Method, device, medium and product for estimating network packet loss rate

Abstract

This application discloses a method, apparatus, medium, and product for estimating network packet loss rate, relating to the field of transmission and bearer technology. The method includes: acquiring signaling statistics data collected at a preset collection point in a first link, including the total number of request packets in a first direction and a second direction of the first link, and the number of retransmissions of request packets in the corresponding directions; wherein the first link includes a first end and a second end; calculating a first overall packet loss rate in the first direction and a second overall packet loss rate in the second direction of the first link based on the signaling statistics data; determining the segmented packet loss rate of segmented links of the first link based on the first overall packet loss rate and the second overall packet loss rate, wherein the segmented links include a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point. The solution of this application solves the problem that existing packet loss rate assessment schemes cannot accurately and quantitatively estimate packet loss rates.

Description

Technical Field

[0001] This application relates to the field of transmission and bearer technology, specifically to a method, apparatus, medium, and product for estimating network packet loss rate. Background Technology

[0002] In the 5G era, the real-time guarantee capability of network quality directly determines the user experience of various services. Among these, network packet loss is one of the core causes of service quality degradation. Therefore, accurate assessment of network packet loss rate is of great significance for network quality assurance. Currently, the mainstream network packet loss rate assessment methods in the industry are mainly divided into the following three types: Solution 1: Calculate the packet loss rate by statistically analyzing the packet sending and receiving data between devices A and B; Solution 2: Estimate the round-trip packet loss rate by sending request packets through the local device and based on the received response packets; Solution 3: Deploy signaling probes in the network to indirectly estimate the packet loss rate through statistical analysis of the entire packet transmission process. Currently, major domestic operators have adopted this signaling probe solution in actual network deployments to improve network quality control and ensure a high-quality service experience.

[0003] However, all three packet loss rate statistics schemes mentioned above have obvious limitations and are difficult to meet the needs of refined quality assurance for 5G networks: Scheme 1 relies on collaborative testing between devices at both ends A and B, and can only count end-to-end packet loss, but cannot locate the specific reasons for poor network quality and fault nodes; Scheme 2 simplifies the testing process and does not require multi-terminal collaboration, but is limited by the testing logic and cannot achieve comprehensive packet loss statistics in a multi-terminal concurrent environment, and the statistical results are one-sided. Summary of the Invention

[0004] At least one embodiment of this application provides a method, apparatus, medium, and product for estimating network packet loss rate, which addresses the problem that existing packet loss rate assessment schemes cannot accurately quantify the rate.

[0005] To solve the above-mentioned technical problems, this application is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a method for estimating network packet loss rate, including:

[0007] Obtain signaling statistics data of the first link collected at a preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction.

[0008] Based on the signaling statistics, calculate the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction.

[0009] Based on the first overall packet loss rate and the second overall packet loss rate, the segmented packet loss rate of the segmented link of the first link is determined, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

[0010] Optionally, after determining the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, the method further includes:

[0011] Based on the segmented packet loss rate, a mapping function relationship between link quality and the corresponding link packet loss rate is established; the larger the value of the link quality, the better the link quality.

[0012] The network fault result is determined based on the mapping function relationship;

[0013] Alternatively, a link judgment threshold can be set. If the segment packet loss rate is greater than or equal to the link judgment threshold, the network fault result is determined to be abnormal link quality; if the segment packet loss rate is less than the link judgment threshold, the network fault result is determined to be normal link quality.

[0014] Optionally, based on the signaling statistics, the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction are calculated, including:

[0015] Based on the signaling statistics, the number of first request messages and the number of first retransmission messages in the first direction of the first link are obtained;

[0016] Based on the number of the first request messages and the number of the first retransmission messages, calculate the first overall packet loss rate of the first link in the first direction;

[0017] Based on the signaling statistics, obtain the number of second request messages and the number of second retransmission messages in the second direction of the first link;

[0018] Based on the number of the second request messages and the number of the second retransmission messages, calculate the second overall packet loss rate of the first link in the second direction.

[0019] The second overall packet loss rate of the path from the second communication terminal to the first communication terminal is calculated based on the number of the second request messages and the number of the second retransmission messages.

[0020] Optionally, based on the signaling statistics, calculating the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction further includes:

[0021] Based on the first retransmission message, determine the first retransmission count of the first link in the first direction;

[0022] If the number of retransmissions exceeds a preset threshold, the ratio of the number of retransmitted packets to the number of request packets is determined as the first overall packet loss rate of the first link in the first direction.

[0023] Based on the second retransmission message, determine the number of second retransmissions of the first link in the second direction;

[0024] If the second retransmission count is greater than the preset threshold, the ratio of the number of second retransmission packets to the number of second request packets is determined as the second overall packet loss rate of the first link in the second direction.

[0025] Optionally, based on the first overall packet loss rate and the second overall packet loss rate, the segmented packet loss rate of the first link is determined, including:

[0026] A set of equations is established for the overall packet loss rate and the segmented packet loss rate; the set of equations is established based on the assumption that the packet loss rate is the same when the same network path is transmitted in the first direction and the second direction.

[0027] Based on the set of equations, the first overall packet loss rate, and the second overall packet loss rate, the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link are determined.

