Method and system for screening downstream devices of home gateway

Abstract

The present application discloses a method and system for screening downstream devices of a home gateway. The method comprises the following steps: real-time acquisition and storage of user probe data; identification of the type of the downstream device based on the probe data; determination of abnormal device detection criteria based on the type of the device and the probe data and detection of abnormal devices based on the criteria; and determination of non-home user scenarios by factor analysis or clustering algorithm on the probe data. The present application also discloses a corresponding system.

Description

Technical Field

[0001] The present application relates to the technical field of gateway downstream device management, and in particular to a management method and system for downstream devices based on home gateway probe data, and more specifically, to a method and system for screening downstream devices of a home gateway. Background Art

[0002] A residential gateway is a piece of hardware that connects a home or small office network to the Internet. A residential gateway provides switching capabilities and enables users to share a DSL or cable modem Internet connection with all computers on the internal network. The residential gateway sits between the DSL or cable modem and the internal network. It is also possible to integrate a DSL or cable modem into a residential gateway. A residential gateway has both modem and routing capabilities, and can connect a home LAN to an external WAN.

[0003] The devices actually used by a family user are important features of the entire family portrait, and are also an important dimension for precision marketing. However, the current home gateway devices in the prior art can only discover devices directly connected to them, and discovering these devices only means that these devices have been connected to this gateway, but cannot indicate whether they belong to the family user. Therefore, if this type of data is used directly for marketing, it will result in low marketing accuracy and unclear troubleshooting focus.

[0004] Therefore, being able to accurately filter out the devices under the home gateway that truly belong to the home user is very important for portraying the family portrait.

[0005] There is an urgent need in the art for a method and system that can truly screen out commonly used devices belonging to a family and connected to a home gateway. Summary of the invention

[0006] In order to solve the above problems, this application uses an anomaly detection algorithm to mark suspected non-resident devices, and on this basis introduces principal component analysis and clustering algorithms (both of which are the two most important unsupervised learning in machine learning, which will not be elaborated here), so as to eliminate non-home user scenarios, thereby achieving the purpose of accurately profiling home devices. Based on this, it is possible to accurately screen the downstream devices of the home gateway, so as to provide more accurate drawings and troubleshooting guarantees in the future.

[0007] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description given later.

[0008] Specifically, according to the first aspect of the present application, a method for screening downlink devices of a home gateway is provided, the method comprising the following steps:

[0009] Collect and store user's probe data in real time;

[0010] Identifying the type of the downlink device based on the probe data;

[0011] determining an abnormal device detection standard based on the type of the device and the probe data and detecting the abnormal device based on the standard; and

[0012] The non-home user scenario is determined by performing factor analysis or clustering algorithm on the probe data.

[0013] According to a preferred embodiment of the present application, identifying the type of the downstream device based on the probe data further includes: using a big data processing framework to process the probe data to identify the type of the downstream device.

[0014] According to a preferred embodiment of the present application, the abnormal device detection standard is determined based on the feature vectors of all downstream devices.

[0015] According to a preferred embodiment of the present application, detecting abnormal devices based on the standard further includes: if the number of active days of the downlink device is less than the abnormal device detection standard, determining it as an abnormal device, and otherwise determining it as a resident device.

[0016] According to a preferred embodiment of the present application, determining non-home user scenarios through factor analysis and clustering algorithms further includes: determining the feature vector of resident equipment; standardizing the feature vector; performing a significance check on the standardized feature vector; if the result is significant, performing factor analysis, otherwise, performing cluster analysis; judging whether the results of factor analysis or cluster analysis are reasonable; and if the result is reasonable, deleting abnormal data.

[0017] According to a second aspect of the present application, a system for screening downstream devices of a home gateway is provided, the system comprising the following modules:

[0018] Probe data collection module, used to collect and store user's probe data in real time;

[0019] A device identification module, used to identify the type of the downstream device based on the probe data;

[0020] an abnormal device detection module, configured to determine an abnormal device detection standard based on the type of the device and the probe data and detect abnormal devices based on the standard; and

[0021] The non-home scene detection module is used to determine the non-home user scene through factor analysis or clustering algorithm.

