Abnormal data determination method and apparatus, storage medium, and device

Abstract

The application discloses a kind of determination method, device, storage medium and equipment of abnormal data, related to communication technical field, for determining the abnormal data caused by non-network reason.The method comprises: obtaining the candidate abnormal data of target service;The measurement MOS of candidate abnormal data is less than the first preset threshold value.According to the service parameter of candidate abnormal data, the reference MOS of candidate abnormal data is determined;Service parameter is used to indicate the quality of service of target service, and reference MOS is positively correlated with quality of service.Determine the target similarity between the measurement MOS of candidate abnormal data and the reference MOS of candidate abnormal data.In the case where target similarity is located in preset mutation interval, determine that candidate abnormal data is target abnormal data caused based on non-network reason.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method, apparatus, storage medium, and device for determining abnormal data. Background Technology

[0002] Currently, in order to improve and optimize instant messaging services in a targeted manner and enhance their quality, it is necessary to analyze and evaluate instant messaging service data. Obtaining instant messaging service quality typically involves using a mean opinion score (MOS) box, a sending test terminal, and a receiving test terminal. Specifically, the MOS box sends the raw instant messaging service data to the sending test terminal via an audio port. The sending test terminal then sends the received raw instant messaging service data to the receiving test terminal via the network. Subsequently, the receiving test terminal records the received raw instant messaging service data to obtain reference data and sends this reference data back to the MOS box. Finally, the MOS box compares the raw instant messaging service data with the reference data to obtain the MOS value, and then evaluates the instant messaging service quality based on the MOS value.

[0003] If issues such as a loose audio port, overheating of the test terminal, or malfunction of the MOS box occur, the instant messaging reference data received by the MOS box will be abnormal. However, the MOS box cannot determine whether the abnormal instant messaging reference data is caused by network issues or non-network issues. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and device for determining abnormal data, used to determine abnormal data caused by reasons other than network issues.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] Firstly, a method for determining anomalous data is provided. This method includes: acquiring candidate anomalous data for a target service; the measured MOS of the candidate anomalous data is less than a first preset threshold. Based on the service parameters of the candidate anomalous data, a reference MOS is determined for the candidate anomalous data; the service parameters are used to indicate the service quality of the target service, and the reference MOS is positively correlated with the service quality. A target similarity is determined between the measured MOS of the candidate anomalous data and the reference MOS of the candidate anomalous data. If the target similarity is within a preset abrupt change range, the candidate anomalous data is determined to be target anomalous data caused by non-network reasons.

[0007] This invention provides a method for determining abnormal data. The method first identifies target service data whose measured MOS (Mean Orientation Scale) is less than a first preset threshold as candidate abnormal data. This filters out abnormal data caused by both network and non-network reasons. Based on service parameters indicating the service quality of the target service, a reference MOS for the candidate abnormal data in the network environment corresponding to those service parameters is determined. Further, the target similarity between the measured MOS of the candidate abnormal data and its reference MOS is determined. If the similarity between the measured MOS of the candidate abnormal data and the reference MOS that should be measured in the corresponding network environment is within a preset abrupt change range, it indicates that the measured MOS of the candidate abnormal data has undergone a sudden change in the corresponding network environment, and the candidate abnormal data is determined to be target abnormal data caused by non-network reasons.

[0008] In one possible implementation, the service parameters include at least one of packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR). The aforementioned "determining the reference MOS of the candidate abnormal data based on the service parameters of the candidate abnormal data" includes: inputting the service parameters of the candidate abnormal data into a target fitting function and outputting the reference MOS of the candidate abnormal data. The target fitting function is obtained by fitting the service parameters and the measured MOS of the multiple sample data of the target service.

[0009] The target fitting function is obtained by fitting service parameters and measured MOS values ​​of multiple sample data points based on the target service. The service parameters of the candidate anomaly data are input into the target fitting function, and the reference MOS of the candidate anomaly data is output. This reference MOS of the candidate anomaly data is the standard MOS value of the candidate anomaly data in the corresponding network environment, thus improving the accuracy of the reference MOS.

