Method for determining logical connection and related equipment

文档序号：814624 发布日期：2021-03-26 浏览：14次中文

阅读说明：本技术 一种确定逻辑连接的方法和相关设备 (Method for determining logical connection and related equipment ) 是由叶瑞杨金鑫匡立马军棋李祥冯皓宇于 2019-09-10 设计创作，主要内容包括：本发明公开了一种确定逻辑连接的方法、网络控制器和系统,该方法包括：确定与目标单板可能具有逻辑连接的候选单板集合；分别计算目标单板与每一个候选单板之间存在逻辑连接的概率；根据计算出的概率从多个候选单板中确定待调测的候选单板,通过控制指令调节目标单板或者目标单板上游的功率,当目标单板的其中一个物理端口与待调测的候选单板的其中一个物理端口的功率变化趋势一致时,确定目标单板上的其中一个物理端口和待调测的候选单板的其中一个第二物理端口之间需要增加逻辑连接。本申请通过机器学习或人工智能算法计算候选单板与目标单板之间存在逻辑连接的概率,并依据概率进行功率调测,可以提高逻辑连接修复的效率和准确性。(The invention discloses a method, a network controller and a system for determining logical connection, wherein the method comprises the following steps: determining a candidate single board set which may have logical connection with a target single board; respectively calculating the probability of logical connection between the target single board and each candidate single board; and determining a candidate single board to be tested from the multiple candidate single boards according to the calculated probability, adjusting the power of the target single board or the upstream of the target single board through a control instruction, and determining that logical connection needs to be added between one physical port on the target single board and one second physical port of the candidate single board to be tested when the power change trend of one physical port of the target single board is consistent with that of one physical port of the candidate single board to be tested. The method and the device calculate the probability of the logical connection between the candidate single board and the target single board through a machine learning or artificial intelligence algorithm, and carry out power regulation and measurement according to the probability, so that the efficiency and the accuracy of logical connection restoration can be improved.)

1. A method of determining a logical connection, the method comprising:

determining a candidate single board set of a target single board, wherein the target single board comprises at least one first physical port, the candidate single board set comprises at least one candidate single board, and each candidate single board in the at least one candidate single board comprises at least one second physical port;

respectively calculating the probability that the logical connection exists between one of the first physical ports of the target single board and each of the second physical ports of each of the candidate single boards, where the logical connection includes a correspondence between an identifier of one of the first physical ports and an identifier of one of the second physical ports;

determining a candidate single board to be tested from the multiple candidate single boards according to the probability, adjusting the power of the target single board or the upstream of the target single board through a control instruction, and determining that the logical connection needs to be added between one of the first physical ports on the target single board and one of the second physical ports of the candidate single boards to be tested when the power change trend of one of the first physical ports of the target single board is consistent with that of one of the second physical ports of the candidate single boards to be tested.

2. The method according to claim 1, wherein a probability that a logical connection exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board is greater than or equal to a preset threshold.

3. The method according to claim 1 or 2, wherein a probability that a logical connection exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board is highest.

4. The method according to any of claims 1-3, wherein the determining the candidate veneer set of the target veneer comprises:

determining a candidate single board set of the target single board according to a connection constraint rule, where the connection constraint rule includes a type of the candidate single board and/or a type of at least one second physical port included in the candidate single board.

5. The method according to any of claims 1-4, wherein said candidate board comprises at least one said second physical port with missing logical connection.

6. The method according to any one of claims 1 to 5, characterized in that the method comprises: the at least one second physical port may be logically connected to the at least one first physical port of the target board.

7. The method according to any of claims 1-6, wherein after said determining the candidate veneer set of the target veneer, said method further comprises:

respectively querying the power of one of the first physical ports of the target board and the power of each of the second physical ports of each of the candidate boards, and deleting the candidate board from the set of candidate boards when the power of one of the first physical ports of the target board and the power of any of the second physical ports of any of the candidate boards are both greater than zero and the power variation trends of the first physical ports of the target board and the power of any of the second physical ports of any of the candidate boards are inconsistent.

8. The method according to any of claims 1 to 7, wherein said separately calculating the probability that the logical connection exists between one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards comprises:

and respectively calculating the probability of the logical connection between one first physical port of the target single board and each second physical port of each candidate single board through a machine learning model, wherein the machine learning model is obtained based on characteristic parameter training of the full-network logical connection.

9. The method of claim 8, wherein the characteristic parameters comprise one or more of a board name, a board type, a board port, a slot, a subrack, a network element ID, and a site ID.

10. The method according to any one of claims 1 to 9, wherein when a power variation trend of one of the first physical ports of the target board is consistent with a power variation trend of one of the second physical ports of at least two candidate boards to be debugged, the method further comprises:

respectively calculating the absolute value of the difference between the power change of one first physical port of the target single board and the power change of one second physical port of the at least two candidate single boards to be measured;

determining that the logical connection needs to be added between one of the first physical ports on the target board and one of the second physical ports of the board to be debugged having the smallest absolute difference value.

11. A network controller, characterized in that the network controller comprises:

a processing module, configured to determine a candidate board set of a target board, where the target board includes at least one first physical port, the candidate board set includes a plurality of candidate boards, and each candidate board in the plurality of candidate boards includes at least one second physical port; the method further comprises determining a candidate single board to be tested from the plurality of candidate single boards; and is further configured to send a control instruction to adjust power of the target board or upstream of the target board

A calculation module, configured to calculate a probability that one of the first physical ports of the target board has a logical connection with one or more second physical ports of each candidate board, where the logical connection includes a correspondence between an identifier of the one of the first physical ports and an identifier of the one of the second physical ports;

a checking module, configured to determine that a logical connection needs to be added between one of the first physical ports on the target board and one of the second physical ports of the candidate board to be measured when power variation trends of the one of the first physical ports of the target board and the one of the second physical ports of the candidate board to be measured are consistent.

12. The network controller of claim 11, wherein the processing module is further configured to:

and determining the candidate single board to be debugged according to whether the probability that the logical connection exists between one second physical port of each candidate single board and one first physical port of the target single board is greater than or equal to a preset threshold value.

13. The network controller of claim 11 or 12, wherein the processing module is further configured to:

14. The network controller according to any of claims 11-13, wherein said candidate board comprises at least one said second physical port with missing logical connection.

15. The network controller according to any of claims 11-14, wherein the processing module is further configured to: respectively inquiring the power of one first physical port of the target single board and the power of each second physical port of each candidate single board;

the verification module is further configured to delete the candidate single board from the candidate single board set when the powers of both the first physical port of the target single board and the second physical port of any candidate single board are greater than zero, and the power variation trends of the first physical port of the target single board and the second physical port of any candidate single board are inconsistent.

16. The network controller of claim 15, wherein the check module is further configured to: and judging whether the candidate single board set only comprises one candidate single board.