[0022] To achieve the foregoing and related ends, the one or more aspects include the features fully described below and particularly pointed out in the appended claims. The following description and drawings set forth in detail certain illustrative features of the one or more aspects. However, these features are merely indicative of several of the various ways in which the principles of the various aspects may be employed, and this description is intended to cover all such aspects and their equivalents. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to understand in detail the manner in which the above-stated features of the present application are used, a more specific description of the above briefly summarized contents may be made with reference to various aspects, some of which are illustrated in the accompanying drawings. However, it should be noted that the accompanying drawings illustrate only certain typical aspects of the present application and should not be considered to limit its scope, as the present description may allow for other equally effective aspects.

[0024] In the attached picture:

[0025] Figure 1 is a flowchart illustrating a method for screening commonly used devices connected to a home gateway according to an embodiment of the present application;

[0026] Figure 2 is a structural block diagram of a system for screening commonly used devices connected to a home gateway according to an embodiment of the present application;

[0027] Figure 3 is a schematic diagram for explaining an abnormal device detection model of a method and system for screening commonly used devices connected to a home gateway according to an embodiment of the present application;

[0028] Figure 4 is a schematic diagram illustrating a non-home scene detection model of a method and system for screening commonly used devices connected to a home gateway according to an embodiment of the present application; and

[0029] Figure 5 Is the explanation Figure 4 An exemplary sample calculation process of the non-home scene detection model shown in . DETAILED DESCRIPTION

[0030] The detailed description set forth below in conjunction with the accompanying drawings is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. This detailed description includes specific details to provide a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well-known modules are shown in block diagram form in order to avoid diluting such concepts.

[0031] It is to be understood that other embodiments would be apparent based on the present disclosure, and that system, structural, process, or mechanical changes may be made without departing from the scope of the present disclosure.

[0032] Reference Figure 1 and 2 , aspects are described with reference to one or more modules and one or more methods that can perform the actions or functions described herein. In one aspect, the term "module" as used herein can be one of the parts that make up a system, can be hardware or software or some combination thereof, and can be divided into other components or modules. Although the following Figure 1 The operations described in the present invention are presented in a specific order and / or as being performed by example modules, but it should be understood that the order of these actions and the modules that perform the actions may vary depending on the implementation. In addition, it should be understood that the following actions or functions may be performed by a specially programmed processor, a processor that executes specially programmed software or a computer-readable medium, or by any other combination of hardware modules and / or software modules that are capable of performing the described actions or functions.

[0033] Figure 1 It is a flowchart illustrating a method 100 for screening commonly used devices connected to a home gateway according to an embodiment of the present application.

[0034] like Figure 1 As shown in , the method 100 mainly includes the following steps.

[0035] Collect and store the user's probe data in real time. Probe data includes but is not limited to the device information connected to the gateway, device connection method information, etc.

[0036] Based on the collected device probe data, the brand, type and model of the gateway downstream device are identified (step 102)

[0037] Subsequently, according to the device probe data collected within a certain period and based on the determined device type, each downstream device is detected to determine suspected abnormal devices and mark them (step 103). Specifically, a big data processing framework can be used to process the collected probe data to calculate the feature vectors of each device downstream of the gateway, such as but not limited to the device type, the number of days the device is online, etc., and the detection standard of abnormal devices is determined based on the overall feature vector.

[0038] By establishing an abnormal device detection model (described in detail below), the anomaly detection algorithm is combined with the degree to which the gateway discovers its downstream devices, and the original anomaly detection algorithm is improved in a scenario-based manner, so that the model can play a role in unique data scenarios, and the model interpretation is more in line with users' actual device usage habits.

[0039] Finally, the use scenarios of the gateway are pre-judged to identify the scenarios of suspected non-family users (step 104). This step mainly uses principal component analysis and clustering algorithms to eliminate non-family user scenarios, thereby achieving the purpose of accurately profiling family devices.

[0040] Specifically, based on the abnormal device detection standard determined according to big data, the resident devices of the gateway can be obtained, thereby obtaining the device type feature vector of each gateway, and through principal component analysis and clustering algorithm, non-home gateway data can be eliminated.

[0041] By introducing a non-home scenario detection model (described in detail below), the scenarios used by the gateway's downstream devices are judged and suspected scenarios are marked, allowing for an accurate portrait of home users, laying the foundation for providing personalized services.

[0042] Figure 2 The structure of the system for screening commonly used devices connected to a home gateway according to an embodiment of the present application is explained in the figure.

[0043] like Figure 2 As shown in , the system 200 according to the present application mainly includes the following modules: a probe data acquisition module 201; a device identification module 202; an abnormal device detection module 203; and a non-home scene detection module 204.