[0010] In one possible implementation, the above determination method further includes: acquiring business parameters and measurement MOS of multiple sample data. The business parameters of each sample data are input into a preset initial fitting function to determine a reference MOS for each sample data, resulting in multiple reference MOS for the sample data. Based on the reference MOS and measurement MOS of the multiple sample data, the determination coefficient of the initial fitting function is determined. If the determination coefficient of the initial fitting function is greater than a second preset threshold, the initial fitting function is determined as the target fitting function. Thus, the reference MOS and measurement MOS of the multiple sample data are fitted. And if the determination coefficient of the initial fitting function is greater than the second preset threshold, the initial fitting function is determined as the target fitting function. This allows for the accurate determination of the target fitting function for the reference MOS.

[0011] In one possible implementation, the above determination method further includes: inputting the business parameters of each sample data point from multiple sample data sets into a target fitting function to determine a reference MOS for each sample data point; determining the difference between the reference MOS and the measured MOS of each sample data point to obtain multiple differences; each difference corresponds to one sample data point; and determining a preset mutation interval based on the distribution of the differences. Thus, a preset mutation interval is obtained.

[0012] In one possible implementation, the above determination method further includes: determining the reference MOS of the target anomaly data as the measurement MOS of the target anomaly data. In this way, updating the measurement MOS of the target anomaly data caused by non-network reasons to the reference MOS corrects for changes in MOS caused by non-network reasons, thus improving the accuracy of the test results.

[0013] Secondly, an apparatus for determining anomalous data is provided. The apparatus includes an acquisition unit and a determination unit. The acquisition unit is used to acquire candidate anomalous data of a target service; the measured MOS of the candidate anomalous data is less than a first preset threshold. The determination unit is used to determine a reference MOS of the candidate anomalous data based on service parameters of the candidate anomalous data; the service parameters are used to indicate the service quality of the target service, and the reference MOS is positively correlated with the service quality. The determination unit is also used to determine the target similarity between the measured MOS of the candidate anomalous data and the reference MOS of the candidate anomalous data. The determination unit is further used to determine that the candidate anomalous data is target anomalous data caused by non-network reasons when the target similarity is within a preset abrupt change range.

[0014] In one possible implementation, the service parameters include at least one of packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR). The determining unit is specifically used to: input the service parameters of the candidate abnormal data into the target fitting function and output the reference MOS of the candidate abnormal data; the target fitting function is obtained by fitting the service parameters and the measured MOS of the multiple sample data of the target service.

[0015] In one possible implementation, the acquisition unit is further configured to acquire the business parameters and the measurement MOS of multiple sample data. The determination unit is further configured to input the business parameters of each sample data into a preset initial fitting function, determine the reference MOS of each sample data, and obtain the reference MOS of multiple sample data. The determination unit is further configured to determine the determination coefficient of the initial fitting function based on the reference MOS and the measurement MOS of the multiple sample data. The determination unit is further configured to determine the initial fitting function as the target fitting function in response to the determination coefficient of the initial fitting function being greater than a second preset threshold.

[0016] In one possible implementation, the determining unit is further configured to: input the business parameters of each sample data in multiple sample data into the target fitting function to determine the reference MOS of each sample data; determine the difference between the reference MOS of each sample data and the measured MOS of each sample data to obtain multiple differences; one difference corresponds to one sample data; and determine a preset mutation interval based on the distribution of the differences.

[0017] In one possible implementation, the determining unit is further configured to: determine the reference MOS of the target anomaly data as the measurement MOS of the target anomaly data.

[0018] Thirdly, an electronic device is provided, comprising: a processor, a memory, and a communication interface; wherein the communication interface is used for communication between a communication device and other devices or networks; the memory is used to store one or more programs, the one or more programs including computer-executable instructions, wherein when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the method for determining abnormal data as described in the first aspect.