[0028] Optionally, after determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, the method further includes:

[0029] If the packet loss rate of the first segment or the packet loss rate of the second segment is less than 0, the corresponding segment packet loss rate will be corrected to 0.

[0030] Optionally, after determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, the method further includes:

[0031] When the first overall packet loss rate or the second overall packet loss rate reaches a preset saturation value, the full packet data of the corresponding communication terminal under the first segment of the first link is obtained, and the total number of global packets and the total number of global retransmission packets of the first segment are calculated.

[0032] Based on the total number of global packets and the total number of global retransmitted packets, the macro packet loss rate of the first segmented link is calculated through the retransmission ratio.

[0033] Substituting the macroscopic packet loss rate into the system of equations, the segment packet loss rate of the second segment of the first link is calculated.

[0034] Secondly, embodiments of this application provide a network packet loss rate estimation device, comprising:

[0035] The first processing module is used to obtain signaling statistics data of the first link collected at a preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction.

[0036] The second processing module is used to calculate, based on the signaling statistics, the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction.

[0037] The third processing module is used to determine the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

[0038] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in any one of the first aspects.

[0039] Fourthly, embodiments of this application provide a computer program product, including computer instructions that, when executed by a processor, implement the steps of the method as described in any one of the first aspects.

[0040] Compared with existing technologies, the network packet loss rate estimation method, apparatus, medium, and product provided in this application embodiment sets up collection points in the link, independently counts the number of request packets and retransmissions of the first link in the first direction and the second direction, and calculates the first and second overall packet loss rates respectively. Since the overall packet loss rate is determined by the superposition of the losses of the two segmented links on both sides of the collection point, the bidirectional independent measurement provides two independent mathematical constraints. Based on the first overall packet loss rate and the second overall packet loss rate, the segmented packet loss rate of the segmented links of the first link is determined, wherein the segmented links include: the first segmented link between the first end and the preset collection point; and the second segmented link between the second end and the preset collection point. The true packet loss rates of the first and second segmented links can be accurately inverted and separated. Accurate quantitative estimation of link segments and fault node location can be achieved without deploying probes across the entire network, solving the problem that existing packet loss rate assessment schemes cannot accurately estimate quantitatively. Attached Figure Description

[0041] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0042] Figure 1 A flowchart illustrating the method for estimating network packet loss rate provided in this application embodiment;

[0043] Figure 2 This is a schematic diagram of the network quality analysis module provided in an embodiment of this application;

[0044] Figure 3 This is one of the schematic diagrams of the network signaling acquisition network architecture provided in the embodiments of this application;

[0045] Figure 4 A second schematic diagram of the network signaling acquisition architecture provided in the embodiments of this application;

[0046] Figure 5 This is a schematic diagram of the network packet loss rate estimation device provided in an embodiment of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and are not used to describe a specified order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, a first object can be one or more. Furthermore, in the specification and claims, "and" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0049] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc.; an indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0050] To enable those skilled in the art to better understand the embodiments of this application, the following description is provided first:

[0051] (1) In the scheme of link packet loss statistics based on message sending or receiving between devices A or B, for the network path to be detected between terminal A and terminal B, one end (such as switch A) can send messages, and the other end (such as switch B) can realize the link packet loss statistics by receiving messages: For example, if terminal A sends 10 messages and terminal B receives 8 messages, the estimated packet loss rate (LOSS) of the network path between A and B can be estimated based on this data as follows: =20%.

[0052] This method of calculating packet loss rate based on inter-device packet transmission and reception between segments A and B requires simultaneous packet tracking on both terminal A and terminal B. Based on the tracked packets, packets for packet loss rate testing are selected and analyzed. This necessitates deploying test environments on both terminals A and B and collaborating on testing, making it quite complex. Furthermore, this method cannot pinpoint the causes of poor network quality, thus failing to provide further guidance for network optimization or fault repair.

[0053] (2) In the scheme of performing link packet loss statistics based on the local device sending request messages and receiving response messages, if the messages transmitted between terminals A and B have a "response-acknowledgment mechanism" similar to the TCP protocol, this scheme requires that the messages transmitted between terminals A and B have a "response-acknowledgment mechanism": terminal A sends a message to terminal B, and terminal B replies with a response message after receiving the message. For example, if terminal A sends 10 messages and receives 6 response messages, then the estimated round-trip packet loss rate between A and B is 1-0.6=0.4 (40%). Assuming that the one-way link packet loss rate between A and B is LOSS, then the following formula is given: The estimated packet loss rate on the network path between terminals A and B can be obtained as follows: =22.5%.

[0054] While packet loss rate statistics based on sending request messages or receiving response messages by the local device eliminates the need for simultaneous packet tracking on two terminals, simplifying the implementation, this method is not very feasible in real-world multi-terminal environments. For example, if it is necessary to statistically analyze the overall network packet loss rate and other network quality data for a large network, this method would require deploying signaling collection on all terminals in the network. Similarly, this method can only count the packet loss phenomenon itself and cannot pinpoint the cause of network packet loss.

[0055] As described in the background section, existing technologies suffer from problems such as complex methods for calculating packet loss and the inability to quantitatively calculate packet loss rates. To address at least one of these problems, this application provides a method, apparatus, medium, and product for estimating network packet loss rates, which can reduce or avoid the occurrence of the above situations and solve the problem that existing packet loss rate assessment schemes cannot accurately estimate quantitatively.