[0044] The following combination Figure 2 To introduce the specific functions and implementation methods of each module in detail.

[0045] Probe data acquisition module 201

[0046] The probe data collection module is mainly used to collect and store the user's probe data in real time. Probe data includes but is not limited to the device information connected to the gateway, the device connection mode information, etc.

[0047] Device identification module 202

[0048] The device identification module is mainly used to identify the brand, type and model of the gateway downstream device based on the collected device probe data.

[0049] Abnormal device detection module 203

[0050] The abnormal device detection module is mainly used to detect resident devices based on the device probe data collected within a certain period to determine suspected abnormal devices and mark them. Specifically, the module can use the big data processing framework to process the collected probe data, thereby calculating the feature vectors of each device connected to the gateway, such as but not limited to the device type, the number of days the device has been online, etc., and determine the abnormal device detection standard based on the overall feature vector. In addition, the module determines whether the device connected to the gateway is an abnormal device based on the abnormal device detection standard.

[0051] Non-home scene detection module 204

[0052] The non-home scene detection module is mainly used to predict the use scenarios of the gateway to identify the scenes of suspected non-home users. This module mainly uses principal component analysis and clustering algorithms to eliminate non-home user scenes, thereby achieving the purpose of accurate home device portrait.

[0053] Specifically, based on the abnormal device detection standard determined by the abnormal device detection module according to big data, the resident devices of the gateway are obtained, thereby obtaining the device feature vectors of each gateway, and non-home gateway data are eliminated through principal component analysis and clustering algorithms.

[0054] Figure 3 The schematic diagram of the abnormal device detection model of the method and system for screening commonly used devices connected to a home gateway according to an embodiment of the present application is further explained in detail.

[0055] like Figure 3 As shown in , according to the abnormal device detection model, for example, samples of downstream devices within two weeks of each gateway are extracted.

[0056] Then, based on these samples, the abnormal upper limit (qu) and lower limit (ql) of each gateway's active days can be calculated. According to the abnormal upper limit qu and abnormal lower limit ql of each gateway, the number of devices nu with active days higher than qu, the number of devices nl with active days lower than ql, and the total number ns are determined.

[0057] Then, based on these samples, the average number of gateway-connected devices in all samples (x) is calculated. Specifically, all samples are calculated by type. For example, for mobile phone type, the value is the total number of mobile phones directly connected to the gateway divided by the total number of users in the entire network.

[0058] Combined with the number of devices nu with active days higher than qu, the number of devices nl with active days lower than ql, the total number of the two ns, and the average number of downstream devices x, the detection standard rule for abnormal devices is determined according to the following formula:

[0059] If abs(ns-nl-x)≤abs(nu-x), then rule=ql;

[0060] Otherwise rule = qu.

[0061] Based on this standard, if the number of active days of the downstream devices of each gateway is ≥ rule, they are resident devices of the gateway.

[0062] Figure 4 A schematic diagram of a non-home scene detection model of a method and system for screening commonly used devices connected to a home gateway according to an embodiment of the present application is further explained in detail.

[0063] like Figure 4 As shown in , according to the non-home scene detection model, the data of the resident devices connected to each gateway are extracted and the feature vectors of the resident devices are determined based on the data (step 401).

[0064] For example, the eigenvector results are as follows:

[0065] [{Number of unknown devices},{Number of OTHER devices},{Number of computers},{Number of pan-intelligent devices},{Number of routers},{Number of mobile phones},{Number of PADs},{Number of network set-top boxes}]

[0066] Next, the feature vectors of these resident devices are normalized (step 402). This can be achieved, for example but not limited to, by using a min_max normalization method. Of course, the commonly used data normalization methods in the art are all within the scope of this application.

[0067] Thereafter, a significance test is performed (step 403). For example, the KMO test and the Bartlett test can be used to perform the KMO test on the standardized data. Figure 5 As shown in , the value after verification is 0.6299059749541638, so the Bartlett test is significant greater than 0.05. As described below, this result is not suitable for factor analysis, so cluster analysis is performed directly.

[0068] At step 404, it is determined whether the result of the significance check is significant.

[0069] If the result is significant, factor analysis is performed at step 405. Factor analysis mainly includes principal component analysis and factor rotation.