[0019] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the instructions are executed, the method for determining abnormal data as described in the first aspect is implemented. Attached Figure Description

[0020] Figure 1 A schematic diagram of a speech evaluation method provided for an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of an active measurement MOS provided for an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of an abnormal data determination system provided in an embodiment of the present invention;

[0023] Figure 4 This is one of the flowcharts illustrating a method for determining abnormal data according to an embodiment of the present invention;

[0024] Figure 5 This is one of the schematic diagrams provided in an embodiment of the present invention for obtaining abnormal data;

[0025] Figure 6 This is a second schematic diagram illustrating the acquisition of abnormal data according to an embodiment of the present invention;

[0026] Figure 7 A second flowchart illustrating a method for determining abnormal data provided in an embodiment of the present invention;

[0027] Figure 8A third flowchart illustrating a method for determining abnormal data provided in an embodiment of the present invention;

[0028] Figure 9 This is a schematic diagram of the structure of an abnormal data determination device provided in an embodiment of the present invention;

[0029] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0030] The technical solutions in some embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments provided in this disclosure are within the scope of protection of this disclosure.

[0031] Unless the context otherwise requires, throughout the specification and claims, the term "comprise" and its other forms, such as the third-person singular "comprises" and the present participle "comprising," are interpreted as open-ended and encompassing, meaning "including, but not limited to." In the description of the specification, terms such as "one embodiment," "some embodiments," "exemplary embodiments," "example," "specific example," or "some examples," etc., are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this disclosure. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics mentioned may be included in any suitable manner in any one or more embodiments or examples.

[0032] Unless the context otherwise requires, throughout the specification and claims, the term "comprising" and its other forms, such as the third-person singular "comprising" and the present participle "comprising," are interpreted as open-ended and inclusive, meaning "including, but not limited to." That is, words such as "comprising" or "including" indicate that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. In the description of the specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this disclosure. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics mentioned may be included in any suitable manner in any one or more embodiments or examples.

[0033] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more.

[0034] In the description of this invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone.

[0035] As used herein, depending on the context, the term “if” may optionally be interpreted as meaning “when”, “in the event of”, “in response to determination”, or “in response to detection”. Similarly, depending on the context, the phrase “if it is determined that…” or “if [the stated condition or event] is detected” may optionally be interpreted as meaning “in the event of determination that…”, “in response to determination that…”, “when [the stated condition or event] is detected”, or “in response to the detection of [the stated condition or event]”.

[0036] The use of “applies to” or “configured to” in this article implies an open and inclusive language that does not preclude applicability to or configuration to devices that perform additional tasks or steps.

[0037] In addition, the use of “based on” implies openness and inclusivity, because processes, steps, calculations or other actions “based on” one or more of the stated conditions or values ​​may in practice be based on additional conditions or values ​​beyond those stated.

[0038] The inventive concept of this invention is described below:

[0039] Traditional narrowband speech audio ranges from 300Hz to 3.4kHz, and the perceptual evaluation of speech quality (PESQ) algorithm is typically used to assess call quality. This algorithm compares the original speech data with reference speech data and calculates the difference. It comprehensively considers the impact of factors such as coding rate, coding transmission, codec matching under multi-speed coding transmission, and noise during the comparison to objectively evaluate the quality of the speech signal. However, PESQ is only suitable for narrowband network communication.

[0040] For 4th generation (4G) and 5th generation (5G) mobile communication systems, voice services are carried by Internet Protocol (IP) based on the IP Multimedia Subsystem (IMS) architecture. The radio access technology for 4G mobile communication networks is Long Term Evolution (LTE), and the voice call service carried on it is called Voice on LTE (VoLTE). The radio access technology for 5G mobile communication networks, as shown above, is New Radio (NR), and the voice call service carried on it is called Voice on NR (VoNR).