17. The network controller according to any of claims 11-16, wherein the calculation module is further configured to:

the test module is further configured to determine that the logical connection needs to be added between one of the first physical ports on the target board and one of the second physical ports of the board to be tested, where the absolute value of the difference is the minimum.

18. A computer device, characterized in that the computer device comprises:

a memory for storing a computer program;

a processor for implementing the method of any one of claims 1 to 10 when executing the computer program.

19. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium for implementing the method according to any one of claims 1-10.

20. A network system, the network system comprising a network controller and at least one network device, the network controller comprising:

The at least one network device is configured to adjust the power of the target board or the upstream of the target board according to the control instruction sent by the network controller.

21. The network system of claim 20, wherein the processing module is further configured to:

22. The network system according to claim 20 or 21, wherein the processing module is further configured to:

23. The network system of any one of claims 20-22, wherein the verification module is further configured to:

when the power of one of the first physical ports of the target board and the power of any of the second physical ports of any of the candidate boards are both greater than zero, and the power variation trends of one of the first physical ports of the target board and any of the second physical ports of any of the candidate boards are not consistent, deleting the candidate board from the candidate board set.

Technical Field

The present invention relates to the field of optical communications technologies, and in particular, to a method and an apparatus for determining a logical fiber connection.

Background

The Wavelength Division network is an optical network adopting a Wavelength Division Multiplexing (WDM) transmission principle, and in the Wavelength Division network, signals are finally transmitted in the form of optical signals. For the service of city 1 to be transmitted to city 2, sites must be deployed in city 1, city 2 and other passing cities, and after the sites are deployed, the sites are connected by fiber cables.

In a wavelength division network, the number of fiber connections inside a site is as many as several hundreds or even thousands, and the fiber connections are connected to board ports with various functions.

Due to the reasons that the logic connection created by network opening is incomplete or part of the logic connection is damaged in the later operation and the like, the logic connection loss of the control side in the current wavelength division network is serious, and a large number of logic connection breakpoints exist. Currently, the missing connection is supplemented and the connection repair is carried out mainly depending on manual work, and the veneer ports of the opposite end of the breakpoint are confirmed according to the detailed connection relation among the veneer ports in the network construction drawing. According to the method, when the network construction drawing is incomplete, missing connections are filled according to drawing comparison, the efficiency is low, the time is consumed, if the deployment site does not have the construction drawing, the site needs to be manually placed to find the physical port corresponding to the logic connection breakpoint, the single board port of the opposite end is found through physical connection fibers, the labor cost is high, the time consumption is long, and the accuracy cannot be guaranteed.

Therefore, how to quickly repair missing logical connections, reduce labor cost, and ensure the accuracy of logical connections is an urgent problem to be solved.

Disclosure of Invention

In order to solve the problems of long time consumption and low accuracy rate caused by manual detection needed for repairing logical connection in the prior art, the embodiment of the invention provides a method for determining logical connection and related equipment. The technical scheme is as follows:

in a first aspect, the present invention provides a method for determining a logical connection judgment, the method being applied to a station, the method including: determining a candidate single board set of a target single board, wherein the target single board comprises at least one first physical port, the candidate single board set comprises a plurality of candidate single boards, and each candidate single board in the plurality of candidate single boards comprises at least one second physical port; respectively calculating the probability of logical connection between one first physical port of the target single board and each second physical port of each candidate single board, wherein the logical connection comprises the corresponding relation between the identifier of one first physical port and the identifier of one second physical port; and determining a candidate single board to be tested from the multiple candidate single boards according to the probability, adjusting the power of the target single board or the upstream of the target single board, and determining that logical connection needs to be added between one of the first physical ports on the target single board and one of the second physical ports of the candidate single boards to be tested when the power change trend of one of the first physical ports of the target single board is consistent with that of one of the second physical ports of the candidate single boards to be tested.

By the mode of combining the probability calculation and the power perturbation, the problems of long time consumption and low accuracy rate of manual logic connection in the prior art are solved.

The method and the device can determine the logical connection aiming at the scenes of logical connection loss (broken fibers) or logical connection errors. The technical scheme of the embodiment of the invention can be applied to a station, and at least an optical fiber is connected between the outbound port and the inbound port of the station. The method may be performed by a network controller, such as a Software Defined Network (SDN), a network manager, and the like.

In a possible design, a probability that a logical connection exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board is greater than or equal to a preset threshold.

In a possible design, the probability that the logical connection exists between one of the second physical ports of the candidate board to be tested and one of the first physical ports of the target board is the highest, so that the verification process can be simplified, and the logical connection between the target board and the candidate board can be determined more efficiently.

In one possible design, after calculating probabilities of logical connections between one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards, all the probabilities are arranged in order.

For example, probability calculation can be performed through an artificial intelligence or machine learning algorithm, power perturbation is performed according to the sequence of the probability, and the efficiency and the accuracy of logic connection repair are improved.

In a possible design, when a trend of power variation of one of the first physical ports of the target board is consistent with a trend of power variation of one of the second physical ports of at least two candidate boards to be tested, the method further includes:

In a possible design, the determining a candidate board set of a target board includes:

determining a candidate single board set of the target single board according to a connection constraint rule, where the connection constraint rule includes a type of a candidate single board of the target single board and/or a type of at least one second physical port included in the candidate single board.

Specifically, the connection constraint rule may include a board type and a port type that may have a connection relationship with a target board within the site, or include one of a board type and a port type that may have a connection relationship with a target board within the site. And determining a candidate veneer set according to the connection constraint rule, and primarily screening veneers possibly having logical connection with the target veneer, thereby further improving the efficiency of logical connection repair.

In one possible design, determining a candidate board set of the target board includes:

determining a candidate single board set of the target single board according to the missing data of the network logical connection, where the candidate single board includes at least one second physical port with missing logical connection.

For example, the candidate board set includes at least a part of boards in the target board site, and may also include all boards in the target board site except the target board.

In a possible design, before determining a candidate board to be measured from the plurality of candidate boards according to the probability and adjusting the power of the target board or the upstream of the target board, the method further includes:

respectively querying power of one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards, and when the power of one of the first physical ports of the target board and the power of any of the second physical ports of any of the candidate boards are both greater than zero, if the power variation trends of one of the first physical ports of the target board and any of the second physical ports of any of the candidate boards are inconsistent, deleting the candidate boards from the candidate board set.

In a possible design, the calculating the probability that the logical connection exists between one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards respectively includes:

In one possible design, the characteristic parameter includes one or more of a board name, a board type, a board port, a slot position, a subrack, a network element ID, and a site ID.

In a second aspect, an embodiment of the present invention provides a method for determining a logical connection, where the method includes: determining a candidate single board set of a target single board, wherein the target single board comprises at least one first physical port, the candidate single board set comprises a plurality of candidate single boards, and each candidate single board in the plurality of candidate single boards comprises at least one second physical port; respectively querying the power of one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards, and when the power of one of the first physical ports of the target board and the power of any of the second physical ports of any of the candidate boards are both greater than zero, if the power change trends of one of the first physical ports of the target board and any of the second physical ports of any of the candidate boards are inconsistent, deleting the candidate boards from the candidate board set; and when the candidate single board only comprises one candidate single board, confirming that the candidate single board and the target single board have a logical connection relation.