[0070] Factor analysis is a statistical technique for studying the extraction of common factors from a group of variables. It is also a well-known analytical technique in the field of statistics, so it will not be described in detail here so as not to affect the description of the main technical solution. Principal component analysis (PCA) is a technique for simplifying data sets. It is a linear transformation. This transformation transforms the data into a new coordinate system so that the first largest variance of any data projection is on the first coordinate (called the first principal component), the second largest variance is on the second coordinate (the second principal component), and so on. Principal component analysis is often used to reduce the dimensionality of a data set while retaining the features of the data set that contribute the most to the variance. This is done by retaining low-order principal components and ignoring high-order principal components. In this way, low-order components can often retain the most important aspects of the data. Given that PCA is a commonly used dimensionality reduction analysis method in this field, it will not be described in detail here so as not to affect the description of the main technical solution.

[0071] If the result is not significant, at step 406, cluster analysis is performed on the result of the significance check. The cluster analysis can adopt the k-means algorithm. The k-means clustering algorithm is a commonly used iterative clustering analysis algorithm, so it is not described here in detail to avoid affecting the description of the main technical solution.

[0072] like Figure 5 As shown in , first determine the initial value (number of categories) k, and then use the inflection point method to obtain the additional Figure 5 The results shown in .

[0073] The horizontal axis x represents the value of the number of clusters k (i.e., the number of categories), and the vertical axis y represents the TSSE value (the sum of squared deviations within the cluster). Figure 5 It can be seen that when k is 3, it is a more reasonable result.

[0074] Regardless of whether factor analysis or cluster analysis is performed, the result is determined at step 407 to determine whether the analysis result is reasonable.

[0075] If the analysis result is reasonable, the abnormal data is deleted in step 408 .

[0076] If the categories are not significant, the process ends.

[0077] If the result at step 407 is not reasonable, the process also ends.

[0078] Through this model, non-home devices can be eliminated, thereby achieving the goal of accurately portraying home devices.

[0079] The implementation of the system and method according to the present application is described in detail below in conjunction with specific embodiments.

[0080] For example, for customer 00E******1C4, 128 mobile device MAC addresses can be identified in one month, and the specific online time is shown in the following table.

[0081] Online days Mobile device MAC number 1 125 6 1 9 1 31 1

[0082] Refer to the attached Figure 5 , with the help of abnormal device detection module 203, it can be determined based on the abnormal device detection standard that the customer has 3 resident mobile phones (i.e., the number of devices with online days greater than 1). Afterwards, a device feature vector is established for the customer's resident devices: {UNKNOWN:10,OTHER:0,PC:0,SMTDEV:0,ROUTER:1,PHONE:3,PAD:0,STB:0}. Based on more than 1.6 million samples, the k-means clustering algorithm can be used to determine that the customer is a normal user.

[0083] Furthermore, for customer 6CD******A07, with the help of abnormal device detection module 203, the customer's resident devices are determined by abnormal device detection standards. Then, the device feature vectors for these resident devices are determined as: {UNKNOWN:3397, OTHER:0, PC:6, SMTDEV:27, ROUTER:3, PHONE:5, PAD:0, STB:0}. According to more than 1.6 million samples, through k-means clustering, it is found that the customer is an abnormal user (that is, determined to be a non-home scenario). According to actual business, the proportion of commercial use of home gateways is not high, so the result is in line with expectations.

[0084] Compared with the prior art, the method and system for screening commonly used devices connected to a home gateway according to the present application mainly have the following advantages.

[0085] First, by establishing an abnormal device detection model, the abnormal device detection algorithm is combined with the degree to which the gateway discovers its downstream devices, and the original abnormal device detection algorithm is improved in a scenario-based manner. This allows the model to play a role in unique data scenarios, and the model interpretation is more in line with users' actual device usage habits.

[0086] Second, a non-home scenario detection model was introduced to label the suspected usage scenarios of the device, providing a basis for subsequent personalized services.

