FTTR network data quality optimization method, device, equipment and medium

Abstract

The invention discloses an FTTR network data quality optimization method and apparatus, a device and a medium. The method comprises the steps of collecting various network performance data of a target terminal device; for any kind of network performance data, obtaining a plurality of missing intervals of the network performance data and calculating the correlation degree between the network performance data and other network performance data; for any missing interval, determining a plurality of target intervals of the missing interval from the network performance data, calculating an influence factor between the missing interval and each target interval based on the correlation degree, and taking the target interval corresponding to the minimum value of the influence factor as a reference interval of the missing interval; and interpolating the network performance data by using a preset data interpolation algorithm, and correcting an interpolated result based on the reference interval of each missing interval to generate corrected network performance data. According to the method, the problem of relatively large interpolation result deviation caused by neglecting index relevance in a traditional univariate interpolation method is solved.

Description

Technical Field

[0001] The present application relates to the technical field of data quality optimization, and in particular to a method, device, equipment and medium for optimizing FTTR network data quality. Background Art

[0002] FTTR (Fiber to the Room) is a fiber-optic communication technology for home or enterprise scenarios. It is designed to transmit high-speed Internet signals directly to terminal devices in each room through optical fiber for use by devices in home or office environments. In actual applications, the network performance data of terminal devices may be missing due to propagation delays or network interruptions. In this case, data interpolation is needed to fill in the missing items.

[0003] Traditional interpolation methods usually only rely on the time series characteristics of the interpolated indicator itself to fill in the missing items. For example, when filling in the missing values ​​of signal strength, calculations are only based on the historical data of signal strength. However, there is a dynamic correlation between different types of network performance data. For example, when the signal strength decreases, the bandwidth utilization rate often increases synchronously. Traditional interpolation methods ignore the interaction between different network performance data, resulting in a large deviation between the interpolation results and the real data, affecting the overall data quality optimization effect. Summary of the invention

[0004] The main purpose of this application is to provide a FTTR network data quality optimization method, device, equipment and medium, aiming to solve the technical problem that the traditional single variable interpolation method ignores the correlation of indicators and causes large deviations in the interpolation results.

[0005] To achieve the above-mentioned purpose, the present application provides a method for optimizing the quality of FTTR network data, including: collecting multiple network performance data of a target terminal device, wherein the target terminal device is a terminal device under the FTTR architecture; for any network performance data of the target terminal device, obtaining multiple missing intervals of the network performance data and calculating the degree of correlation between the network performance data and other network performance data; for any missing interval, determining multiple target intervals of the missing interval from the network performance data, calculating the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the degree of correlation between the network performance data and other network performance data, and taking the target interval corresponding to the minimum value of the impact factor as the reference interval of the missing interval; interpolating the network performance data using a preset data interpolation algorithm, and correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0006] Optionally, the calculation of the impact factor between the missing interval and each target interval based on the network performance data, other network performance data and the degree of correlation between the network performance data and other network performance data includes: calculating the degree of difference in actual numerical values ​​between the missing interval and each target interval based on the network performance data; calculating the degree of difference in correlation indicators between the missing interval and each target interval based on other network performance data and the degree of correlation between the network performance data and other network performance data; and determining the impact factor between the missing interval and each target interval based on the actual numerical difference and the correlation indicator difference.

[0007] Optionally, the method of calculating the degree of difference in correlation indicators between the missing interval and each target interval based on other network performance data and the degree of correlation between the network performance data and other network performance data includes: for any target interval, taking the time interval corresponding to the missing interval as the missing time period, and taking the time interval corresponding to the target interval as the target time period, and calculating the numerical difference between the missing time period and the target time period of other network performance data; determining the degree of difference in correlation indicators between the missing interval and the target interval based on each numerical difference and the degree of correlation between the network performance data and other network performance data.

[0008] Optionally, the calculating the degree of difference in the correlation index between the missing interval and each target interval based on other network performance data and the degree of correlation between the network performance data and other network performance data includes: calculating the degree of difference in the correlation index between the missing interval and each target interval using the following formula (1):

[0009] In the formula, Indicates that the missing interval is The degree of difference in the correlation indicators between the target intervals, Indicates the number of network performance data. In addition to the network performance data, the degree of correlation between the network performance data and the network performance data, Indicates that the missing time period is The network performance data corresponds to numerical values, Indicates that the target time period is The network performance data corresponds to numerical values, Indicated in The number of network performance data shared by the missing time period and the target time period.