It should be noted that the candidate board set after the power verification usually includes a plurality of candidate boards.

In a possible design, when the candidate single board includes at least two candidate single boards, respectively calculating a probability that a logical connection exists between one of the first physical ports of the target single board and each of the second physical ports of each of the candidate single boards, where the logical connection includes a correspondence between an identifier of one of the first physical ports and an identifier of one of the second physical ports; and determining a candidate single board to be tested from the multiple candidate single boards according to the probability, adjusting the power of the target single board or the upstream of the target single board, and determining that logical connection needs to be added between one of the first physical ports on the target single board and one of the second physical ports of the candidate single boards to be tested when the power change trend of one of the first physical ports of the target single board is consistent with that of one of the second physical ports of the candidate single boards to be tested.

In a third aspect, an embodiment of the present invention provides a network controller, where the network controller includes:

a processing module, configured to determine a candidate board set of a target board, where the target board includes at least one first physical port, the candidate board set includes multiple candidate boards, and each of the multiple candidate boards includes at least one second physical port; the method further comprises determining a candidate single board to be tested from the plurality of candidate single boards; and is further configured to send a control instruction to adjust power of the target board or upstream of the target board

a checking module, configured to determine that a logical connection needs to be added between one of the first physical ports on the target board and one of the second physical ports of the candidate board to be tested when power variation trends of the one of the first physical ports of the target board and the one of the second physical ports of the candidate board to be tested are consistent.

In a possible design, the processing module is further configured to determine the candidate board to be debugged according to whether a probability that a logical connection exists between one of the second physical ports of each candidate board and one of the first physical ports of the target board is greater than or equal to a preset threshold.

In a possible design, the processing module is further configured to determine a candidate board set of the target board according to a connection constraint rule, where the connection constraint rule includes a type of the candidate board and/or a type of at least one second physical port included in the candidate board.

In a possible design, the processing module is further configured to determine a candidate board set of the target board according to missing data of network logical connection, where the candidate board includes at least one second physical port with missing logical connection; and is further configured to query the power of one of the first physical ports of the target board and the power of each of the second physical ports of each of the candidate boards, respectively.

In a possible design, the checking module is further configured to, when the powers of both the one of the first physical ports of the target board and the any one of the second physical ports of the any one of the candidate boards are greater than zero, delete the candidate board from the set of candidate boards if the power change trends of the one of the first physical ports of the target board and the any one of the second physical ports of the any one of the candidate boards are inconsistent.

In a possible design, the calculation module is further configured to calculate an absolute value of a difference between a power change of one of the first physical ports of the target board and a power change of one of the second physical ports of the at least two candidate boards to be measured, respectively.

In a possible design, the checking module is further configured to determine that the logical connection needs to be increased between one of the first physical ports on the target board and one of the second physical ports of the board to be measured with the smallest absolute difference value, and the processing module is further configured to rank, in order, probabilities that the one of the first physical ports has logical connection with one or more of the second physical ports of each candidate board

In one possible design, the processing module is further to: and respectively querying the power of one first physical port of the target single board and the power of each second physical port of each candidate single board.

In one possible design, the verification module is further configured to:

when the powers of one of the first physical ports of the target board and any of the second physical ports of any of the candidate boards are both greater than zero, if the power change trends of one of the first physical ports of the target board and any of the second physical ports of any of the candidate boards are not consistent, deleting the candidate board from the candidate board set.

In a possible design, the check module is further configured to determine whether the candidate board set only includes one candidate board.

In a fourth aspect, an embodiment of the present invention provides a computer device, where the computer device includes: memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method as described in any one of the possible designs of any one of the first and second aspects when executing the computer program.

In a fifth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, the computer program being configured to implement the method in any one of the possible designs of any one of the first and second aspects.

In a sixth aspect, embodiments of the invention provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform a method as set forth in any one of the possible designs of the first and second aspects.

In a seventh aspect, an embodiment of the present invention provides a system for determining a logical connection, where the system includes a network controller and at least one network device as described in any possible design of the third aspect, and the network device is configured to adjust power of the target board or upstream of the target board according to the control instruction sent by the network controller.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic diagram of a multi-site network connection provided by an embodiment of the present invention;

FIG. 2 is a schematic diagram of the internal structure of site A in FIG. 1;

fig. 3 is a schematic diagram of the internal structure of the station B in fig. 1;

FIG. 4A is a schematic diagram of a logical connection breakpoint at site A;

FIG. 4B is a schematic diagram of another logical connection breakpoint at site A;

FIG. 5 is a flow chart of a method for determining logical connections provided by an embodiment of the present invention;

FIG. 6 is a flow chart of a method for determining logical connections provided by another embodiment of the present invention;

FIG. 7A is a flowchart of a method for establishing a connection probability model according to an embodiment of the present invention;

FIG. 7B is a flowchart of another method for establishing a connection probability model according to an embodiment of the present invention;

FIG. 8 is a flow chart of a power verification method provided by an embodiment of the invention;

fig. 9 is a schematic diagram of a logic structure of a network controller according to an embodiment of the present invention.

Fig. 10 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present invention.

Detailed Description

While the present application has been described in some detail in the embodiments for the purpose of illustrating the invention and for the purpose of providing a thorough understanding of the present application, the present application is capable of being practiced otherwise than as specifically described in the embodiments and as such may be readily utilized by those skilled in the art without departing from the spirit and scope of the present application.

Embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

Fig. 1 is a schematic diagram of a multi-site network connection provided in an embodiment of the present invention. Different cities communicate with each other by deploying sites. Referring to fig. 1, a schematic diagram of a multi-site network connection provided by the present invention is shown. Site A, site B, site C, site D, site E, site F and site G are deployed between city 1 and city 2. The stations are connected through optical fibers to realize the transmission of optical signals. The stations can be connected through cables to realize the transmission of electric signals.

The station related by the invention is a station formed by combining a network element at the same place or a plurality of network elements which are connected through optical fibers and network cables and can communicate. For example, a plurality of network elements located in the same computer room are used as a station. A site may also be referred to as a single site. A network element usually comprises a plurality of boards, and thus a station can also be considered to be composed of one or more boards. The veneer to which the present invention relates includes but is not limited to: an Optical Wavelength conversion Unit (OTU), a Wavelength Selective Switching demultiplexer (WSD), a Wavelength Selective Switching multiplexer (WSM), an Optical Amplifier (OA), and so on.

