Detailed Description
The following detailed description is made with reference to the accompanying drawings and is provided to assist in a comprehensive understanding of various example embodiments of the disclosure. The following description includes various details to aid in understanding, but these are to be considered merely examples and are not intended to limit the disclosure, which is defined by the appended claims and their equivalents. The words and phrases used in the following description are only intended to provide a clear and consistent understanding of the present disclosure. In addition, descriptions of well-known structures, functions and configurations may be omitted for clarity and conciseness. Those of ordinary skill in the art will recognize that various changes and modifications of the examples described herein can be made without departing from the spirit and scope of the present disclosure.
Fig. 1 shows a schematic diagram of a customer abnormal access network, specifically illustrating a topology of a passive optical network of a fiber to the home (Fiber To The Home, FTTH), according to an embodiment of the present disclosure.
In the passive Optical network shown in fig. 1, an Optical line terminal (Optical LINE TERMINAL, OLT) may be connected to a plurality of primary splitters (Optical Branching Device, OBD), and a plurality of secondary splitters may be connected to each primary splitter, and a port may be provided to one or more users under each secondary splitter. The optical network shown in fig. 1 includes two stages of optical splitters, but the invention is not limited thereto and the optical network may include more stages of optical splitters, wherein the final stage of optical splitters may provide ports for one or more subscribers.
As shown in fig. 1, there is a port user with an abnormal resource, which occupies (e.g., connects to) a port of the secondary splitter a in the field, but which is erroneously shown to occupy a port of the secondary splitter b in the resource management system of the optical network. For such abnormal users, it is necessary to identify that the occupation of the last-stage splitter port in the resource management system is inconsistent with the site.
Therefore, there is a need to focus on the accuracy of the existing data of the resource management system. In an embodiment according to the present disclosure, the correlation between the location of the last-stage optical splitter and the user distribution is analyzed taking into consideration the data collected by the PON network manager (such as downlink light attenuation and access distance, which will be described in detail later) and the resource management system data in combination.
The data related to the correspondence between the beam splitters and the users collected in the prior art is derived from a resource management system. For example, the splitter hang distance for each user is the distance the user displays in the resource management system to the corresponding OBD. Such data derived from the resource management system cannot be used for feature analysis of port users due to inaccuracy of the data in the resource management system.
The inventors of the present application have realized that the downlink optical attenuation corresponding to the port users and the access distance of the optical network units (Optical Network Unit, ONUs) to the OLT can be used to reflect the correlation between the location of the last-stage optical splitter and the user profile. The downlink light degradation may refer to, for example, light degradation from an ONU to an OLT, and the ONU may refer to, for example, a cat of a user. The data of downlink light attenuation and access distance can be collected by PON network management. By selecting the downlink light attenuation corresponding to each port user and the access distance from the ONU to the OLT as the characteristic data of the port user, the port occupation condition of the final-stage optical splitter can be obtained through a big data analysis method.
Fig. 2 shows a feature data diagram of port users under the same penultimate optical splitter, according to an embodiment of the present disclosure. The penultimate beam splitter is the last stage of the final beam splitter. For example, in fig. 1, the second stage beam splitter is the last stage beam splitter, and the first stage beam splitter is the penultimate beam splitter.
In the two-dimensional coordinate system shown in fig. 2, each data point may correspond to one port user, the abscissa of each data point is downlink light attenuation corresponding to the port user, and the ordinate of each data point is an access distance from the ONU corresponding to the port user to the OLT. The data shown in fig. 2 is normalized and is not an actual value. Under the one-stage beam splitter as shown in fig. 2 there are 5 final-stage beam splitters, and accordingly the data points of the user in fig. 2 appear approximately as 5 clusters, each cluster corresponding to a center point (as shown by the 5 five-pointed stars in fig. 2). Thus, as will be described below in connection with FIG. 3, the center point of each cluster may be assigned to one category, and the data points may be further assigned to 5 categories.
Fig. 3 illustrates a flowchart of a method for detecting an abnormal port user in a resource management system of an optical network, according to an embodiment of the present disclosure.
At step S301, downlink light attenuation and access distances corresponding to a plurality of port users may be acquired from a network manager, where the downlink light attenuation and access distance of each port user constitute characteristic data of the port user.