[0010] Optionally, before correcting the interpolated result based on the reference interval of each missing interval to generate the corrected network performance data, the method also includes: for any missing interval, obtaining the time interval between the reference interval of the missing interval and the missing interval; correcting the interpolated result based on the reference interval of each missing interval to generate the corrected network performance data includes: for any missing interval, correcting each interpolated data point in the missing interval based on the reference interval of the missing interval and the time interval to obtain a corrected missing interval; generating the corrected network performance data based on each corrected missing interval.

[0011] Optionally, the step of correcting each interpolated data point in the missing interval based on the reference interval of the missing interval and the time interval comprises: correcting each interpolated data point in the missing interval using the following formula (2):

[0012] In the formula, Indicates the missing interval The corrected value of the data point, Indicates the missing interval The interpolation result of data points is Indicates that the missing interval is within the reference interval and The value of the data point corresponding to the data point, The time interval between the reference interval representing the missing interval and the missing interval.

[0013] Optionally, each target interval of the missing interval has the same number of data as that contained in the missing interval.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a FTTR network data quality optimization device, including: a data acquisition module, used to collect multiple network performance data of a target terminal device, wherein the target terminal device is a terminal device under the FTTR architecture; a correlation degree construction module, used to obtain multiple missing intervals of the network performance data for any type of network performance data of the target terminal device and calculate the correlation degree between the network performance data and other network performance data; a reference interval acquisition module, used to determine multiple target intervals of the missing interval from the network performance data for any missing interval, calculate the impact factor between the missing interval and each target interval based on the network performance data, other network performance data and the correlation degree between the network performance data and other network performance data, and use the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; an interpolation correction module, used to interpolate the network performance data using a preset data interpolation algorithm, and correct the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0015] The present application also provides a FTTR network data quality optimization device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method in any one of the above possible implementations.

[0016] The present application also provides a computer-readable storage medium, comprising: a computer program stored therein, wherein when the computer program is executed by a processor, the method in any possible implementation manner described above is implemented.

[0017] The present application proposes a method, device, equipment and medium for optimizing the quality of FTTR network data. First, for any network performance data of the target terminal device, multiple missing intervals of the network performance data are obtained and the correlation between the network performance data and other network performance data is calculated; secondly, for any missing interval, multiple target intervals of the missing interval are determined from the network performance data, the impact factor between the missing interval and each target interval is calculated, and the target interval corresponding to the minimum impact factor is used as the reference interval of the missing interval; finally, the network performance data is interpolated using a preset data interpolation algorithm, and the interpolated result is corrected based on the reference interval of each missing interval to generate the corrected network performance data. The present application solves the technical problem that the traditional univariate interpolation method ignores the correlation of indicators, resulting in large deviations in the interpolation results. It can effectively restore the real state when the data is missing and improve the credibility and accuracy of the completed data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of the FTTR network data quality optimization method provided in Example 1 of the present application; Figure 2 A flowchart of a method for optimizing FTTR network data quality provided in Example 2 of the present application; Figure 3 A structural block diagram of a FTTR network data quality optimization device provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of a FTTR network data quality optimization device provided in one embodiment of the present application.

[0019] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0021] Traditional interpolation methods usually only rely on the time series characteristics of the interpolated indicator itself to fill in the missing items. For example, when filling in the missing values ​​of signal strength, calculations are only based on the historical data of signal strength. However, there is a dynamic correlation between different types of network performance data. For example, when the signal strength decreases, the bandwidth utilization rate often increases synchronously. Traditional interpolation methods ignore the interaction between different network performance data, resulting in a large deviation between the interpolation results and the real data, affecting the overall data quality optimization effect.

[0022] To solve the above problems, the present application provides a method, device, equipment and medium for optimizing FTTR network data quality. The present application scheme is introduced in detail below.

[0023] Figure 1 The flowchart of the FTTR network data quality optimization method provided in the first embodiment of the present application can be applied to the FTTR network architecture, wherein the FTTR architecture can include a management platform, a master gateway, a slave gateway, a fiber optic splitter, and a terminal device. The operator's fiber optic accesses the master gateway, and the master gateway distributes the optical signal to each slave gateway through a fiber optic splitter. Each slave gateway is connected to each terminal device in each room by wire or wireless means to provide network services for it. The FTTR network data quality optimization method can be executed by a data quality optimization device that is in communication with each terminal device. For example, the data quality optimization device can be a FTTR network data quality optimization device. Figure 1 The FTTR network data quality optimization method may include the following steps: S11, collecting various network performance data of the target terminal device, where the target terminal device is a terminal device under the FTTR architecture.