[0041] VoLTE / VoNR supports adaptive multi-rate-wideband (AMR-WB) encoding and decoding from 50Hz to 7kHz, providing a wider audio range. The perceptual objective listening quality assessment (POLQA) algorithm is recommended. This method is designed for wideband voice. The reference signal and degraded signal undergo filtering, time alignment, sampling rate estimation, objective perception, and scoring to obtain a POLQA score, which is then mapped to the MOS (Mean Orientation). The reference signal is called the corpus, and typically has three sampling rates: 8k, 16k, and 48k. Considering the different selection of corpora for the test services, the resulting MOS will vary. Table 1 shows the MOS algorithm and corpus configuration for each voice service. Unless otherwise specified, all tests involved in this invention follow the configuration in the table below.

[0042] Table 1 shows the MOS algorithm and corpus configuration for each voice service evaluation.

[0043]

[0044] 2G refers to the second-generation (2G) mobile communication system, and 3G refers to the third-generation (3G) mobile communication system. HR stands for half-rate, FR for full-rate, ERF for enhanced full-rate, AMR-NB for Adaptive Multi-Rate narrow-band, and EVS for enhanced voice services.

[0045] Currently, end-to-end perception of instant messaging services (such as voice or video services) typically employs the MOS scoring method. This embodiment of the invention uses voice services as an example for illustration. Figure 1 As shown, speech quality evaluation methods are divided into subjective evaluation and objective evaluation. Objective evaluation includes active mode and passive mode.

[0046] Subjective evaluation involves human intervention to perceive speech quality, which is then scored to obtain the Mean Square of Opinions (MOS), thus completing the subjective evaluation. The correspondence between MOS and user perception is shown in Table 2.

[0047] Table 2 shows the correspondence between MOS and user experience.

[0048]

[0049] However, subjective evaluation requires a significant amount of manpower and time, and the scoring is inherently arbitrary. Therefore, subjective evaluation is rarely used.

[0050] Objective evaluation can be categorized into active and passive modes. For example, ... Figure 2 As shown, the active mode typically employs a MOS box, a transmitting test terminal, and a receiving test terminal. The MOS box sends the raw voice data to the transmitting test terminal via its audio port. The transmitting test terminal then sends the received raw voice data to the receiving test terminal via the network. Subsequently, the receiving test terminal records the received raw voice data to obtain reference voice data and sends the reference voice data to the MOS box. Finally, the MOS box compares the raw voice data and the reference voice data to obtain the MOS value, and then evaluates the voice quality based on the MOS value. Typically, the algorithm for comparing the raw voice data and the reference voice data can be perceptual speed quality measure (PSQM), PSQM+, perceptual analysis measurement system (PAMS), PESQ, or POLQA.

[0051] Passive mode converts the delay, jitter, and packet loss rate of the actual speech output signal into MOS (Mean of Speech Quality), and evaluates speech quality based on MOS. E-Model is typically used.

[0052] As can be seen, the active mode obtains the MOS value by comparing the original speech data and the reference speech data, and then evaluates the speech quality based on the MOS value. Compared with the passive mode, which converts the delay, jitter, and packet loss rate of the actual speech output signal into MOS, the MOS-based evaluation of speech quality is more accurate. Furthermore, the MOS converted in the passive mode takes into account wireless environmental factors, such as reference signal receiving power (RSRP) and signal to interference plus noise ratio (SINR). Therefore, the MOS converted in the passive mode cannot accurately represent the speech quality in a wireless network environment. Therefore, the active mode is currently commonly used to evaluate speech quality.

[0053] However, if issues such as a loose audio port, overheating of the test terminal, or malfunction of the MOS box occur, the instant messaging service reference data received by the MOS box will be abnormal. This makes it impossible for measurement personnel in active mode to determine whether the abnormal instant messaging service reference data is caused by network issues or non-network-related problems.