Fig. 2 is a schematic diagram of the internal structure of the station a in fig. 1. The station A is internally provided with single boards such as an OTU, a WSM, a WSD and an OA, and is connected with the station B and the station E through an outbound optical amplifier and an inbound optical amplifier. According to the connection constraint rule of the site A, the opposite end of the output port of the OTU single board is the input port of the WSM single board, and the opposite end of the input port of the OTU single board is the output port of the WSD single board; the opposite end of the input port of the WSM single board is an output port of the OTU single board or the WSD single board, and the opposite end of the output port of the WSM single board is an input port of the OA single board; the opposite end of the input port of the WSD single board is an output port of the OA single board, and the opposite end of the output port of the WSD single board is an input port of the OTU single board or the WSM single board; the opposite end of the OA single-board input port is an output port of the WSM single-board, and the opposite end of the OA single-board output port is an input port of the WSD single-board. The connection constraint rule may also be referred to as a fiber connection relationship or a connection constraint, and refers to a connection relationship between two adjacent boards in a site, and may include a board type and a port type that may have a connection relationship with a target board in the site, or include one of the board type and the port type that may have a connection relationship in the site. For example, the connection constraint may be which types of boards generally have connection relationships with which types of boards, such as an OTU board generally has connection relationships with WSD and WSM boards. Or which type of board does not have a connection relationship with which type of board, for example, the output port of the OA board does not have a direct connection relationship with the WSM board. The connection constraint may also be which types of ports generally have a connection relationship with which types of ports, for example, a connection relationship between an input port and an output port. Or a certain type of port usually has no connection relation with which type of port, for example, there is no connection relation between an output port and an output port. The fiber connection constraint may be generally obtained through fiber connection data and single board data of the site memory amount network. It should be noted that the above is only an example of the connection relationship of the boards in the site a, it should be understood that the types of the boards in the site are not limited to the above four types, and the connection relationship between the boards is more complicated, and the present invention is described by taking this simple connection relationship as an example.

Fig. 3 is a schematic diagram of the internal structure of the station B in fig. 1. Fig. 3 is a schematic diagram of the internal structure of the station B in fig. 1. The site B is internally provided with three single boards of OA, WSM and WSD, and is connected with the site C through an outbound optical amplifier and an inbound optical amplifier, and the site B is not internally provided with an OTU single board. The connection relationship between the boards in the site B is relatively simple, for example, the opposite-end board of the input port of the OA board only has one type of WSM board, and the opposite-end board of the output port of the OA board only has one type of WSD board. It should be understood that the connection relationship between the single boards in different sites is different, and therefore, the connection fiber constraints of different sites are different.

Fig. 4A and 4B are schematic diagrams of network architectures with breakpoints occurring in logical connections, where fig. 4A illustrates a port where a breakpoint of a logical connection is located as an input port, and fig. 4B illustrates a breakpoint port of a logical connection as an output port. As shown in fig. 4A and 4B, the network architecture includes two planes, an upper layer is a control plane, and a lower layer is a data plane. The data plane includes physical entities such as network devices, single boards, optical fibers, cables, and the like, and is used for carrying service data. The control plane includes a control channel, a control signaling, and the like, and is used for controlling establishment of a service path, forwarding of service data, and the like in the data plane. The data plane may include a plurality of network devices, each of which in turn includes a plurality of boards, each board including a plurality of physical ports. Physical entities of a data plane, such as network equipment, a single board, a physical port, an optical fiber, a cable, and the like, can be mapped to a control plane one by one, and the physical entities mapped to the control plane can be called logical entities. In the present application, "connected" may be physically or logically connected, unless otherwise specified. The "port" in the present application may be a physical port or a logical port, as not particularly indicated. The control plane and the data plane are the connection relation of the single boards in the site A. The control plane includes a logical connection relationship between different boards, or a logical connection relationship between different ports on a board. The logical connection relationship may be a connection relationship stored in a memory such as a storage on the control plane, for example, the identifier of the bound two ports indicates that the logical connection relationship exists between the bound two ports. The data plane includes physical connection relationships between different boards, or physical connection relationships between different ports on a board. The physical connection relationship may be a connection relationship by optical fiber connection or a connection relationship by physical transmission media such as cable and radio electromagnetic wave. The single boards in the control plane are in one-to-one correspondence with the single boards in the data plane, and therefore, the physical connections and the logical connections are also in one-to-one correspondence. Fig. 4A and 4B show a process of an optical signal in a service under site a and a service on site a, as an example, an optical signal of a WSD board is split into three signals by the WSD board, where a first signal is output by an output port D1, enters the OTU board through an OTU1 board input port T11 for wavelength conversion, and then is subjected to a service under an output port T12; the second service is output from the input port T13 of the OTU board through the output port T14 after wavelength conversion, and then reaches the input port M1 of the WSM board, and is combined with the service from other input ports to form a third signal, which is subjected to power amplification by the OA board and then is continuously transmitted. As shown in the figure, the physical connection of the data plane is normal, the input port M1 of the WSM board in the control plane is a port of a logical connection breakpoint, that is, the logical connection between M1 and the opposite end is missing, and according to the connection fiber constraint in the site a, the opposite end of the input port of the WSM board may be an output port of the WSD board or an output port of the OTU board. As shown in fig. 4B, when the output port D1 of the WSD is a port of a logical connection breakpoint, according to the connection constraint of the site a, the opposite end of the WSD may be connected to an input port of an OTU board or an input port of a WSM board.

Fig. 5 is a flowchart of a method for determining a logical connection according to an embodiment of the present invention. The method for determining logical connection provided by the embodiment of the invention is applied to a site, and particularly applied to the connection between ports of at least two single boards. Referring to fig. 5, an optical network is taken as an example, in which "physical connection" refers to optical fiber connection, and "logical connection" refers to a logical entity after the optical fiber connection is mapped to a control plane. The embodiment of the invention can judge the logic connection aiming at the scenes of logic connection loss (logic connection breakpoints) or logic connection misconnection. The technical scheme of the embodiment of the invention can be applied to a station, and at least an optical fiber is connected between the outbound port and the inbound port of the station. The method can be executed by a network controller, such as a Software Defined Network (SDN), a network manager, and the like, and includes the following steps:

501: determining a candidate single board set of a target single board, wherein the target single board comprises at least one first physical port, the candidate single board set comprises at least one candidate single board, and each candidate single board in the at least one candidate single board comprises at least one second physical port.

For example, the target board is a board where a port with the logical connection breakpoint occurs, the candidate board set includes an opposite board that may be connected to the port with the logical connection breakpoint of the target board, and the candidate board set may include one or more boards. The first physical port may be any one port on the target board, or may be a port on the target board where a logical connection breakpoint occurs. The second physical port may be any one port on the candidate board, or may be a port in the candidate board, which may have a connection relationship with the port on the target board where the fiber is broken. The term "may" as used herein means that the probability may be estimated according to a certain rule, and the probability cannot be determined.

Any single board in the entire network may be used as the candidate single board set, for example, the candidate single board set includes all single boards in the entire network except the target single board, and the second physical port is any physical port in the candidate single board set.