[0024] Among them, the network performance data is used to characterize the network connection status and network operation status of each terminal device. The various network performance data can be optical power data, optical fiber attenuation data, signal interference data, etc.

[0025] During the specific implementation process, the network performance data of the target terminal device is obtained through the management platform or the management interface of each terminal device.

[0026] It should be noted that, in this embodiment, the main gateway or the management platform can send a synchronous sampling instruction to the terminal device through the network protocol to ensure that the timestamps of the collected network performance data are aligned.

[0027] S12, for any type of network performance data of the target terminal device, obtain multiple missing intervals of the network performance data and calculate the correlation between the network performance data and other network performance data.

[0028] Among them, the missing interval represents the interval composed of continuous data missing items in the network performance data due to propagation delay or network interruption, and the degree of correlation can reflect the dynamic correlation between the two network performance data.

[0029] In the specific implementation process, taking any network performance data of the target terminal device as an example, firstly, multiple missing intervals of the network performance data are obtained. It can be understood that in this embodiment, a single missing data item is also recorded as a missing interval.

[0030] Furthermore, the Pearson correlation coefficient is used to calculate the correlation between the network performance data and other network performance data.

[0031] It should be noted that, in other embodiments, the correlation between network performance data may also be calculated using partial correlation coefficients or mutual information, and this embodiment does not specifically limit the algorithm used to calculate the correlation. Since two network performance data may be negatively correlated, that is, the obtained correlation coefficient value is a negative number, this embodiment first takes the absolute value of the result obtained by calculating the correlation coefficient, and then uses the obtained absolute value as the correlation between the two network performance data.

[0032] It should be further explained that when the change range of a certain network performance data at both ends of a missing interval is small, but other network performance data associated with it in the corresponding period fluctuates significantly, if only the data at adjacent time points of the missing interval are used for interpolation, the dynamic correlation characteristics cannot be captured, causing the interpolation result to deviate from the actual physical state. For example: in the missing interval of a certain section of optical power data, due to the sudden change of ambient temperature, the optical fiber attenuates abnormally. At this time, the optical power values ​​at both ends of the missing interval only show a slight change of ±0.1dB, but the bit error rate in the same period increases sharply from 1E-9 to 1E-6. Directly using the optical power values ​​at both ends of the missing interval for data interpolation will miss the actual nonlinear attenuation changes.

[0033] Based on this, this embodiment uses the following method to improve the problems existing in the traditional interpolation algorithm (i.e., using data from adjacent time points of the missing interval for interpolation): first, extract a target interval with the same length as the missing interval from the network performance data, and then use the interval in the target interval that best meets the following two conditions as the reference interval for the missing interval: 1. The difference between the data at both ends of the interval and the data at both ends of the missing data is small; 2. The change trend of other network performance data in the time period corresponding to the interval is similar to the change trend of other network performance data in the time period corresponding to the missing interval. Finally, the reference interval is used to correct the data interpolated by the traditional interpolation algorithm for the missing interval, which can effectively restore the actual state when the data is missing and improve the credibility and accuracy of the completed data.

[0034] S13, for any missing interval of any type of network performance data, determine multiple target intervals of the missing interval from the network performance data, calculate the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data, and use the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; S14, interpolating the network performance data using a preset data interpolation algorithm, and correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0035] Among them, the impact factor represents the correction credibility of each target interval when the target interval is used to correct the result after interpolation of the missing interval.

[0036] In a specific implementation process, multiple target intervals for each missing interval are determined from the network performance data, each target interval contains the same number of data as the missing interval, and each target interval has no missing data.

[0037] It should be noted that, in this embodiment, the target interval of each missing interval does not include the missing interval, and there may be sampling time overlap between the target intervals. For example, if a certain network performance data uses seconds as the sampling time unit, a missing interval in the network performance data is from the 7th second to the 10th second, and there is no missing data from the 1st second to the 6th second of the network performance data, then starting from the first sampling point of the network performance data, the target interval of the missing interval is: from the 1st second to the 4th second, from the 2nd second to the 5th second, from the 3rd second to the 6th second, and so on.

[0038] It can be understood that, taking any missing interval as an example, the target interval of the missing interval and the missing interval are both included in the same network performance data.