[0054] To address the aforementioned problems, this invention provides a method for determining abnormal data. The method first identifies target service data whose measured MOS (Mean Orientation Parameter) is less than a first preset threshold as candidate abnormal data. This filters out abnormal data caused by both network and non-network reasons. Based on service parameters indicating the service quality of the target service, a reference MOS for the candidate abnormal data in the network environment corresponding to those service parameters is determined. Further, the target similarity between the measured MOS of the candidate abnormal data and its reference MOS is determined. If the similarity between the measured MOS of the candidate abnormal data and the reference MOS that should be measured in the corresponding network environment is within a preset abrupt change range, it indicates that the measured MOS of the candidate abnormal data has undergone a sudden change, and the candidate abnormal data is determined to be target abnormal data caused by non-network reasons.

[0055] Based on the above-described inventive concept, the method for determining abnormal data provided in the embodiments of the present invention can be applied to an abnormal data determination system. Figure 3 A schematic diagram of a system for identifying this abnormal data is shown. Figure 3 As shown, the abnormal data determination system 10 can identify candidate abnormal data as target abnormal data caused by reasons other than network issues. The abnormal data determination system 10 includes an electronic device 11, a transmitting terminal device 12, and a receiving terminal device 13. The electronic device 11 and the transmitting terminal device 12 can be connected via a wired connection. The electronic device 11 and the receiving terminal device 13 can be connected via a wired connection. The transmitting terminal device 12 and the receiving terminal device 13 can be connected via a wireless connection.

[0056] Electronic device 11 can be a server, a MOS box, or other devices that implement the defined method of the present invention, and this application does not limit it. For example, electronic device 11 can be a mobile terminal, personal computer (PC), tablet computer, laptop computer, or other devices used by the user.

[0057] The system architecture and business scenarios described in the embodiments of this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0058] The method for determining abnormal data provided in the embodiments of the present invention is described below. The actions, terminology, etc., involved in the various embodiments of the present invention can be referred to mutually without limitation. The message names or parameter names in the messages exchanged between the various devices in the embodiments of the present invention are merely examples; other names may be used in specific implementations without limitation.

[0059] This invention provides a method for determining abnormal data, which can identify candidate abnormal data as target abnormal data caused by non-network reasons, such as... Figure 4 As shown, the method includes the following steps S201-S205.

[0060] S201. Electronic devices acquire candidate abnormal data for the target service.

[0061] Among them, the measured MOS of candidate abnormal data is less than the first preset threshold.

[0062] As one possible implementation, the electronic device acquires the measurement MOS of multiple raw data of the target service and determines whether the measurement MOS of each raw data is less than a first preset threshold. Further, if the measurement MOS is less than the first preset threshold, the electronic device determines the raw data corresponding to that measurement MOS as candidate abnormal data and acquires the candidate abnormal data.

[0063] For example, the electronic device acquires the measurement MOS of 10 raw data points and determines whether the measurement MOS of each raw data point is less than a first preset threshold. Further, the electronic device determines that two measurement MOS values ​​are less than the first preset threshold and identifies the raw data corresponding to these two measurement MOS values ​​as candidate abnormal data. Subsequently, the electronic device acquires these two candidate abnormal data points.

[0064] In some embodiments, the electronic device determines that the raw data corresponding to the measured MOS is candidate anomalous data, and acquires a portion of the candidate anomalous data. Furthermore, the electronic device saves the portion of the candidate anomalous data.

[0065] For example, such as Figure 5 As shown, the electronic device acquires three candidate anomaly data points and then acquires the data for the first six seconds of each of these three candidate anomaly data points. Furthermore, the electronic device saves the first six seconds of data for each of the three candidate anomaly data points. If the candidate anomaly data is less than six seconds long, the electronic device acquires and saves the entire candidate anomaly data set. Combined with... Figure 6 Candidate anomaly data A has only 3 seconds of data. At this point, the electronic device saves the entire candidate anomaly data A.

[0066] For example, the first preset threshold is 2.5.

[0067] It should be noted that the first preset threshold is stored in the electronic device in advance by the maintenance personnel.

[0068] S202. The electronic equipment determines the reference MOS of the candidate abnormal data based on the service parameters of the candidate abnormal data.