Or screening out partial single boards as a candidate single board set according to a certain rule. For example, the type of the candidate board and/or the type of the at least one second physical port included in the candidate board may be determined according to a connection constraint rule. Taking fig. 4A as an example, an M1 input port of the WSM board is a port of a logical connection breakpoint, and according to the connection constraint rule in the site a, an opposite end of the logical connection breakpoint port (M1 port) of the WSM board may be an output port of an OTU-type board or an output port of a WSD-type board, so that all OTU boards and WSD boards that satisfy the connection constraint rule in the site a form a candidate board set a. Since the port M1 where the logical connection breakpoint occurs is an input port, what may be connected with M1 is an output port. Therefore, it may be determined that the output ports on the OTU board and the WSD board in the candidate board set a may have a connection relationship with M1.

In addition, if the board with the missing logical connection in the network may be a candidate board, determining a candidate board set of the target board according to the missing data of the logical connection of the network, where the candidate board includes at least one second physical port with the missing logical connection.

It should be noted that the boards involved in the embodiment of the present invention include, but are not limited to, the boards of the types described above, and the connection between the boards may also be more complex, based on which, the determination regarding the fiber-breaking point to the board port of the opposite end is only an example, and the set a may also include other types of boards, which is not specifically limited in this invention.

502: and respectively calculating the probability of the logical connection between one first physical port of the target single board and each second physical port of each candidate single board.

In this embodiment, the logical connection includes a correspondence between an identifier of the one of the first physical ports and an identifier of the one of the second physical ports.

The probability of logical connection between the target single board and the candidate single board can be calculated through a machine learning or artificial intelligence algorithm. For example, the probability that a logical connection exists between one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards is calculated through a machine learning model. The probability is a specific probability value, for example, any one of 20%, 50%, 90%, and the like.

For example, the characteristic parameters substituted into the machine learning model may include one or more of a board name, a board type, a board port, a slot, a subrack, a network element ID, and a station ID. The machine model is an algorithm model and is used for calculating the probability that a logical connection exists between a first physical port of a target single board and a second physical port of a candidate single board. Specifically, the probability that two board ports have logical connection can be obtained by inputting the target board characteristic parameter and the candidate board characteristic parameter into the algorithm model. The present invention is specifically illustrated and described in the following embodiments with respect to the generation of a machine learning model.

For example, after the probability that each second physical port of each candidate board has logical connection with the target board port is obtained, all the probabilities are arranged in order.

It should be noted that, although this step is executed before 503, in actual operation, it may also be executed after 503 power check. As shown in fig. 6, after power verification, the candidate single boards in the candidate single board set exclude obviously wrong candidate single boards, the remaining single boards form a new candidate single board set, when the new candidate single board set includes at least two single boards, each candidate single board characteristic parameter in the set and the target single board characteristic parameter are respectively substituted into the machine learning model, the probability that the logical connection exists in the combined connection is calculated, and the probability results are arranged in sequence. In this way, after removing obviously wrong candidate boards, only substituting the candidate board characteristic parameters and the target board characteristic parameters retained in the set into the machine learning model, and then performing power perturbation confirmation according to the probability, thereby achieving the same effect as the embodiment shown in fig. 5, avoiding unnecessary workload, and increasing the accuracy of confirming logical connection.

503: and carrying out power check on each candidate single board in the candidate single board set.

The step is an optional step, and aims to further screen the candidate single board set and eliminate physical single boards which are not connected with the target single board according to the power change condition in the network. For example, after each candidate board in the candidate board set is combined with the target board, power verification is performed, where the combining refers to separately querying port power of each candidate board and the fiber-broken board, and is not performed by physically connecting each candidate board with the target board. The power check flow can be seen in fig. 8. It should be understood that if the candidate board has a logical connection with the target board, both the second physical port of the candidate board and the first physical port of the target board should have power and have consistent power variation trend. Therefore, obvious wrong candidate single boards can be preliminarily eliminated by inquiring and comparing the power change trends of the candidate single board port and the fiber-broken single board port of each combined fiber connection. The power trend may be looked up and derived from the controller power data.

For example, after each candidate board in the candidate board set is combined with the target board, whether the power changes of the first physical port of the target board and the second physical port of the candidate board are consistent or not is respectively queried and compared.

In addition, the step aims to eliminate obviously wrong combined fiber connection, improve the accuracy of fiber connection repair, and select whether to execute the combined fiber connection according to specific conditions in actual operation.

504: and judging whether the candidate single board set only comprises one candidate single board.

For example, it may be determined whether power perturbation check is required according to the number of boards included in the candidate board set:

in some cases, if the candidate single board set only contains one candidate single board, it is determined that there is a logical connection between the candidate single board and the target single board, and the process ends.

In other cases, if the candidate board set includes at least two candidate boards, step 505 is executed.

It should be noted that this step is an optional step, and step 505 may be directly performed on the candidate board set without performing this step.

505: and determining the candidate single board to be tested from the plurality of candidate single boards according to the probability.

For example, the probability that a logical connection exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board is greater than or equal to a preset threshold, for example, the threshold is 90%.

Specifically, in this embodiment, a path is selected in a site where the target board is located, where the path at least includes the target board and the candidate board to be tested. The path is a path of optical transmission in a station, is a unidirectional path and comprises two or more single boards, and is used for determining a power regulation point.

As an example, when the first physical port of the target board is an input port, the path at least includes the target board and a candidate board, where the candidate board is a board with the highest probability of having a logical connection with the target board in the candidate board set, and the candidate board is a candidate board to be tested. As shown in fig. 4A, when M1 is a target board port, if the probability that the OTU1 board in the candidate board set has a logical connection with the target board is the highest, a path is selected, where the path includes at least the target board port M1 and the candidate board OTU1, and thus the selected path may be T13 → T14 → M1. When the target board port is an output port, if the probability that the OTU1 board in the candidate board set is logically connected to the target board is the highest, the OTU1 board is the candidate board to be tested. A path is selected, which includes at least the target board port D1 and the candidate board OTU1, so the selected path may be D1 → T11 → T12.

506: and adjusting the power of the target single board or the upstream thereof through the control signaling, and judging whether the power variation trend of the first physical port of the target single board is consistent with that of the second physical port of the single board to be measured.

In this embodiment, a plurality of attenuators and a plurality of OA boards are provided in a station and used as key debugging devices in a wavelength division system, where the attenuators may implement attenuation of internal channels of the boards, that is, different attenuations may be set for optical signals of each port, and the OA boards may adjust an overall optical path to implement power gain of the overall optical signals. In this embodiment, the adjusting of the power at the upstream of the port of the target board refers to performing adjustment based on the port of the target board. It should be understood that when the target board fiber-cut side (input side or output side) has multiple ports, the power of the path based on the target port is usually adjusted by the attenuator, and when the target board fiber-cut side has a unique port, the power of the path based on the target port can be adjusted by the attenuator or the OA board. This embodiment will be described by taking an example of adjusting the power upstream of the target board port through the attenuator.