[0039] Furthermore, the influence factor between the missing interval and each target interval is calculated based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data.

[0040] In one embodiment, in step S13, calculating the influence factor between the missing interval and each target interval based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data may specifically include: S131, calculating the actual numerical difference between the missing interval and each target interval based on the network performance data; S132, calculating the difference degree of the correlation index between the missing interval and each target interval based on other network performance data and the correlation degree between the network performance data and other network performance data; S133. Determine the impact factor between the missing interval and each target interval based on the difference degree between the actual numerical value and the difference degree between the correlation index.

[0041] The degree of difference in actual values ​​represents the numerical difference between the observed values ​​(actual values) at both ends of the missing interval and the target interval, and the degree of difference in associated indicators represents the difference between the changing trend of other network performance data in the time period corresponding to the missing interval and the changing trend of other network performance data in the time period corresponding to the missing data.

[0042] In the specific implementation process, the absolute value of the difference between the observation value at one end of the missing interval and the observation value at one end of each target interval is first calculated, and then the absolute value of the difference between the observation value at the other end of the missing interval and the observation value at the other end of each target interval is calculated, and the sum of the two absolute values ​​of the difference is used as the actual numerical difference between the missing interval and each target interval. For example: there is a missing interval from the 7th second to the 10th second in a certain network performance data, and there is no missing data from the 1st second to the 6th second of the network performance data. Taking the target interval from the 2nd second to the 5th second of the missing interval as an example, the absolute value of the difference between the observation value at one end of the missing interval (i.e., the data corresponding to the 6th second) and the observation value at one end of the target interval (i.e., the data corresponding to the 1st second) is first calculated, and then the absolute value of the difference between the observation value at the other end of the missing interval (i.e., the data corresponding to the 11th second) and the observation value at the other end of the target interval (i.e., the data corresponding to the 6th second) is calculated, and the two absolute values ​​of the difference are added to obtain the actual numerical difference between the missing interval and the target interval.

[0043] It should be noted that if there is no data at one end of the target interval, for example, the target interval is from the 1st second to the 4th second, the first data of the target interval is used as the observation value at one end of the target interval. In addition, if there is no data at one end of the missing interval, for example, if the missing interval starts from the 1st second, only the absolute value of the difference between the observation value at the other end of the missing interval and the observation value at the other end of the target interval is calculated, and the absolute value of the difference is used as the actual numerical difference between the missing interval and the target interval.

[0044] Furthermore, taking any target interval as an example, the time interval corresponding to the target interval is taken as the target time period, and the time interval corresponding to the missing interval is taken as the missing time period, and the numerical difference between other network performance data in the missing time period and the target interval in the target time period is calculated.

[0045] It is understandable that the calculation of the numerical difference between the other network performance data in the missing time period and the target interval in the target time period is to obtain the difference between the change trend of other network performance data in the time period corresponding to the target interval and the change trend of other network performance data in the time period corresponding to the missing interval. For example, taking any other network performance data as an example, if the numerical difference between the network performance data in the missing time period and the target time period is small, it means that the change trend of the network performance data in the missing time period is similar to that in the target time period.

[0046] Furthermore, the degree of difference in the correlation index between the missing interval and the target interval is determined based on the numerical difference between other network performance data in the missing time period and the target time period and the degree of correlation between the network performance data in the target time period and other network performance data.

[0047] Specifically, the missing interval For example, the missing interval and the first target interval can be calculated using the following formula (1): The degree of difference in correlation indicators between target intervals :

[0048] In the formula, Indicates the missing interval and the The degree of difference in the correlation indicators between the target intervals, Indicates the number of network performance data. Indicates that in addition to the network performance data, The degree of correlation between the network performance data and the network performance data, Indicates that the missing time period is The network performance data corresponds to numerical values, Indicates that the target time period is The network performance data corresponds to numerical values, Indicated in The number of network performance data shared by the missing time period and the target time period.

[0049] It should be noted that, after obtaining the correlation between the network performance data and other network performance data, this embodiment linearly normalizes all correlations, that is, the correlation values ​​fall within the range of 0 to 1. The network performance data is the first Values Does not exist (i.e., is a missing value) or The first network performance data in the target time period Values If it does not exist (i.e. it is a missing value), it will not be calculated. , that is, the sum of Differences between non-existent (missing) data are not included.