[0056] This application provides a method, apparatus, medium, and product for estimating network packet loss rate. The method and apparatus are based on the same concept, and since the principles underlying the problem-solving are similar, their implementations can be referred to interchangeably; repeated details will not be repeated.

[0057] Please refer to Figure 1 This application provides a method for estimating network packet loss rate, comprising:

[0058] Step 11: Obtain signaling statistics data of the first link collected at the preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction.

[0059] Specifically, the preset collection point is the location of the signaling collection probe deployed in the network. It can read the raw signaling data of the two-way communication between the first communication terminal (denoted as terminal A) and the second communication terminal (denoted as terminal B) at the preset collection point from an external internet log collection platform. The raw signaling data specifically includes the total number of request messages sent from terminal A to terminal B (denoted as...). The number of request messages retransmitted from end A to end B (denoted as ) ), and the total number of request messages sent from B to A (denoted as ). The number of request messages retransmitted from B to A (denoted as ) ).

[0060] Step 12: Based on the signaling statistics, calculate the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction.

[0061] Using the preset collection point as the starting point for statistics, based on the original signaling data obtained in step 11, the first overall packet loss rate of the path from the first communication terminal to the second communication terminal, and the second overall packet loss rate of the path from the second communication terminal to the first communication terminal are calculated respectively, that is, the first overall packet loss rate of the path from terminal A to terminal B is calculated (denoted as...). ), and the second overall packet loss rate of the path from B to A (denoted as ). The calculation process is based on the correlation between the total number of request packets and the number of retransmissions, combined with the statistical patterns of packet retransmissions, to achieve a quantitative estimate of the overall packet loss rate.

[0062] Step 13: Determine the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

[0063] The paths on both sides of the preset collection point are denoted as the first side path (network 1) and the second side path (network 2), respectively. The data analysis submodule is based on the calculations obtained in step 12. and A system of logical mapping equations is constructed to relate the overall packet loss rate to the segmented packet loss rate. By solving the system of equations, the first segmented packet loss rate of the first side path (denoted as ) is obtained. ) and the second segment packet loss rate of the second side path (denoted as ) This enables independent quantitative estimation of segmented packet loss rate.

[0064] In this application, based on the result obtained in step 13 and By establishing a mapping relationship between link quality and segmented packet loss rate, or by setting a link judgment threshold, the link quality of the paths on both sides of the preset collection point is evaluated, thereby determining the network fault result and realizing the delineation and location of the faulty link.

[0065] Optionally, the network packet loss rate estimation method described in this application can be based on a preset network signaling acquisition probe deployment mode, and can achieve quantitative estimation of segmented network packet loss rate and delineation and location of faulty links through a built-in or external "network quality analysis module". (Refer to...) Figure 2 As shown, the network quality analysis module includes a data reading submodule, a data analysis submodule, and a fault location submodule. These submodules work together to complete the reading, analysis, and fault location of the original signaling data.

[0066] Optionally, step 12 above includes:

[0067] Based on the signaling statistics, the number of first request messages and the number of first retransmission messages in the first direction of the first link are obtained;

[0068] Based on the number of the first request messages and the number of the first retransmission messages, calculate the first overall packet loss rate of the first link in the first direction;

[0069] Based on the signaling statistics, obtain the number of second request messages and the number of second retransmission messages in the second direction of the first link;

[0070] Based on the number of the second request messages and the number of the second retransmission messages, calculate the second overall packet loss rate of the first link in the second direction.

[0071] The second overall packet loss rate of the path from the second communication terminal to the first communication terminal is calculated based on the number of the second request messages and the number of the second retransmission messages.

[0072] In this embodiment of the application, a preset collection point is used as the statistical starting point to extract the first request message (i.e., the message sent from end A to end B) from the original signaling data. ) and the first repeater message (i.e. ),in This represents the total number of request messages sent from endpoint A to endpoint B. This represents the total number of retransmitted messages in this direction. Based on the obtained... and Based on the statistical patterns of message retransmission, the first overall packet loss rate is derived. The estimation formula is as follows. For details, please refer to... Figure 3 The network signaling acquisition architecture shown assumes that within a certain period of time, the acquisition probe receives a total of [number] signals sent from end A to end B. A request message, in which a detection Each message is a retransmitted message. Let the overall packet loss rate of the message sent from point A to point B along the network transmission path be calculated using the collection point as the starting point for the statistics. Assuming a message can be retransmitted a maximum of n times, the total number of retransmitted messages N should be equal to the number of non-retransmitted messages MN in the M request messages, based on the overall packet loss rate of the path. The sum of the expected number of packets that undergo one retransmission + two retransmissions + ... + infinitely many retransmissions, i.e.: , formula (1).

[0073] Using the same method as described above for determining the first overall packet loss rate, the second request message (i.e., the message sent from B to A) is extracted from the original signaling data. ) and second transmission message (i.e. ),in This represents the total number of request messages sent from client B to client A. This represents the total number of retransmitted packets in this direction. The same calculation logic as used to determine the first overall packet loss rate is employed, based on... and This gives us the second overall packet loss rate. The formula allows for the calculation of the second overall packet loss rate of the path from the second communication terminal to the first communication terminal based on the number of the second request packets and the number of the second retransmission packets. Here, the calculation process is based on the correlation between the total number of request packets and the number of retransmissions, combined with the statistical patterns of packet retransmissions, to achieve a quantitative estimate of the overall packet loss rate.