Specifically, if there is a logical connection between the candidate board and the target board, the power of the upstream of the target board is adjusted, and the optical power variation trends of the port of the target board and the port of the candidate board should be kept consistent. Based on this, the power of the upstream of the target single board is adjusted, and whether the candidate single board is logically connected with the target single board can be determined according to whether the optical power variation trends of the port of the target single board and the port of the candidate single board are consistent.

As shown in fig. 4A, when the fiber-broken single board port is M1, the selected path for determining the adjustment point may be T13 → T14 → M1, the path includes the target single board port M1 and the candidate single board OTU1, and the adjustment point is determined to be the T14 port of the OTU single board. When the power attenuation is adjusted to reduce the optical power of the T14 port to a small extent, it is queried whether the input power of the M1 port is reduced correspondingly, and according to the fiber connection shown in fig. 4A, a logical connection exists between the OTU1 and the target single board port M1, so that when the attenuation of the T14 port is adjusted, the output ports T14 and M1 of the OTU1 single board will have consistent power changes. In addition, when the target board port is M1, the selected path for determining the adjustment point may also be another path, and certainly, since there is no logical connection between the candidate board in the other path and the target board, when the attenuation of the candidate board port is adjusted, the M1 port does not change correspondingly.

As shown in fig. 4B, when the target board port is D1, the selected path for determining the adjustment point is D1 → T11 → T12, the path includes the target board port D1 and the candidate board OTU1, the adjustment point is determined to be the D1 port of the WSD board, and when the attenuation is adjusted so that the optical power of the D1 port is reduced, whether the power of the input port or the output port of the candidate board OTU1 is correspondingly reduced is queried. According to the fiber connection shown in fig. 4B, there is a logical connection between the port of the OTU1 board and the D1 port of the target board, so when the attenuation of the D1 port is adjusted, the power of both the input port and the output port of the OTU1 board will change correspondingly.

Preferably, in order to ensure the stability of the transmission service, only small amplitude adjustment is performed on the attenuation of the signal, so that the output optical power is changed slightly. The present invention is not particularly limited to the specific value of the attenuation adjustment.

507: determining one of the second physical ports of the candidate single board and one of the first physical ports of the target single board does not require adding a logical connection.

In this embodiment, if the power variation trends of the second physical port of the candidate board and the first physical port of the target board are different, there is no logical connection between the two. For example, the steps 505 and 506 are repeated to check other candidate boards until a candidate board whose port power is consistent with the first physical port power variation trend of the target board is detected. For example, the checking order is performed according to the high-low order of the probability of the logical connection existing in the candidate single board set.

The different power variation trends of the candidate single-board port and the target single-board port mean that the attenuation of the signal is adjusted to enable the optical power of the target single-board port to be reduced by a small margin (for example, 2dB), but the optical power of the candidate single-board port is basically unchanged or even increased after being inquired. As shown in fig. 4A, when the target board port is W1, if it is calculated according to a machine learning model that it is an OTU2 board with the highest probability of having a logical connection, a path OTU2 → M1 is selected according to step 505, and it is determined that the adjustment point is the line-side output port of the OTU2 board, and when the optical power attenuation of the line-side output port is adjusted, although the output power of the candidate board OTU2 changes the same, the target board port W1 does not have a consistent power change, so it is determined that no logical connection needs to be added between the line-side output port of the OTU2 board and the W1 port. At this time, the flow will re-execute step 505 and 506, select the candidate board with the highest probability of having the logical connection in the candidate board set to perform the power perturbation check again, for example, T13 → T14 → M1, determine the power adjustment point T14, and perform the power perturbation check.

508: determining that a logical connection needs to be added to one of the second physical ports of the candidate single board and one of the first physical ports of the target single board.

In this embodiment, if the power variation trend of one of the second physical ports of the candidate board is consistent with that of one of the first physical ports of the target board, a logical connection exists between the one of the second physical ports of the candidate board and the one of the first physical ports of the target board. Therefore, the accuracy and the uniqueness of the repair result can be ensured by carrying out the power perturbation check on the candidate single boards in the candidate single board set.

In this embodiment, that the power of the second physical port of the candidate board is consistent with the power variation trend of the first physical port of the target board means that when the power of the target board or the upstream power thereof is adjusted to reduce the power of the output optical signal, if it is found that the power of the first physical port of the target board and the power of the second physical port of the candidate board are both reduced, the power variation trends of the second physical port of the candidate board and the first physical port of the target board are consistent. In this embodiment, specific values of the power changes of the second physical port of the candidate board and the first physical port of the target board are not specifically limited, and the two values are not required to be equal.

For example, when the trend of the power change of one of the first physical ports of the target board is consistent with the trend of the power change of one of the second physical ports of at least two candidate boards to be measured, the absolute value of the difference between the power change of one of the first physical ports of the target board and the power change of one of the second physical ports of at least two candidate boards to be measured is calculated respectively; determining that a logical connection needs to be added between one of the first physical ports on the target board and one of the second physical ports of the board to be measured with the smallest absolute difference value.

In this embodiment, a path including at least one first physical port of the target board and one second physical port of the candidate board in one site is taken as an example for explanation, and in fact, there may be multiple paths, and one path may be selected from the multiple paths.

According to the method provided by the embodiment, the probability that each candidate single board is logically connected with the target single board is calculated by using the machine learning model, and the only logical connection is determined by combining the trend comparison of the port powers of the candidate single board and the target single board and the power perturbation, so that the fiber breakage repair can be rapidly carried out, the previous cycle is reduced to the hour, the efficiency is improved by nearly one hundred times, the fiber connection inspection and repair time is greatly saved, the labor cost is greatly reduced, and the accuracy of the fiber connection repair is improved.

The model recommendation and the power perturbation are combined, so that the uncertainty of the model recommendation can be reduced, meanwhile, the power perturbation verification is carried out according to the probability, the influence of perturbation operation on services can be greatly reduced, and the whole scheme has high feasibility and reliability.

Fig. 6 is a flowchart of a method for determining a logical connection according to another embodiment of the present invention. In the embodiment shown in fig. 6, step 502 is performed after step 503, other steps are the same as those in the embodiment shown in fig. 5, and the specific method and the embodiment may refer to the description of fig. 5, which is not repeated herein.

Fig. 7A and 7B show a process of establishing the machine learning model, which is obtained by using a program algorithm. It should be noted that the probabilistic model can be obtained through a plurality of algorithms, and this embodiment is described by taking a machine learning method as an example, and specifically includes the following steps:

the method comprises the following steps: all connection relations in the network-wide fiber connection data are converged, and the physical connections and the logical connections in the network correspond to each other one by one, so that the connection relations can be physical connections or logical connections. Each connection relation comprises two single boards and the connection between the two single boards. One of the boards is regarded as a source board, the other board is regarded as a sink board, and the source board or the sink board port is assumed to be a port of the logical connection breakpoint, and the board where the logical connection breakpoint port is located constitutes a target board set. The whole network connection fiber data converges the connection relation of the whole network single boards, and each connection fiber comprises a source single board and a destination single board, so the whole network connection fiber data is actually a pair of single boards with direct connection relation in the whole network.