[0050] Understandably, Indicates the number of types of network performance data other than the network performance data in the missing interval. Indicates The numerical difference between the missing time period and the target time period of network performance data (i.e., the sum of the numerical differences of the corresponding bits), if It is smaller, indicating that the numerical difference (that is, the sum of the numerical differences of the corresponding bits) is smaller, and the change trend of the network performance data in the missing time period is similar to that in the target time period. Indicates the degree of correlation. By taking the degree of correlation as the weight, more attention can be paid to the network performance data with a greater degree of correlation with the network performance data. The smaller it is, the closer the change trends of other network performance data in the missing time period are to those in the target time period.

[0051] Furthermore, the actual numerical difference between the missing interval and each target interval and the difference between the associated indicators are weighted and summed to obtain the impact factor between the missing interval and each target interval, and the target interval corresponding to the minimum impact factor is used as the reference interval of the missing interval. Among them, the weights of the actual numerical difference and the associated indicator difference are both decimals between 0 and 1 (excluding 0 and 1), and the sum of the two weights is 1.

[0052] It should be noted that in order to avoid the influence of different dimensions (i.e., there is a difference in the numerical range between the actual numerical difference degree and the correlation indicator difference degree), this embodiment can first perform linear normalization on both the numerical difference degree and the correlation indicator difference degree, that is, linearly adjust the numerical difference degree to the range of 0 to 1, and also linearly adjust the correlation indicator difference degree to the range of 0 to 1.

[0053] It can be understood that the smaller the difference between the actual values ​​of the missing interval and a target interval, the smaller the difference between the data at both ends of the target interval and the data at both ends of the missing data; the smaller the difference between the correlation indicators of the missing interval and a target interval, the similarity between the change trend of other network performance data in the time period corresponding to the target interval and the change trend of other network performance data in the time period corresponding to the missing interval; the smaller the influence factor is obtained by multiplying the actual value difference with the correlation indicator difference to construct the influence factor, the similarity between the value fluctuation of the target interval and the missing interval and the similarity between the influence of other network performance data, and thus the target interval with the smallest influence factor can be selected as the reference interval of the missing interval, so as to correct the difference of the reference interval.

[0054] Therefore, this embodiment obtains the reference interval of each missing interval based on the dynamic correlation between each network performance data, and then the reference interval can be used to correct the data interpolated by using the traditional interpolation algorithm.

[0055] During the specific implementation process, the network performance data is interpolated using a preset data interpolation algorithm, wherein the preset data interpolation algorithm may be a linear interpolation algorithm. In other embodiments, the preset data interpolation algorithm may also be other interpolation algorithms, such as a spline interpolation algorithm or a polynomial interpolation algorithm. This embodiment does not specifically limit the preset data interpolation algorithm.

[0056] Furthermore, the interpolated results are corrected based on the reference intervals of each missing interval to generate corrected network performance data.

[0057] Specifically, take the first Take a data point as an example, preset adjustment parameters, and use the following formula (3) to correct the data point:

[0058] In the formula, Indicates the missing interval The corrected value of the data point, Indicates the missing interval The interpolation result of data points is Indicates that the missing interval is within the reference interval and The value of the data point corresponding to the data point, represents the adjustment parameter. is a decimal between 0 and 1, for example, It can be 0.7.

[0059] It can be understood that if the missing interval is from the 7th second to the 10th second and the reference interval is from the 20th second to the 23rd second, the data point corresponding to the 8th second data in the missing interval is the data point at the 21st second in the reference interval, and the data point corresponding to the 9th second data in the missing interval is the data point at the 22nd second in the reference interval.

[0060] It should be noted that if the reference interval of the missing interval is within the same range as the Data points corresponding to the data points If it does not exist (i.e. it is a missing value), then this data point is not applicable. Make corrections, that is, at this time, the missing interval The corrected value of the data point Still equal to .

[0061] The embodiment of the present application proposes a method for optimizing the quality of FTTR network data. First, for any network performance data of any terminal device, multiple missing intervals of the network performance data are obtained and the correlation between the network performance data and other network performance data is calculated; secondly, for any missing interval, multiple target intervals of the missing interval are determined from the network performance data, the impact factor between the missing interval and each target interval is calculated, and the target interval corresponding to the minimum impact factor is used as the reference interval of the missing interval; finally, the network performance data is interpolated using a preset data interpolation algorithm, and the interpolated result is corrected based on the reference interval of each missing interval to generate the corrected network performance data. The embodiment of the present application solves the technical problem that the traditional univariate interpolation method ignores the correlation of indicators, resulting in large deviations in the interpolation results. It can effectively restore the real state when the data is missing and improve the credibility and accuracy of the completed data.