[0087] According to various aspects, elements, or any part of elements, or any combination of elements of the present disclosure, a "processing system" including one or more processors can be implemented. Examples of processors include: microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform various functionalities described throughout the present disclosure. One or more processors in a processing system can execute software. Software should be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether it is described in software, firmware, middleware, microcode, hardware description language, or other terms. Software may reside on a computer-readable medium. The computer-readable medium may be a non-transient computer-readable medium. As examples, non-transitory computer-readable media include: magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips), optical disks (e.g., compact disks (CDs), digital versatile disks (DVDs)), smart cards, flash memory devices (e.g., memory cards, memory sticks, key drives), random access memory (RAM), read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), registers, removable disks, and any other suitable medium for storing software and / or instructions that can be accessed and read by a computer. As examples, computer-readable media may also include carrier waves, transmission lines, and any other suitable medium for transmitting software and / or instructions that can be accessed and read by a computer. The computer-readable medium may reside in the processing system, be external to the processing system, or be distributed across multiple entities including the processing system. The computer-readable medium may be implemented in a computer program product. As examples, a computer program product may include a computer-readable medium in packaging materials. Those skilled in the art will recognize how to best implement the described functionality presented throughout this disclosure depending on the specific application and the overall design constraints imposed on the overall system.

[0088] It should be understood that the specific order or hierarchy of steps in the disclosed methods is an illustration of an exemplary process. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the methods or methodologies described herein may be rearranged. The attached method claims present elements of the various steps in a sample order and are not meant to be limited to the specific order or hierarchy presented unless specifically stated herein.

[0089] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the universal principles defined herein may be applied to other aspects. Therefore, the claims are not intended to be limited to the various aspects shown herein, but should be granted the full scope consistent with the language of the claims, wherein the singular reference to the element is not intended to mean "there is and only one" (unless specifically stated) but "one or more". Unless specifically stated otherwise, the term "some" refers to one or more. The phrase "at least one" quoting a list of items refers to any combination of these items, including a single member. As an example, "at least one of a, b or c" is intended to cover: at least one a; at least one b; at least one c; at least one a and at least one b; at least one a and at least one c; at least one b and at least one c; and at least one a, at least one b and at least one c. The elements of the various aspects described throughout this disclosure are all structurally and functionally equivalent schemes currently or hereafter known to those of ordinary skill in the art, and are expressly incorporated herein by reference, and are intended to be covered by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims.

Claims

1. A method for screening downstream devices of a home gateway, characterized in that: The method comprises the following steps: Collect and store user's probe data in real time; identifying the type of the downlink device based on the probe data; Determine an abnormal device detection standard based on the type of the device and the probe data and detect abnormal devices based on the standard; and determine a non-home user scenario by performing factor analysis or a clustering algorithm on the probe data; wherein the abnormal device is detected based on the standard, including: if the number of active days of the downstream device is less than the abnormal device detection standard, it is determined to be an abnormal device, and otherwise, it is determined to be a resident device; the non-home user scenario is determined by factor analysis and a clustering algorithm, including: determining a feature vector of a resident device; standardizing the feature vector; performing a significance check on the standardized feature vector; if the result is significant, perform a factor analysis, otherwise, perform a cluster analysis; determine whether the result of the factor analysis or cluster analysis is reasonable; and if the result is reasonable, delete the abnormal data.

2. The method according to claim 1, characterized in that Identifying the type of the downstream device based on the probe data further includes: The probe data is processed using a big data processing framework to identify the type of the downstream device.

3. The method according to claim 1, characterized in that The abnormal device detection standard is determined based on the feature vectors of all downstream devices.

4. A system for screening downstream devices of a home gateway, characterized in that: The system includes the following modules: Probe data collection module, used to collect and store user's probe data in real time; A device identification module, used to identify the type of the downstream device based on the probe data; an abnormal device detection module, configured to determine an abnormal device detection standard based on the type of the device and the probe data and detect abnormal devices based on the standard; and a non-home scenario detection module, for determining non-home user scenarios by factor analysis or clustering algorithms; The detecting of abnormal devices based on the standard includes: if the number of active days of the downstream device is less than the abnormal device detection standard, determining it as an abnormal device, and otherwise, determining it as a resident device; the determining of non-home user scenarios through factor analysis and clustering algorithms includes: determining the feature vector of the resident device; standardizing the feature vector; performing significance check on the standardized feature vector; if the result is significant, performing factor analysis, otherwise, performing cluster analysis; judging whether the result of factor analysis or cluster analysis is reasonable; and if the result is reasonable, deleting the abnormal data.

5. The system according to claim 4, characterized in that Identifying the type of the downstream device based on the probe data further includes: The probe data is processed using a big data processing framework to identify the type of the downstream device.

6. The system according to claim 4, characterized in that The abnormal device detection standard is determined based on the feature vectors of all downstream devices.

Images