[0069] Among them, the business parameters are used to indicate the business quality of the target business, and the reference MOS is positively correlated with the business quality.

[0070] As one possible approach, the electronic device acquires the service parameters of the candidate abnormal data and determines the reference MOS of the candidate abnormal data based on the service parameters of the candidate abnormal data.

[0071] The specific implementation of this step can be found in the subsequent description of the embodiments of the present invention, and will not be repeated here.

[0072] Understandably, after acquiring candidate anomaly data for the target service, the electronic device determines the candidate anomaly data based on the reference MOS in the network environment corresponding to the service parameters.

[0073] S203. The electronic device determines the target similarity between the measurement MOS of the candidate abnormal data and the reference MOS of the candidate abnormal data.

[0074] One possible approach is for the electronic device to determine the target similarity between the measured MOS and the reference MOS of the candidate anomaly data by using the difference between the measured MOS and the reference MOS of the candidate anomaly data.

[0075] Specifically, the electronic device acquires the measurement MOS and reference MOS of the candidate abnormal data, and inputs the measurement MOS and reference MOS of the candidate abnormal data into the first similarity formula to obtain the target similarity between the measurement MOS and reference MOS of the candidate abnormal data.

[0076] The above-mentioned first similarity formula can be expressed as shown in Formula 1 below:

[0077]

[0078] Where d is the target similarity and y is the measured MOS. For reference MOS.

[0079] Understandably, electronic devices determine the degree of deviation between the measured MOS of candidate anomalous data and the MOS referenced for that candidate anomalous data in the corresponding network environment.

[0080] S204. The electronic device determines whether the target similarity is within a preset mutation range.

[0081] Specifically, the electronic device determines whether the target similarity d is located in the abnormal range of sudden increase or sudden decrease.

[0082] Understandably, electronic devices determine whether the deviation between the measured MOS of candidate abnormal data and the reference MOS of the candidate abnormal data in the corresponding network environment exceeds the normal range.

[0083] S205. When the similarity of the target is within a preset mutation range, the electronic device determines the candidate abnormal data as target abnormal data caused by non-network reasons.

[0084] Specifically, the electronic device determines that the target similarity d is within the abrupt increase range [d1:d1>μ+n*σ] or the abrupt decrease range [d1:d1<μ-n*σ], thus identifying the candidate anomalous data as target anomalous data caused by non-network reasons. Here, d1 is the abrupt increase similarity, and n is a value in n-sigma, which can be 1, 2, or 3. Understandably, if the electronic device determines that the measured MOS of the candidate anomalous data deviates significantly from the reference MOS of the candidate anomalous data in the corresponding network environment, it determines that the measured MOS of the candidate anomalous data being less than a first preset threshold is due to non-network reasons. Therefore, the electronic device determines that the candidate anomalous data is target anomalous data caused by non-network reasons.

[0085] This invention provides a method for determining abnormal data. The method first identifies target service data whose measured MOS (Mean Orientation Parameter) is less than a first preset threshold as candidate abnormal data. This filters out abnormal data caused by both network and non-network reasons. Based on service parameters indicating the service quality of the target service, a reference MOS for the candidate abnormal data in the network environment corresponding to those service parameters is determined. Further, the target similarity between the measured MOS of the candidate abnormal data and its reference MOS is determined. If the similarity between the measured MOS of the candidate abnormal data and the reference MOS that should be measured in the corresponding network environment is within a preset abrupt change range, it indicates that the measured MOS of the candidate abnormal data has undergone a sudden change, and the candidate abnormal data is determined to be target abnormal data caused by non-network reasons.

[0086] In one design, in order to determine the reference MOS, the service parameters include at least one of packet loss rate, jitter, delay, reference signal received power RSRP, and signal-to-interference-plus-noise ratio (SINR). S202 provided in the embodiments of the present invention specifically includes the following S2021.