[0062] Based on the above embodiments, Figure 2 This is a flow chart of the FTTR network data quality optimization method provided in Example 2 of this application. Figure 2 Based on Figure 1 The corresponding preferred embodiment of the FTTR network data quality optimization method can be executed by a data quality optimization device connected to each terminal device for communication. The data quality optimization device can be, for example, a FTTR network data quality optimization device. Figure 2 The FTTR network data quality optimization method may include the following steps: S21, collecting various network performance data of the target terminal device, where the target terminal device is a terminal device under the FTTR architecture; S22. For any type of network performance data of the target terminal device, obtain multiple missing intervals of the network performance data and calculate the correlation between the network performance data and other network performance data; S23, for any missing interval of any type of network performance data, determine multiple target intervals of the missing interval from the network performance data, calculate the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data, and use the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; S24, interpolating the network performance data using a preset data interpolation algorithm; S25. For any missing interval, obtain the time interval between the reference interval of the missing interval and the missing interval; S26, correcting each interpolated data point in the missing interval based on the reference interval and the time interval of the missing interval to obtain a corrected missing interval; S27. Generate corrected network performance data based on each corrected missing interval.

[0063] It should be noted that as the sampling time increases, the surrounding environment of each network performance data is constantly changing and the network performance data is continuously affected by environmental parameters (such as temperature, mechanical stress, etc.). When the sampling time interval between the reference interval and the missing interval is large, the cumulative difference in environmental parameters will cause the direct correction to fail. For example: if the optical power data collected 12 hours ago is directly used to interpolate a missing interval in the early morning, the difference in the fiber core expansion coefficient caused by the temperature difference between day and night will cause the corrected optical power data to deviate slightly from the true value.

[0064] Based on this, this embodiment first obtains the time interval between the reference interval and the missing interval on the basis of the above embodiment, and then corrects the data after interpolation of the missing interval according to the time interval and the reference interval.

[0065] In the specific implementation process, the time interval between the reference interval of each missing interval and each missing interval is first obtained, and all time intervals are linearly normalized, that is, the time interval is mapped to the range of 0 to 1, where the time interval is the absolute value of the difference between the first sampling time in the missing time and the first sampling time in the reference interval.

[0066] Furthermore, for any missing interval, each interpolated data point in the missing interval is corrected based on the reference interval of the missing interval and the time interval between the missing interval and the reference interval of the missing interval to obtain a corrected missing interval, and corrected network performance data is generated based on each corrected missing interval.

[0067] Specifically, take the first Taking a data point as an example, the following formula (2) can be used to correct the data point:

[0068] In the formula, Indicates the missing interval The corrected value of the data point, Indicates the missing interval The interpolation result of data points is Indicates that the missing interval is within the reference interval and The value of the data point corresponding to the data point, Represents the time interval between the missing interval and the reference interval.

[0069] Based on the above embodiments, this exemplary embodiment considers the sampling time interval between the reference interval and the missing interval, and then corrects each data point after interpolation in the missing interval in combination with the time interval. This can improve the accuracy of the correction results of the network performance data while suppressing the cumulative errors caused by the influence of environmental parameters.

[0070] Based on the above embodiments, Figure 3 FIG. 1 is a structural block diagram of a FTTR network data quality optimization device according to an implementation mode of the present application, such as Figure 3 As shown, the FTTR network data quality optimization device 300 may include: a data acquisition module 310, a correlation degree construction module 320, a reference interval acquisition module 330 and an interpolation correction module 340, wherein: The data collection module 310 is used to collect various network performance data of the target terminal device, and the target terminal device is a terminal device under the FTTR architecture; The correlation building module 320 is used to obtain multiple missing intervals of any network performance data of the target terminal device and calculate the correlation between the network performance data and other network performance data; The reference interval acquisition module 330 is used to determine multiple target intervals of any missing interval of any type of network performance data from the network performance data, calculate the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data, and use the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; The interpolation correction module 340 is used to interpolate the network performance data using a preset data interpolation algorithm, and correct the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0071] In an exemplary embodiment, the reference interval acquisition module 330 can also be used to calculate the actual numerical difference between the missing interval and each target interval based on the network performance data; calculate the correlation index difference between the missing interval and each target interval based on other network performance data and the correlation between the network performance data and other network performance data; determine the influencing factor between the missing interval and each target interval based on the actual numerical difference and the correlation index difference.