For example, an OA in the aggregated full-network fiber connection data is used as all the connection fibers of the source board, assuming that an OA output port is a fiber-broken port, and the target board OA forms a target board set, as shown in table 1 below:

TABLE 1

Name of single board

Of the single-plate type

Single board port

Source slot position

Source frame

Source network element ID

Source site ID

OAU103

OUT

OAU103

TDC

OAU103

OUT

OAU101

TDC

OAU101

OUT

OAU101

TDC

OAU101

OUT

OAU101

TDC

Step two: according to the connection constraint rule, an opposite-end veneer set which is possibly connected with the target veneer port is determined from the whole network veneer data to form a candidate veneer set of the target veneer port, it should be noted that the candidate veneer set is different from the candidate veneer set in the foregoing embodiment in concept, the candidate veneer set in this embodiment is used for training a machine learning model, the target veneer is not really broken, but paired veneers having a direct connection relationship in the whole network are respectively assumed as the target veneer. The data of the whole network single boards are a set of all single boards in the network, and the connection constraint, which may also be referred to as a connection relationship, refers to a connection relationship between different single boards in a site, specifically, a type of single board is usually connected with which type of single board, or usually does not have a connection relationship with which type of single board, and for types of single boards that may be connected, ports that are usually connected and ports that may not be connected.

For example, all boards except the target board in the whole network are selected as the candidate board set of the target board.

For example, taking the first board OAU103 in table 1 as an example of a target board, determining a candidate board subset of the target board OAU103 from the full-network board data according to the fiber connection constraint is shown in table 2:

TABLE 2

Name of single board

Of the single-plate type

Single board port

Groove position

Sub-frame

Network element ID

Site ID

OAU103

OUT

OAU103

TDC

FIU

OADM

OAU101

TDC

ITL

OADM

OUT

FIU

OADM

TDC

OAU101

OUT

OAU101

TDC

Assume that table 2 is the data of the full-network boards in the site, wherein four boards of site a1 constitute the candidate board subset of target board OAU 103. Other single boards are excluded from the candidate single board set due to the mismatch of single board types or the mismatch of ports.

Step three: and combining the target single board port in the target single board set and each single board port in the candidate single board subset to form a fiber connection data set. It should be understood that the extended fiber connection data set includes a plurality of fiber connection data subsets, each subset includes a target board and its possible connected opposite-end boards, and each subset includes the correct opposite-end board of the target board and other possible extended opposite-end boards.

For example, assuming that the target board is a1 and the correct opposite board thereof is B1, according to the fiber connection constraint, the target board a may have, in addition to B1, similar boards B2 and B3 and other types of boards C1, C2, D1 and D2, the correct board and other possible opposite boards B2, B3, C1, C2, D1 and D2 jointly form a fiber connection data subset a1, similarly, all possible opposite boards of the target board a2 form a fiber connection data subset a2, all possible opposite boards of the target board A3 form a fiber connection data subset A3, and so on, and all fiber connection data subsets in the whole network jointly form a fiber connection data set.

Specifically, taking the OAU103 target board and its candidate board subset as an example, the formed fiber connection data subset is shown in table 3:

TABLE 3

The fiber connection data subset is a data set after the target single board parameter is combined with the candidate single board parameter respectively to connect the fibers. The last column "Label" in table 3 is that after comparing the combined continuous fiber with the original correct continuous fiber sample, the existing labeled value is 1, and the nonexistent labeled value is 0, so as to facilitate the feature extraction later.

It should be understood that the fiber connection data set includes many fiber connection data subsets, and each fiber connection data subset includes a target single board and a combined fiber connection of all candidate single boards in the candidate single board subset.

Step IV: randomly extracting a certain number of samples (fiber connection data subsets) from the fiber connection data set as a training set, and using the rest samples (fiber connection data subsets) as a testing set, wherein the training set is used for machine training, and the testing set is used for program verification to obtain the actual accuracy of the algorithm. The training set comprises fiber-connected data sets

As an example, 80% of samples of the fiber-connected data set may be randomly extracted as a training set, and the remaining 20% of samples may be used as a testing set, and it should be noted that, in this embodiment, there is no specific limitation on the extraction ratio of the samples, and the extraction ratio is determined according to specific situations in actual operation.

Referring to fig. 7B, fig. 7B shows a process of training and verifying the machine learning model.

Step five: and extracting the characteristics of the opposite-end single board port with the correct optical fiber connection and the parameters of the opposite-end single board port with the wrong optical fiber connection by processing the characteristic parameters of the correct optical fiber connection and the characteristic parameters of the wrong optical fiber connection for model screening, eliminating and determining the correct opposite-end single board. In this embodiment, the specific board ports, including the target board port and the candidate board port, are identified by the board name, the board type, the board port, the slot position, the sub-frame, the network element ID, the site ID, and other features. The characteristic parameters of the fiber connection include, but are not limited to, a source board name, a source board type, a source port, a source slot, a source sub-frame, a source network element type, a sink board name, a sink board type, a sink port, a sink slot, a sink sub-frame, a sink network element type, and the like. And performing feature extraction by adopting a statistical analysis method or other algorithms based on the feature parameters to obtain an initial machine learning model.

The probability model is used for calculating the correct probability of the fiber connection, specifically, characteristic parameters (a source single board name, a source single board type, a source port, a source slot position, a source sub-frame, a source network element type, a sink single board name, a sink single board type, a sink port, a sink slot position, a sink sub-frame, a sink network element type and the like) of the fiber connection are input into the model, and the correct probability of the fiber connection can be obtained through model internal calculation.

Step sixthly, the accuracy of the initial machine learning model is verified. Specifically, the characteristic parameters of each fiber in the test set are input into the determined initial machine learning model, the correct probability of each fiber is output as a target variable, and the target variable is compared with an actual sample, wherein the actual sample is a Label in the fiber connection data set, so as to obtain the accuracy of the algorithm. For example, if the machine learning model calculates that the correct probability of a certain fiber connection is 99%, and the Label in the fiber connection dataset is 1, 99% is the target variable, 1 is the actual sample, and the target variable is matched with the actual sample. And when all the connecting fibers in the test set are verified, the accuracy of the initial machine learning model can be obtained. And when the accuracy of the initial model is lower than the expectation, re-performing the fifth step, and adjusting or adding the feature extraction process until the accuracy of the adjusted machine learning model reaches the expectation.

Step (c): and outputting the machine learning model.

Referring to fig. 8, a power verification method provided in the embodiment of the present invention is applied to the fiber connection repair method shown in fig. 5 and 6, specifically, the method includes the deployment process of steps 503 and 602, and includes:

801: combining and connecting each candidate single board in the candidate single board set with the target single board, and inquiring the power of a first physical port of the target single board and the power of a second physical port of the candidate single board.