At step S302, the feature data of the plurality of port users may be divided into k categories based on a k-means cluster analysis method and a nearest neighbor classification method, where k is the number of final stage splitters, and the divided categories of the feature data of each port user indicate the final stage splitter to which the port user corresponds. Referring to FIG. 2, for the data points in FIG. 2, a k-means cluster analysis may be performed to divide the data points into 5 categories, each identified in FIG. 2 by a different shape of the point.
In an embodiment according to the present disclosure, classifying the feature data of the plurality of port users into k categories based on a k-means cluster analysis method and a nearest neighbor classification method may include taking downlink light attenuation and an access distance in the feature data of each port user as two-dimensional coordinates representing data points of the port user, performing k-means cluster analysis on a set of data points of the plurality of port users to obtain k center points, the k center points corresponding to the k categories, and classifying the feature data of each port user into a category corresponding to a center point closest to the data point corresponding to the feature data among the k center points by nearest neighbor classification.
A method of performing k-means cluster analysis on a set of data points for a plurality of port users to obtain k center points is shown in fig. 4. A flowchart of an iteration of performing a k-means cluster analysis is shown in fig. 4.
At step S401, k points may be selected in the two-dimensional coordinate system as initial center points.
At step S402, for each data point, a distance of the data point to each center point may be calculated and assigned to the center point closest to the distance of the data point.
At step S403, for each of the k center points, the two-dimensional coordinates of all data points assigned to that center point may be averaged to obtain updated two-dimensional coordinates of that center point.
At step S404, it may be determined whether the two-dimensional coordinates of the center point to which each data point is assigned change. If so, the process may return to step S402, and the steps S402 and S403 may be cyclically performed. If not, the loop is jumped out. At this time, the data points of the multiple port users are already classified into k categories, where the two-dimensional coordinates of the k center points are the two-dimensional coordinates of the k center points in the iteration, where the k center points in the iteration may form a set { Ci1,Ci2,…,Cij,…,Cik }, where j indicates the sequence number of the center point and i indicates the sequence number of the iteration.
In embodiments according to the present disclosure, the clustering result may be unstable due to the random initialization of the center point strategy in the k-means cluster analysis. FIG. 5 shows a schematic diagram of the results of ten iterations of the k-means cluster analysis on the same profile as shown in FIG. 2, where each of the 5 categories resulting from each cluster analysis are outlined in FIG. 5 with a dashed circle for identification, in accordance with an embodiment of the present disclosure. From fig. 5 it can be seen that the effect of the initial center point on the k-means cluster analysis, the initial center point of each iteration of the k-means cluster analysis is random and different, and thus the result of the cluster analysis is accordingly unstable. Therefore, a plurality of iterations of k-means cluster analysis may be performed, i.e., n iterations shown in fig. 4 may be performed, and the coordinates of k center points in n iterations may be averaged to obtain the coordinates of k final center points { C1,C2,…,Cj,…,Ck }. For example, for fig. 5, the coordinates obtained in 10 iterations for 5 center points may be averaged separately to obtain the final 5 center points. The item_count marked in the small graph in fig. 5 is the iteration number for which the k-means cluster reaches convergence.
After obtaining k center points, nearest neighbor classification can be performed, namely, the characteristic data of each port user is divided into categories corresponding to the center point, which is closest to the data point corresponding to the characteristic data, in the k center points, so that the port users are divided into k categories.
Returning to fig. 3, after the port users are divided into k categories, at step S303, it may be determined whether the last-stage beam splitter indicated by the divided category of the feature data of each port user is identical to the corresponding last-stage beam splitter of the port user in the resource management system, and if not, it is determined that the port user is an abnormal port user. The port occupancy of the users of most of the last-stage splitters under the penultimate splitter may be uniform in the resource management system and in the field, so that the last-stage splitter indicated by the class corresponding to the center point may be set to the last-stage splitter corresponding to the majority of the port users under the class in the resource management system (or in the field). Therefore, the corresponding relation between the class of the cluster analysis and the final-stage beam splitter can be obtained by using the result of the cluster analysis, so that abnormal port users, of which the result of the cluster analysis is inconsistent with the records in the resource management system, can be found. For example, for the optical network shown in fig. 1, it may be found that a certain port user is connected to the secondary splitter a in the field according to the result of the cluster analysis, and is connected to the secondary splitter b in the resource management system, and the port user may be determined as an abnormal port user.