[0072] In an exemplary embodiment, the reference interval acquisition module 330 can also be used to calculate the numerical difference between the missing time period and the target time period of other network performance data for any target interval, taking the time interval corresponding to the missing interval as the missing time period and the time interval corresponding to the target interval as the target time period; and determine the degree of difference in the correlation indicators between the missing interval and the target interval based on the numerical differences.

[0073] In an exemplary embodiment, the reference interval acquisition module 330 may also use the following formula (1) to calculate the degree of difference in the association index between the missing interval and each target interval:

[0074] In the formula, Indicates the missing interval and the The degree of difference in the correlation indicators between the target intervals, Indicates the number of network performance data. Indicates that in addition to the network performance data, The degree of correlation between the network performance data and the network performance data, Indicates that the missing time period is The network performance data corresponds to numerical values, Indicates that the target time period is The network performance data corresponds to numerical values, Indicated in The number of network performance data shared by the missing time period and the target time period.

[0075] In an exemplary embodiment, the interpolation correction module 340 may also be used to obtain, for any missing interval, a time interval between a reference interval of the missing interval and the missing interval.

[0076] In an exemplary embodiment, the interpolation correction module 340 may also use the following formula (2) to correct each interpolated data point in the missing interval:

[0077] In the formula, Indicates the missing interval The corrected value of the data point, Indicates the missing interval The interpolation result of data points is Indicates that the missing interval is within the reference interval and The value of the data point corresponding to the data point, The time interval between the reference interval representing the missing interval and the missing interval.

[0078] In an exemplary embodiment, each target interval of the missing interval in the correlation degree construction module 320 contains the same number of data as the missing interval.

[0079] Those skilled in the art should understand that the division of the various modules in the embodiment is only a division of logical functions, and can be fully or partially integrated into one or more actual carriers in actual application, and these modules can be all implemented in the form of software called by the processing unit, or all implemented in the form of hardware, or implemented in the form of a combination of software and hardware. It should be noted that each module in a FTTR network data quality optimization device in this embodiment corresponds one-to-one to each step in a FTTR network data quality optimization method in the aforementioned embodiment. Therefore, the specific implementation method of this embodiment can refer to the implementation method of the aforementioned FTTR network data quality optimization method, which will not be repeated here.

[0080] Based on the above embodiments, Figure 4 FIG. 1 is a schematic diagram of the structure of a FTTR network data quality optimization device according to an implementation mode of the present application, such as Figure 4 As shown, the electronic device may include: a processor (processor) 410, a communication interface (Communications Interface) 420, a memory (memory) 430 and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call the logic instructions in the memory 430 to execute a FTTR network data quality optimization method, which includes: collecting multiple network performance data of the target terminal device, the target terminal device is a terminal device under the FTTR architecture; for any network performance data of the target terminal device, obtaining multiple missing intervals of the network performance data and calculating the correlation between the network performance data and other network performance data; for any missing interval, determining multiple target intervals of the missing interval from the network performance data, calculating the impact factor between the missing interval and each target interval based on the network performance data, other network performance data and the correlation between the network performance data and other network performance data, and taking the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; using a preset data interpolation algorithm to interpolate the network performance data, and correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0081] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0082] On the basis of the above embodiments, on the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a FTTR network data quality optimization method provided by the above methods, the method including: collecting multiple network performance data of a target terminal device, the target terminal device is a terminal device under a FTTR architecture; for any network performance data of the target terminal device, obtaining multiple missing intervals of the network performance data and calculating the correlation between the network performance data and other network performance data; for any missing interval, determining multiple target intervals of the missing interval from the network performance data, calculating the impact factor between the missing interval and each target interval based on the network performance data, other network performance data and the correlation between the network performance data and other network performance data, and taking the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; using a preset data interpolation algorithm to interpolate the network performance data, and correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0083] On the basis of the above embodiments, on another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to execute a FTTR network data quality optimization method provided by the above methods, the method comprising: collecting a variety of network performance data of a target terminal device, the target terminal device being a terminal device under a FTTR architecture; for any network performance data of the target terminal device, obtaining a plurality of missing intervals of the network performance data and calculating the degree of correlation between the network performance data and other network performance data; for any missing interval, determining a plurality of target intervals of the missing interval from the network performance data, calculating an impact factor between the missing interval and each target interval based on the network performance data, other network performance data and the degree of correlation between the network performance data and other network performance data, and taking the target interval corresponding to the minimum value of the impact factor as a reference interval of the missing interval; interpolating the network performance data using a preset data interpolation algorithm, and correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