[0074] Furthermore, based on the original signaling data, and taking the preset collection point as the statistical starting point, the first overall packet loss rate of the path from the first communication terminal to the second communication terminal, and the second overall packet loss rate of the path from the second communication terminal to the first communication terminal are calculated respectively, further including:

[0075] Based on the first retransmission message, determine the first retransmission count of the first link in the first direction;

[0076] If the number of retransmissions exceeds a preset threshold, the ratio of the number of retransmitted packets to the number of request packets is determined as the first overall packet loss rate of the first link in the first direction.

[0077] Based on the second retransmission message, determine the number of second retransmissions of the first link in the second direction;

[0078] If the second retransmission count is greater than the preset threshold, the ratio of the number of second retransmission packets to the number of second request packets is determined as the second overall packet loss rate of the first link in the second direction.

[0079] In this embodiment, based on the first retransmission message, the number of first retransmissions of the request message sent from end A to end B (i.e., the average number of retransmissions or the maximum number of retransmissions per message) is counted, and a preset threshold is set. When the number of first retransmissions exceeds the preset threshold, it indicates that the message retransmission behavior can truly reflect the network packet loss situation, and at this time, a preset threshold is adopted. ≈ / As the first overall packet loss rate, assuming a maximum retransmission count of n for a packet, when n is sufficiently large (e.g., the number of terminal retransmissions ≥ 3 in the TCP protocol), the relationship between the number of retransmitted packets, the number of request packets, and the overall packet loss rate can be approximately simplified, thus yielding... ≈ / If the number of first retransmissions is less than or equal to a preset threshold, it needs to be corrected using other statistical methods to ensure the accuracy of the estimate. Using the same logic, based on the second retransmission message, count the number of second retransmissions of the request message sent from B to A. When the number of second retransmissions exceeds a preset threshold, use... As a second overall packet loss rate, it ensures the reliability of packet loss rate estimation.

[0080] Specifically, refer to Figure 3 The network signaling acquisition architecture shown assumes that within a certain period of time, the acquisition probe receives a total of [number] signals sent from end A to end B. A request message, in which a detection Each message is a retransmitted message. Let the overall packet loss rate of the message sent from point A to point B along the network transmission path be calculated using the collection point as the starting point for the statistics. Assuming a message can be retransmitted a maximum of n times, the total number of retransmitted messages N should be equal to the number of non-retransmitted messages MN in the M request messages, based on the overall packet loss rate of the path. The sum of the expected number of packets that undergo one retransmission + two retransmissions + ... + infinitely many retransmissions, because Less than 1, when the number of retransmissions n is sufficiently large (taking the TCP protocol as an example, the number of retransmissions for most terminals is greater than or equal to 3). As the value approaches 0, the above formula (1) can be approximated as follows: Formula (1a); Based on the above formula (1a), we can obtain: , formula (2).

[0081] Using the same method as described above for determining the first overall packet loss rate, the second request message (i.e., the message sent from B to A) is extracted from the original signaling data. ) and second transmission message (i.e. ),in This represents the total number of request messages sent from client B to client A. This represents the total number of retransmitted packets in this direction. The same calculation logic as used to determine the first overall packet loss rate is employed, based on... and The second overall packet loss rate was obtained. The estimation formula is: the overall packet loss rate of a message sent from end B to end A along the network transmission path, with the collection point as the starting point of the statistics. It can be estimated based on the following formula: Formula (3), where, It is the number of request messages collected by the collection point from B to A. It is the number of retransmission messages in the request message sent by B to A collected by the collection point.

[0082] Optionally, step 13 above includes:

[0083] A set of equations is established to represent the overall packet loss rate and the segmented packet loss rate. This set of equations is based on the assumption that the packet loss rate is consistent across the same network path in both the first and second directions. Specifically, a set of equations is constructed to describe the logical mapping relationship between the overall packet loss rate and the segmented packet loss rates of the paths on both sides of a preset collection point. This logical mapping relationship is based on the assumption that the packet loss rates of request and response messages are consistent across the same network path. Here, the first direction can be selected as the forward direction of the first link, and the second direction can be selected as the reverse direction of the first link.

[0084] Based on the set of equations, the first overall packet loss rate, and the second overall packet loss rate, the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link are determined.

[0085] It should be noted that, using the same scenario as an example: within a certain period of time, the acquisition probe receives a total of [number] data sent from end A to end B. A request message, in which a detection This message is a retransmitted message. At this point, let's assume... All retransmitted messages were due to A not receiving a response message within the timeout period, and were retransmitted accordingly. (See reference...) Figure 3 The signaling acquisition architecture diagram shows that terminal A did not receive a response message, which may be due to the following three reasons:

[0086] 1) The request message sent by terminal A was lost in the "Network 2" path on the right side of the collection point;

[0087] 2) Packet loss occurred in the response message from B in the "Network 2" path to the right of the collection point;

[0088] 3) Packet loss occurred in the response message from B in the "Network 1" path on the left side of the collection point;

[0089] Given that the packet loss rate depends on the health of the network itself, assuming that for the same network path, request and response packets have the same packet loss rate, we can let:

[0090] 1) The packet loss rate of the request message sent by end A and the response message replied by end B in the "Network 2" path to the right of the collection point is 0. 2) The packet loss rate of the response message from B in the "Network 1" path to the left of the collection point is... ;

[0091] Then we have the following system of equations containing logical operations:

[0092] Formula (4); where, the operator The conversion relationship between the "OR" operation and arithmetic operations is as follows: Formula (4a); then formula (4) can be expressed by the following arithmetic operations:

[0093] , formula (5).

