at the beginning of the preset statistical period, and starting the counter;

During the statistics period, the number of received packets in the counter is incremented by one for each received message; when the statistical period ends, the sequence number of the downlink end message in the user data stream is received and obtained;

And calculating, according to the downlink direction start message sequence number and the end message sequence number, the number of receivable packets; and calculating the number of lost packets and/or the packet loss rate according to the number of the receivable packets and the number of received packets.

The method according to claim 1, wherein the received downlink start message sequence number and the received downlink direction end message sequence number are carried in a GTP message or a GRE message. .

3. The method according to claim 1, wherein the multi-component is a 5-tuple composed of a source IP, a destination IP, a source port, a destination port, and a ten-tune type.

The method according to claim 1, wherein the method further comprises: when the statistical period ends, adding 1 to the sequence number of the end message as a start report of the next statistical period. The serial number of the text, and when the sequence number of the start message actually received in the next statistical period is inconsistent with the preset start sequence number, the start sequence serial number actually received in the next statistical period is corrected.

5. The method of claim 1, wherein the method further comprises: sorting the four texts before the beginning of the statistical period.

The method according to claim 1, wherein the method further comprises: performing statistics on the packet loss rate within a preset time period, and transmitting the statistics to the operation and maintenance center OM.

The method according to claim 1, wherein the method further comprises: comparing the packet loss rate with a preset packet loss rate threshold, if the packet loss rate exceeds the preset lost rate Packet rate threshold, Then, the traffic limiting message is sent to the peer end.

8. A link detecting device, comprising:

a first unit, configured to identify at least one user data stream by using a multi-group;

a second unit, configured to receive and obtain a sequence number of the downlink direction message in the user data stream in a preset statistical period;

The third unit is configured to count the actually received packets in the statistical period;

a fourth unit, configured to calculate, according to a message sequence number received and obtained by the second unit, a number of receivables in the statistical period;

The fifth unit is configured to calculate, according to the calculation result of the third unit and the fourth unit, the number of lost packets and/or the packet loss rate in the statistical period.

The link detecting apparatus according to claim 8, wherein the apparatus further comprises: a sixth unit, configured to sort the message.

The link detecting apparatus according to claim 8, wherein the apparatus further comprises: a seventh unit, configured to compare the packet loss rate with a preset packet loss rate threshold, if If the packet loss rate exceeds the preset packet loss rate threshold, the traffic limiting message is sent to the peer end.

A communication system, comprising the link detecting apparatus according to any one of claims 8 to 10.