[0084] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for optimizing FTTR network data quality, characterized in that: include: Collecting various network performance data of a target terminal device, where the target terminal device is a terminal device under a FTTR architecture; For any type of network performance data of the target terminal device, obtain multiple missing intervals of the network performance data and calculate the correlation between the network performance data and other network performance data; For any missing interval, determine multiple target intervals of the missing interval from the network performance data, calculate the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the degree of correlation between the network performance data and other network performance data, and use the target interval corresponding to the minimum impact factor as the reference interval of the missing interval; The network performance data is interpolated using a preset data interpolation algorithm, and the interpolated result is corrected based on the reference interval of each missing interval to generate corrected network performance data.

2. The method according to claim 1, characterized in that The calculating the impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the correlation between the network performance data and other network performance data includes: Calculate the actual numerical difference between the missing interval and each target interval based on the network performance data; Calculate the difference degree of the correlation index between the missing interval and each target interval based on other network performance data and the correlation degree between the network performance data and other network performance data; The influence factor between the missing interval and each target interval is determined based on the difference degree between the actual value and the difference degree between the associated index.

3. The method according to claim 2, characterized in that The calculating the difference degree of the correlation index between the missing interval and each target interval based on other network performance data and the correlation degree between the network performance data and other network performance data includes: For any target interval, the time interval corresponding to the missing interval is taken as the missing time period, and the time interval corresponding to the target interval is taken as the target time period, and the numerical difference between the missing time period and the target time period of other network performance data is calculated; The degree of difference in the correlation index between the missing interval and the target interval is determined based on the degree of correlation between each numerical difference and the network performance data and other network performance data.

4. The method according to claim 3, characterized in that The calculating the difference degree of the correlation index between the missing interval and each target interval based on other network performance data and the correlation degree between the network performance data and other network performance data includes: The following formula (1) is used to calculate the difference degree of the correlation index between the missing interval and each target interval: In the formula, Indicates that the missing interval is The degree of difference in the correlation indicators between the target intervals, Indicates the number of network performance data. In addition to the network performance data, the degree of correlation between the network performance data and the network performance data, Indicates that the missing time period is The network performance data corresponds to numerical values, Indicates that the target time period is The network performance data corresponds to numerical values, Indicated in The number of network performance data shared by the missing time period and the target time period.

5. The method according to claim 1, characterized in that Before correcting the interpolated result based on the reference interval of each missing interval to generate the corrected network performance data, the method further includes: For any missing interval, obtaining the time interval between a reference interval of the missing interval and the missing interval; The step of correcting the interpolated result based on the reference interval of each missing interval to generate corrected network performance data comprises: For any missing interval, correct each interpolated data point in the missing interval based on the reference interval of the missing interval and the time interval to obtain a corrected missing interval; Corrected network performance data is generated based on each corrected missing interval.

6. The method according to claim 5, characterized in that The step of correcting each interpolated data point in the missing interval based on the reference interval of the missing interval and the time interval comprises: The following formula (2) is used to correct each data point after interpolation in the missing interval: In the formula, Indicates the missing interval The corrected value of the data point, Indicates the missing interval The interpolation result of data points is Indicates that the missing interval is within the reference interval and The value of the data point corresponding to the data point, The time interval between the reference interval representing the missing interval and the missing interval.

7. The method according to claim 1, characterized in that Each target interval of the missing interval has the same number of data as the missing interval.

8. A FTTR network data quality optimization device, characterized in that: include: A data collection module, used to collect various network performance data of a target terminal device, wherein the target terminal device is a terminal device under the FTTR architecture; A correlation degree building module, for obtaining, for any network performance data of a target terminal device, a plurality of missing intervals of the network performance data and calculating the correlation degree between the network performance data and other network performance data; A reference interval acquisition module, for determining, for any missing interval, a plurality of target intervals of the missing interval from the network performance data, calculating an impact factor between the missing interval and each target interval based on the network performance data, other network performance data, and the degree of correlation between the network performance data and other network performance data, and taking the target interval corresponding to the minimum impact factor as a reference interval for the missing interval; The interpolation correction module is used to interpolate the network performance data using a preset data interpolation algorithm, and to correct the interpolated result based on the reference interval of each missing interval to generate corrected network performance data.

9. A FTTR network data quality optimization device, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images