[0094] For the sake of simplicity, we will use the following methods respectively: , , , Abbreviated as: , , , Formula (5) can be written as: , formula (6).

[0095] Based on the above system of equations, the following transformation is performed:

[0096] , formula (6-a);

[0097] Formula (6-b);

[0098] If both LA and LB are less than 1, dividing the two equations (6-b) gives:

[0099] Formula (6-c);

[0100] Substituting (6-c) into (6-b), we get:

[0101] Formula (6-d);

[0102] Substituting the packet loss rate estimation formulas from formulas (2) and (3) above into the formula above and simplifying it, we can obtain the packet loss rate of the first segment. Second segment packet loss rate The estimation formula for 2 can be used to obtain... Figure 3 Under the generalized network architecture shown, the estimation formulas for segmented packet loss rates for "Network 1" on the left and "Network 2" on the right of the collection point are as follows:

[0103] Formula (7); where: .

[0104] Based on formula (7), segmented independent and quantitative numerical estimation of packet loss rate of links under different network paths (network 1 or network 2) can be realized in the signaling acquisition architecture of the existing network, thereby solving the problems existing in the current technology.

[0105] Furthermore, the estimation method based on formula (7) has two problems: Problem 1: The actual packet retransmission behavior deviates too much from the inherent packet loss rate of the network. The link packet loss rate estimation formula obtained by formula (7) is essentially a reverse derivation of the inherent packet loss rate of the network based on the network packet loss behavior that has already occurred. Since there is a deviation between a single network packet loss behavior and the inherent packet loss rate of the network, it may lead to significant errors in the estimation results under certain circumstances. Consider Figure 3 The network architecture used in the proposed solution has the following characteristics: Figure 4 The message transmission scenario shown:

[0106] According to formula (7), the packet loss rates of network 1 and network 2 can be estimated as follows:

[0107] Formula (8); It should be noted that in the above estimation results, the packet loss rate of the first segment is... A negative value clearly deviates from reality, the root cause being that... / , / The numerical sample combination does not conform to the packet loss rate of the first segment of any network 1. Or the packet loss rate of the second segment of network 2 Expected network retransmission under the given value.

[0108] Furthermore, to address the problems existing in the aforementioned special scenarios, this application, after determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, further includes:

[0109] If the packet loss rate of the first segment or the packet loss rate of the second segment is less than 0, the corresponding segment packet loss rate will be corrected to 0.

[0110] In this embodiment of the application, since the packet loss rate is a non-negative value and cannot be less than 0, the calculated first segment packet loss rate is... Or the packet loss rate of the second segment Make a judgment: If <0 or <0 indicates that the sample combination of the current original signaling data does not meet the expected situation of network packet loss, there is a statistical bias or there is no packet loss in the current segment of the network. At this time, the corresponding segment packet loss rate should be corrected to 0. At this time, the estimated packet loss rate value that is less than 0 should be corrected to 0. At this time, formula (7) is further adjusted to:

[0111] Formula (9); that is: Formula (9a).

[0112] This correction step ensures that the estimated segmented packet loss rate matches the actual network scenario, thus improving the accuracy of the estimation.

[0113] Optionally, after determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, the method further includes:

[0114] When the first overall packet loss rate or the second overall packet loss rate reaches a preset saturation value, the full packet data of the corresponding communication terminal under the first segment of the first link is obtained, and the total number of global packets and the total number of global retransmission packets of the first segment are calculated.

[0115] Based on the total number of global packets and the total number of global retransmitted packets, the macro packet loss rate of the first segmented link is calculated through the retransmission ratio.

[0116] Substituting the macroscopic packet loss rate into the system of equations, the segment packet loss rate of the second segment of the first link is calculated.

[0117] In this embodiment of the application, when the preset saturation value is optionally set to 1, and the first overall packet loss rate or the second overall packet loss rate reaches the preset saturation value, that is, the formula for... = (Right now =1), or = (Right now In the scenario where =1), formula (6-b) cannot be derived to formula (6-c), but the following conditional relationship can be obtained: Formula (11), in order to accurately estimate the other value of L2 and L1 when one of L1 and L2 is close to 1, let the total number of messages received from the left link of the collection point and sent to the right link / the number of retransmitted messages be M in the total message statistics collected by the signaling acquisition probe. L / N L The total number of messages received from the right link of the acquisition point and the number of retransmitted messages sent to the left link are M respectively. R / N R ,by Figure 4 The diagram shows multiple pairs of transmit / receive ends A. (1) / B (1) In a network structure of A2 / B2, A3 / B3..., M L / N L、 M R / N R It can be expressed by the following formula: , Formula (12), where n is the number of communication terminal pairs. The same calculation logic as above for determining the overall packet loss rate is used, based on... and Calculate the macroscopic packet loss rate of the first-side path, based on and Calculate the macroscopic packet loss rate of the second-side path, i.e., macroscopic packet loss rate ≈ total number of retransmitted packets / total number of packets. When When =1, it can be determined =1, substituting the macroscopic packet loss rate of the first side path into the simplified equation system, and then calculating the second side path... ;when When =1, it can be determined =1, substituting the macroscopic packet loss rate of the second-side path into the simplified equation system, we can calculate the first-side path. .