In this embodiment, the power data for each combined source and sink port may be queried and derived from the controller's full network performance data.

For example, each candidate board in the candidate board set and the target board are combined and connected to the fiber, and the power of the source port and the sink port of the combined fiber is queried. The combined fiber comprises a source single board and a destination single board, and specifically, when a first physical port of a target single board is an input port, a candidate single board is the source single board of the combined fiber, and a source port is the input port of the candidate single board; the target single board is a combined fiber-connected host single board, and the host port is an output port of the target single board; when the fiber-broken port is an output port, the target board is a fiber-connected source board, the source port is an input port of the target board, the candidate board is a fiber-connected sink board, and the sink port is an output port of the candidate board.

For example, each candidate board in the candidate board set and the target board are combined and connected to the fiber, and the power of any physical port of the target board and the power of any physical port of the candidate board are queried.

802: and judging whether the first physical port and the second physical port both have power.

In this embodiment, when both the first physical port and the second physical port have power, whether the combined fiber connection is correct can be confirmed by comparing whether the power variation trends of the first physical port and the second physical port are consistent. However, when at least one of the two ports has no power data, it is impossible to confirm the correctness of the combined fiber connection by a power check method, and in this embodiment, no processing is performed on the candidate veneer corresponding to the combined fiber connection, and the candidate veneer is retained in the candidate veneer set, and then other combined fiber connections are checked again. And after the verification of all combined connecting fibers is finished, the correctness of the connecting fibers can be determined again through the power perturbation.

803: and when the first physical port of the target single board and the second physical port of the candidate single board both have power, judging whether the power changes of the two ports are consistent.

It should be understood that if the combined fiber connection is correct, the power changes of the first physical port of the target board and the second physical port of the candidate board of the fiber connection should be consistent. Based on this, it is able to determine whether the power variation of the first physical port of the combined fiber-connected target board and the second physical port of the candidate board in the full network performance data is consistent, and further determine whether the corresponding candidate board is the opposite board to which the target board is correct.

804: and deleting the candidate single board corresponding to the combined connecting fiber in the candidate single board set.

And if the power change trends of the first physical port of the target single board and the second physical port of the candidate single board are inconsistent, deleting the candidate single board corresponding to the combined connecting fiber from the candidate single board set.

805: and reserving the candidate single board corresponding to the combined connecting fiber in the candidate single board set.

It should be noted that this embodiment may be a component of the method embodiments shown in fig. 5 and fig. 6, and as an implementation step before the power perturbation, the combined fiber connections with obvious errors may be preliminarily eliminated through power verification.

Fig. 9 is a schematic diagram of a logic structure of a network controller according to an embodiment of the present invention. Referring to fig. 9, the network controller may be applied to a station having at least an optical fiber (physical connection) connected between an outbound optical drop and an inbound optical drop, and includes a processing module 901, a checking module 902, and a calculating module 903, and is configured to execute the method flow shown in fig. 5 or fig. 6.

Fig. 10 is a computer device provided in an embodiment of the present application, where the computer device includes a processor 1010, a memory 1020, and a port 1030, as shown in fig. 10:

processor 1010 includes one or more processing cores. The processor 1010 executes various functional applications and data processing by executing software programs and modules. The Processor 1010 includes an arithmetic logic Unit, a register Unit, a control Unit, and the like, and may be a separate central processing Unit, or may be an Embedded Processor such as a Microprocessor (MPU), a Microcontroller (MCU), or a Digital Signal Processor (EDSP).

Specifically, in the present application, the processor 1010 is configured to execute instructions and/or application programs stored in the memory 1020, for example, may be used to determine a candidate board set of a target board; the candidate single board is also used for calculating the probability of the existence of the logical connection between the candidate single board and the target single board; and determining the candidate single board to be tested according to the probability.

For example, the processor 1010 may be further configured to determine that a probability that a logical connection exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board is greater than or equal to a preset threshold; for example, the processor is configured to determine that a highest probability exists between one of the second physical ports of the candidate board to be debugged and one of the first physical ports of the target board.

For example, the processor 1010 is configured to determine the candidate board set of the target board according to a connection constraint rule, where the connection constraint rule includes a type of a candidate board of the target board and/or a type of at least one second physical port included in the candidate board.

For example, the processor 1010 is further configured to query the power of one of the first physical ports of the target board and the power of each of the second physical ports of each of the candidate boards, and when the power of one of the first physical ports of the target board and the power of each of the second physical ports of each of the candidate boards are both greater than zero, if the power change trends of one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards are consistent, the candidate boards are retained in the candidate board set.

The Memory 1020 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM) or a ROM, a magnetic Memory, a flash Memory, a magnetic disk or an optical disk. Memory 920 may be used to store instructions executable by software programs and modules, and the like.

For example, memory 1020 includes application program modules required for at least one function. The application program modules are used to store instructional information and/or computer programs that can be executed on the processor 1010 to implement the aforementioned methods. For example, the memory 1020 may include a preset machine learning model, and the probability that the logical connection exists between one of the first physical ports of the target board and each of the second physical ports of each of the candidate boards is calculated through the machine learning model, where the machine learning model is obtained based on the feature parameter training of the full-network logical connection.

The interface 1030 is used for the computer device 100 to issue a control instruction to the device layer network device, so as to adjust the power of one of the second physical ports of the board to be tested.

The network device 100 may further include a storable operating system (not shown), which may be an operating system such as a Real Time eXecutive (RTX), LINUX, UNIX, WINDOWS, or OS X, and may need to process basic transactions such as managing and configuring memory, determining priorities of system resources, controlling input devices and output devices, operating a network and managing a file system, and providing an operating interface for a user to interact with the system.

For example, network device 100 may also include input/output components (not shown). The input/output components include a display for displaying information and an input device such as a mouse, keyboard, etc. for a user to input information. Wherein the display and input devices are coupled to the processor 1010 via buses 1-40.

According to the device provided by the embodiment, the probability of logical connection between each candidate single board and the target single board is calculated by using the machine learning model, and the only logical connection is determined by combining the trend comparison of the port powers of the candidate single board and the target single board and the power perturbation, so that the fiber breakage repair can be rapidly carried out, the previous cycle is reduced to the hour, the efficiency is improved by nearly one hundred times, the fiber connection inspection and compensation time is greatly saved, the labor cost is greatly reduced, and the accuracy of the fiber connection repair is improved.

It will be understood by those skilled in the art that all or part of the steps of the above embodiments may be implemented by hardware, or may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, such as a read only memory, a magnetic or optical disk, and the like.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

27页详细技术资料下载

上一篇：一种医用注射器针头装配设备

下一篇：一种基于光纤色散的光波束形成预加重装置

Method for determining logical connection and related equipment

相关技术

网友询问留言