The feature data may be preprocessed before classifying the feature data of the plurality of port users into k categories.
In an embodiment according to the present disclosure, the encoding of the penultimate beam splitter, the encoding of the final beam splitter, and the user access number may be obtained from a resource management system, and the feature data may be preprocessed according to these data. For the optical network shown in fig. 1, the characteristic data can be divided into a plurality of data sets according to the codes of the primary beam splitter, and the data sets are exported and stored in batches, so that the data sets are convenient for subsequent cluster analysis. When cluster analysis is performed, cluster analysis can be performed on data of the secondary beam splitters belonging to the same primary beam splitter.
In the embodiment of the disclosure, the number of the second-stage optical splitters under the same first-stage optical splitter and the number of the users under the same second-stage optical splitter can be counted, and the characteristic data of the port users corresponding to the last-stage optical splitter of only one second-stage optical splitter under the first-stage optical splitter can be screened out, so that analysis is simplified.
In an embodiment according to the present disclosure, abnormal feature data of a port user in which the feature data does not conform to a predetermined rule may be screened out. For example, if the access distance and the code of the primary beam splitter are missing, the characteristic data of the port user is rejected. For example, if the access distance is less than or equal to 0, the feature data of the port user is rejected.
After detecting the abnormal port user, the subsequent work arrangement can be performed according to the detected abnormal port user. For example, the last-stage beam splitter corresponding to the abnormal port user in the resource management system may be updated to match the analyzed last-stage beam splitter.
In an embodiment according to the present disclosure, a service package area and a belonging area corresponding to a port user may be obtained from a resource management system. Thus, the abnormal port detection method of the present disclosure may be utilized to obtain the resource accuracy of each region for further evaluation and processing, such as preferentially repairing regions with low accuracy.
Fig. 6 illustrates an exemplary configuration of a computing device 600 capable of implementing embodiments in accordance with the present disclosure.
Computing device 600 is an example of a hardware device that can employ the above aspects of the present disclosure. Computing device 600 may be any machine configured to perform processing and/or calculations. Computing device 600 may be, but is not limited to, a workstation, a server, a desktop computer, a laptop computer, a tablet computer, a Personal Data Assistant (PDA), a smart phone, an in-vehicle computer, or a combination thereof.
As shown in fig. 6, computing device 600 may include one or more elements that may be connected to or in communication with bus 602 via one or more interfaces. Bus 602 may include, but is not limited to, an industry standard architecture (Industry Standard Architecture, ISA) bus, a micro channel architecture (Micro Channel Architecture, MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus. Computing device 600 may include, for example, one or more processors 604. The one or more processors 604 may be any kind of processor and may include, but is not limited to, one or more general purpose processors or special purpose processors (such as special purpose processing chips). The processor 602 may be, for example, processing circuitry in an AP configured to implement the flow steps as shown in fig. 3.
Computing device 600 may also include or be connected to a non-transitory storage device 614, which non-transitory storage device 614 may be any storage device that is non-transitory and that may enable data storage, and may include, but is not limited to, disk drives, optical storage devices, solid state memory, floppy disks, flexible disks, hard disks, magnetic tape, or any other magnetic medium, compact disk or any other optical medium, cache memory and/or any other memory chip or module, and/or any other medium from which a computer may read data, instructions, and/or code. Computing device 600 may also include Random Access Memory (RAM) 610 and Read Only Memory (ROM) 612. The ROM 612 may store programs, utilities or processes to be executed in a nonvolatile manner. The RAM 610 may provide volatile data storage and stores instructions related to the operation of the computing device 600.
In summary, according to a first aspect of the present disclosure, there is provided a method for detecting an abnormal port user in a resource management system of an optical network, including acquiring downlink light attenuation and access distance corresponding to a plurality of port users from a network manager, wherein the downlink light attenuation and access distance of each port user constitute characteristic data of the port user, classifying the characteristic data of the plurality of port users into k classes based on a k-means cluster analysis method and a nearest neighbor classification method, wherein k is a number of final-stage splitters, the classified class of the characteristic data of each port user indicates a final-stage splitter corresponding to the port user, and determining whether the final-stage splitter indicated by the classified class of the characteristic data of each port user is identical to the corresponding final-stage splitter of the port user in the resource management system, and if not, determining that the port user is an abnormal port user.