[0118] Specifically, referring to the formula (9a) above, the network quality of L1 / L2 can be estimated based on the overall packet retransmission behavior of the entire network, then N A =M A or N B =M B Then, formula (11) further becomes:

[0119] , formula (13).

[0120] After applying the modified schemes of formulas (9a) and (12), formula (7) is further adjusted to:

[0121] , formula (14).

[0122] Optionally, after step 13 above, the method further includes:

[0123] Based on the segmented packet loss rate, a mapping function relationship between link quality and the corresponding link packet loss rate is established; the larger the value of the link quality, the better the link quality.

[0124] The network fault result is determined based on the mapping function relationship;

[0125] Alternatively, a link judgment threshold can be set. If the segment packet loss rate is greater than or equal to the link judgment threshold, the network fault result is determined to be abnormal link quality; if the segment packet loss rate is less than the link judgment threshold, the network fault result is determined to be normal link quality.

[0126] In this embodiment, the specific process of "determining the network fault result" is detailed. This application provides two specific implementation methods, as follows:

[0127] Method 1: Determining network fault results based on a link quality mapping function. Based on the segmented packet loss rate, establish a mapping function relationship between link quality and the corresponding link packet loss rate; the higher the link quality value, the better the link quality. (Based on segmented packet loss rate) and Establish a mapping function between link quality and packet loss rate. The link quality value is bounded and positively correlated with the quality of the link. An example mapping function is as follows: ,in, For the link quality of the first side path, The link quality of the second-side path is represented by a value ranging from 0 to 1. When LOSS=0, Q=1 (optimal link quality); when LOSS=1, Q=0 (worst link quality). This value is calculated using a mapping function. and Based on the link quality value, the network quality of both paths is evaluated, and the fault result is determined: the lower the link quality value, the more severe the packet loss on the corresponding path and the higher the probability of failure.

[0128] Specifically, in the first method described above, the optimized link packet loss estimation formula obtained based on the above formula (13) can be further implemented. Figure 3 The segment quality of the links in different network path segments (Network 1, Network 2). Based on different actual business scenarios, the link quality Q1 / Q2 of Network Path 1 / Network Path 2 can be expressed as a function of the corresponding path segment packet loss rate LOSS1 / LOSS2: If the link quality Q is bounded, and the value of Q is proportional to the link quality, then a functional relationship between link quality Q and packet loss rate (LOSS) can be expressed as (but is not limited to): In this example, the value of Q ranges from 0 to 1. When LOSS equals 0, the link quality is optimal, which is 1; when LOSS equals 1, the link quality is worst, which is 0.

[0129] Method 2: Determining network fault results based on link judgment thresholds. This application can set link judgment thresholds (e.g., Q1T, Q2T) for the first and second side paths respectively, according to actual business scenario requirements. These thresholds can be flexibly adjusted based on network service quality requirements. The segmented packet loss rate is compared with the corresponding link judgment threshold (e.g., Q1T, Q2T). For example, if... If the value is ≥Q1T, then the link quality of the first-side path is determined to be abnormal, and troubleshooting is required; if If the value is less than Q1T, then the link quality of the first-side path is considered normal; similarly, according to... The comparison results with Q2T determine the link quality and fault status of the second-side path, and finally output the link packet loss rate information and fault location and delimitation results.

[0130] In one specific embodiment, reference is made to Figure 4 The network architecture shown is Figure 4 The network architecture shown uses preset data collection points (here) Figure 4 The network is divided into two sections, network 1 (represented by the optical splitting acquisition point) and network 2 (represented by the optical splitting acquisition point). Network 1 is located on the first communication end (A end), and network 2 is located on the second communication end (B end). The network includes multiple pairs of concurrent communication terminals, such as A2, A3, B2, and B3, to simulate a multi-user service scenario in a real network. The optical splitting acquisition point is deployed between the two network sections and collects all raw signaling data of the bidirectional communication between A end and B end using a lossless optical splitting method. This process does not interfere with the normal transmission of service traffic and relies on preset acquisition points to complete the acquisition of raw signaling data for bidirectional communication between the two ends.

[0131] Secondly, the optical splitting acquisition point classifies and statistically analyzes the bidirectional message data: for the uplink direction from A to B, it collects and records the total number of uplink request messages MA and the number of uplink retransmission messages NA, where MA=10 and NA=1 in the example; for the downlink direction from B to A, it collects and records the total number of downlink request messages MB and the number of downlink retransmission messages NB, where MB=10 and NB=9 in the example. Simultaneously, the acquisition point collects message and retransmission data from multiple pairs of terminals, such as A2, A3, and B2, B3, forming the full message data for Network 1 and Network 2, providing data support for subsequent correction estimations in abnormal scenarios. The raw signaling data covers the total number of request messages sent between the two ends and the number of retransmission messages in the corresponding transmission direction. It also aggregates the full message information of all communication terminals under a single link at the acquisition point, ensuring the computational correction needs under abnormal operating conditions.

[0132] Next, the collected raw signaling data is uploaded to the Internet access log collection platform, generating standardized Internet access log records and completing data storage and preprocessing. Subsequently, the network quality analysis module reads the data from the platform and executes the core estimation process: taking the optical splitting collection point as the statistical starting point, it calculates the overall packet loss rate corresponding to the two transmission directions by combining the number of bidirectional request packets and retransmission packets; based on the premise that the packet loss rate of request packets and response packets in the same network path is consistent, it establishes a system of correlation equations between the overall packet loss rate and the segmented packet loss rate on both sides of the collection point, and solves the equations to obtain the segmented packet loss rate corresponding to the two network segments. For the abnormal scenario of excessively high downlink and downlink retransmission ratios in the example, it retrieves full packet data from multiple terminals to calculate the macroscopic packet loss rate of the link, thereby correcting unreasonable estimates generated in conventional calculations and ensuring that the packet loss rate calculation results are consistent with the actual network operating status.

[0133] Finally, the network quality analysis module combines the solved segmented packet loss rate to establish a mapping relationship between link quality and packet loss parameters, or configure link judgment threshold standards to comprehensively evaluate the link operation quality of network 1 and network 2 respectively, output network anomaly judgment results, and accurately complete the division and location of fault paths.

[0134] This application Figure 4 This paper fully demonstrates the entire process of the method in this application, from data acquisition, preprocessing, packet loss rate estimation and correction to fault location. It embodies "single-point acquisition, bidirectional estimation, and segmented delimitation," and solves the technical problems of complex data statistics, inability to make quantitative estimations, and difficulty in locating fault links in the prior art.

[0135] In summary, this application addresses the pain points of existing network packet loss rate assessments, such as complex statistics, inability to quantitatively estimate, and difficulties in fault location. Based on existing network signaling acquisition probes, it integrates or adds a network quality analysis module. Through algorithmic formulas and mathematical modeling, it achieves link packet loss rate estimation based on packet retransmission information, accurate estimation of segmented path packet loss rate (including optimization and correction), and can determine network fault results based on packet loss rate. Compared to existing technologies, this application only requires centralized deployment of probes at a single point, reducing statistical complexity and enabling quantitative estimation of packet loss rate and segmented delineation of link faults, aligning with the needs of 5G-A experience operation and tiered network protection.

[0136] The various methods of the embodiments of this application have been described above. Apparatus for implementing the above methods will now be provided.

[0137] Please refer to Figure 5 This application also provides a network packet loss rate estimation device, comprising:

[0138] The first processing module 51 is used to obtain signaling statistics data of the first link collected at a preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction.

[0139] The second processing module 52 is used to calculate, based on the signaling statistics, the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction.

[0140] The third processing module 53 is used to determine the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

[0141] Optionally, the above-mentioned device further includes:

[0142] The fourth processing module is used to establish a mapping function relationship between link quality and corresponding link packet loss rate based on the segmented packet loss rate; the larger the value of the link quality, the better the link quality.

[0143] The fifth processing module is used to determine the network fault result based on the mapping function relationship;

[0144] Alternatively, the sixth processing module is used to set a link judgment threshold. If the segment packet loss rate is greater than or equal to the link judgment threshold, the network fault result is determined to be an abnormal link quality; if the segment packet loss rate is less than the link judgment threshold, the network fault result is determined to be a normal link quality.

[0145] Optionally, the second processing module 52 described above is specifically used for:

[0146] Based on the signaling statistics, the number of first request messages and the number of first retransmission messages in the first direction of the first link are obtained;

[0147] Based on the number of the first request messages and the number of the first retransmission messages, calculate the first overall packet loss rate of the first link in the first direction;

[0148] Based on the signaling statistics, obtain the number of second request messages and the number of second retransmission messages in the second direction of the first link;

[0149] Based on the number of the second request messages and the number of the second retransmission messages, calculate the second overall packet loss rate of the first link in the second direction.

[0150] Optionally, the second processing module 52 described above is further specifically used for:

[0151] Based on the first retransmission message, determine the first retransmission count of the first link in the first direction;

[0152] If the number of retransmissions exceeds a preset threshold, the ratio of the number of retransmitted packets to the number of request packets is determined as the first overall packet loss rate of the first link in the first direction.

[0153] Based on the second retransmission message, determine the number of second retransmissions of the first link in the second direction;

[0154] If the second retransmission count is greater than the preset threshold, the ratio of the number of second retransmission packets to the number of second request packets is determined as the second overall packet loss rate of the first link in the second direction.

[0155] Optionally, the aforementioned third processing module 53 is specifically used for:

[0156] A set of equations is established for the overall packet loss rate and the segmented packet loss rate; the set of equations is established based on the assumption that the packet loss rate is the same when the same network path is transmitted in the first direction and the second direction.

[0157] Based on the set of equations, the first overall packet loss rate, and the second overall packet loss rate, the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link are determined.

[0158] Optionally, the apparatus of this application further includes:

[0159] The seventh processing module is used to correct the corresponding segment packet loss rate to 0 when the first segment packet loss rate or the second segment packet loss rate is less than 0.