In some embodiments, classifying the feature data of the plurality of port users into k categories based on a k-means cluster analysis method and a nearest neighbor classification method includes taking downlink light attenuation and access distance in the feature data of each port user as two-dimensional coordinates representing data points of the port user, performing k-means cluster analysis on a set of data points of the plurality of port users to obtain k center points, wherein the k center points correspond to the k categories, and classifying the feature data of each port user into categories corresponding to center points closest to the data points corresponding to the feature data in the k center points through nearest neighbor classification.
In some embodiments, performing k-means cluster analysis on the set of data points of the plurality of port users to obtain k center points includes performing an iteration including S1, selecting k points in a two-dimensional coordinate system as initial center points, S2, for each data point, calculating a distance from the data point to each center point, and assigning the data point to a center point closest to the data point, S3, for each of the k center points, averaging two-dimensional coordinates of all data points assigned to the center point to obtain updated two-dimensional coordinates of the center point, S4, repeating S2 to S3 until the two-dimensional coordinates of the assigned center point of each data point no longer changes, at which time the two-dimensional coordinates of the k center points are the two-dimensional coordinates of the k center points in the iteration, the k center points of the iteration forming a set { Ci1,Ci2,…,Cij,…,Cik }, wherein j indicates a sequence number of the center point, and i indicates a sequence number of the iteration.
In some embodiments, performing k-means cluster analysis on the set of data points of the plurality of port users to obtain k center points further comprises performing n iterations, and averaging the coordinates of the k center points in the n iterations, respectively, to obtain the coordinates of k final center points { C1,C2,…,Cj,…,Ck }.
In some embodiments, the feature data of the plurality of port users is preprocessed before being divided into k categories.
In some embodiments, the final stage beam splitter is assigned to the same penultimate beam splitter.
In some embodiments, the preprocessing includes screening out characteristic data of port users corresponding to a last stage splitter, wherein only one last stage splitter is below a next-to-last stage splitter.
In some embodiments, the preprocessing includes screening out feature data of port users in which the feature data does not meet a predetermined rule.
According to a second aspect of the present disclosure there is provided an apparatus for detecting an abnormal port user in a resource management system of an optical network, comprising a memory and a processor, the memory being communicatively coupled to the processor, the memory having stored therein a program which, when executed by the processor, causes the processor to perform the method according to the first aspect.
According to a third aspect of the present disclosure, there is provided a computer-readable storage medium for detecting an abnormal port user in a resource management system of an optical network, comprising computer-executable instructions which, when executed by one or more processors, cause the one or more processors to perform the method according to the first aspect.
According to a fourth aspect of the present disclosure, there is provided a computer program product for detecting an abnormal port user in a resource management system of an optical network, comprising a computer program which, when executed by a processor, implements the method of the first aspect.
By utilizing the method and the device disclosed by the invention, the detection of the inconsistent user of the resource occupation and the site occupation can be realized, and at least some of the following advantages are realized:
The abnormal port users can be accurately found, errors in the resource management system are corrected, and service opening and configuration are facilitated;
After receiving the fault maintenance request, maintenance personnel can use the detected abnormal port user as a basis for resource checking task deployment, so that resource waste and manpower waste are avoided;
the whole condition of the user resource occupation accuracy can be evaluated according to the region, the inconvenience of spot check is eliminated, and the efficiency is improved.
The subject matter of the present disclosure is provided as examples of apparatuses, systems, and methods for performing the features described in the present disclosure. Other features or variations in addition to those described above are contemplated. It is contemplated that the implementation of the components and functions of the present disclosure may be accomplished with any emerging technology that may replace any of the above-described implementation technologies.
In addition, the foregoing description provides examples without limiting the scope, applicability, or configuration set forth in the claims. Changes may be made in the function and arrangement of elements discussed without departing from the spirit and scope of the disclosure. Various embodiments may omit, replace, or add various procedures or components as appropriate. For example, features described with respect to certain embodiments may be combined in other embodiments.
Similarly, although operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In some cases, multitasking and parallel processing may be advantageous.