[0160] Optionally, the apparatus of this application further includes:

[0161] The eighth processing module is used to obtain the full message data of the corresponding communication terminal under the first segment link of the first link when the first overall packet loss rate or the second overall packet loss rate reaches a preset saturation value, and to summarize and calculate the total number of global messages and the total number of global retransmission messages of the first segment link.

[0162] The ninth processing module is used to calculate the macro packet loss rate of the first segmented link based on the global total number of packets and the global total number of retransmitted packets through the retransmission ratio relationship.

[0163] The tenth processing module is used to substitute the macroscopic packet loss rate into the system of equations to calculate the segment packet loss rate of the second segment of the first link.

[0164] It should be noted that the device in this embodiment corresponds to the method described above, and the implementation methods in each of the above embodiments are applicable to the embodiments of this device, achieving the same technical effect. The device provided in this application embodiment can implement all the method steps implemented in the above method embodiments and achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiments will not be described in detail here.

[0165] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described network packet loss rate estimation method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0166] This application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the various processes of the above-described network packet loss rate estimation method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0167] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0168] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0169] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for estimating network packet loss rate, characterized in that, include: Obtain signaling statistics data of the first link collected at a preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction. Based on the signaling statistics, calculate the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction. Based on the first overall packet loss rate and the second overall packet loss rate, the segmented packet loss rate of the segmented link of the first link is determined, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

2. The method according to claim 1, characterized in that, After determining the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, the method further includes: Based on the segmented packet loss rate, a mapping function relationship between link quality and the corresponding link packet loss rate is established; the larger the value of the link quality, the better the link quality. The network fault result is determined based on the mapping function relationship; Alternatively, a link judgment threshold can be set. If the segment packet loss rate is greater than or equal to the link judgment threshold, the network fault result is determined to be abnormal link quality; if the segment packet loss rate is less than the link judgment threshold, the network fault result is determined to be normal link quality.

3. The method according to claim 1, characterized in that, Based on the signaling statistics, calculate the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction, including: Based on the signaling statistics, the number of first request messages and the number of first retransmission messages in the first direction of the first link are obtained; Based on the number of the first request messages and the number of the first retransmission messages, calculate the first overall packet loss rate of the first link in the first direction; Based on the signaling statistics, obtain the number of second request messages and the number of second retransmission messages in the second direction of the first link; Based on the number of the second request messages and the number of the second retransmission messages, calculate the second overall packet loss rate of the first link in the second direction.

4. The method according to claim 3, characterized in that, Based on the signaling statistics, the calculation of the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction further includes: Based on the first retransmission message, determine the first retransmission count of the first link in the first direction; If the number of retransmissions exceeds a preset threshold, the ratio of the number of retransmitted packets to the number of request packets is determined as the first overall packet loss rate of the first link in the first direction. Based on the second retransmission message, determine the number of second retransmissions of the first link in the second direction; If the second retransmission count is greater than the preset threshold, the ratio of the number of second retransmission packets to the number of second request packets is determined as the second overall packet loss rate of the first link in the second direction.

5. The method according to claim 1, characterized in that, Based on the first overall packet loss rate and the second overall packet loss rate, the segmented packet loss rate of the segmented link of the first link is determined, including: A set of equations is established for the overall packet loss rate and the segmented packet loss rate; the set of equations is established based on the assumption that the packet loss rate is the same when the same network path is transmitted in the first direction and the second direction. Based on the set of equations, the first overall packet loss rate, and the second overall packet loss rate, the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link are determined.

6. The method according to claim 5, characterized in that, After determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, the method further includes: If the packet loss rate of the first segment or the packet loss rate of the second segment is less than 0, the corresponding segment packet loss rate will be corrected to 0.

7. The method according to claim 5, characterized in that, After determining the first segment packet loss rate and the second segment packet loss rate of the segmented link of the first link based on the equation set, the first overall packet loss rate, and the second overall packet loss rate, the method further includes: When the first overall packet loss rate or the second overall packet loss rate reaches a preset saturation value, the full packet data of the corresponding communication terminal under the first segment of the first link is obtained, and the total number of global packets and the total number of global retransmission packets of the first segment are calculated. Based on the total number of global packets and the total number of global retransmitted packets, the macro packet loss rate of the first segmented link is calculated through the retransmission ratio. Substituting the macroscopic packet loss rate into the system of equations, the segment packet loss rate of the second segment of the first link is calculated.

8. A device for estimating network packet loss rate, characterized in that, include: The first processing module is used to obtain signaling statistics data of the first link collected at a preset collection point; wherein, the first link includes a first end and a second end, the preset collection point is located in the first link, and the signaling statistics data includes the total number of request packets in the first direction and the second direction of the first link and the number of retransmissions of request packets in the corresponding direction. The second processing module is used to calculate, based on the signaling statistics, the first overall packet loss rate of the first link in the first direction and the second overall packet loss rate of the first link in the second direction. The third processing module is used to determine the segmented packet loss rate of the segmented link of the first link based on the first overall packet loss rate and the second overall packet loss rate, wherein the segmented link includes: a first segmented link between the first end and the preset collection point; and a second segmented link between the second end and the preset collection point.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images