[0087] S2021. The electronic device inputs the service parameters of the candidate abnormal data into the target fitting function and outputs the reference MOS of the candidate abnormal data.

[0088] The target fitting function is obtained by fitting the business parameters and the measurement MOS of multiple sample data of the target business.

[0089] As one possible approach, the electronic device acquires the packet loss rate, jitter, latency, RSRP, and SINR of the candidate abnormal data, inputs these parameters into a target fitting function, and outputs a reference MOS for the candidate abnormal data.

[0090] In some embodiments, the electronic device acquires the packet loss rate, jitter, latency, and RSRP of the candidate abnormal data, and inputs the packet loss rate, jitter, latency, and RSRP of the candidate abnormal data into the target fitting function, and outputs the reference MOS of the candidate abnormal data.

[0091] It should be noted that this invention does not impose specific limitations on service parameters. Electronic devices can input one of the following service parameters—packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR)—to a target fitting function; or they can input multiple of these parameters to a target fitting function; or they can input other parameters indicating the service quality of the target service to the target fitting function.

[0092] In one design, to determine the target fitting function, such as Figure 7 As shown in the figure, the method for determining abnormal data provided in this embodiment of the invention further includes the following steps S206-S2010.

[0093] S206. Business parameters and measurement MOS of multiple sample data acquired by electronic equipment.

[0094] One possible approach is for the electronic device to acquire multiple sample data points and obtain the corresponding business parameters and measurement MOS for each sample data point.

[0095] Specifically, the electronic device tests multiple samples of raw speech data under stable laboratory or testing conditions to obtain multiple sample recorded speech data. Furthermore, the electronic device acquires the speech packet loss rate, speech jitter, speech delay, RSRP, SINR, and measures MOS of the multiple sample recorded speech data.

[0096] It should be noted that the voice packet loss rate is the uplink packet loss rate, which is calculated by the base station based on the sequence numbers (SNs) of the Packet Data Convergence Protocol (PDCP) packets sent by the test terminal. If there are discontinuous SNs, the base station considers packet loss to have occurred. RSRP and SINR represent the environmental quality of the wireless network where the test terminal is located. For 4G test terminals, the RSRP and SINR values ​​measured by the test terminal can be collected; for 5G terminals, the synchronization signal reference signal receiving power (SS-RSRP) and synchronization signal interference plus noise ratio (SS-SINR) of the test terminal can be collected.

[0097] S207. The electronic device inputs the business parameters of each sample data into a preset initial fitting function to determine the reference MOS of each sample data, and obtains the reference MOS of multiple sample data.

[0098] One possible approach is for the electronic device to perform trend line fitting using regression equations on the service parameters and measured MOS of multiple sample data points, obtaining an initial fitting function between the reference MOS and the service parameters of the sample data. Further, the electronic device inputs the service parameters of each sample data point into the initial fitting function to determine the reference MOS for each sample data point, thus obtaining the reference MOS for multiple sample data points.

[0099] In some embodiments, the electronic device uses a regression equation to fit the voice packet loss rate, voice jitter, voice delay, RSRP, SINR, and measured MOS to obtain a first initial fitting function for the reference MOS of the sample data and the service parameters of the sample data.

[0100] The first initial fitting function mentioned above can be represented by the following formula 2:

[0101] f(mos) = f1 (packet loss rate) + f2 (jitter) + f3 (delay) + f4 (RSRP) + f5 (SINR) Formula 2

[0102] Where f(mos) is a function associated with the measured MOS, f1(packet loss rate) is a function associated with the packet loss rate, f2(jitter) is a function associated with jitter, f3(delay) is a function associated with delay, f4(RSRP) is a function associated with RSRP, and f5(SINR) is a function associated with SINR.

[0103] In other embodiments, the electronic device uses a regression equation to fit the voice packet loss rate, voice jitter, voice delay, SINR, and measured MOS to a trend line, thereby obtaining a second initial fitting function for the reference MOS of the sample data and the service parameters of the sample data.

[0104] The second initial fitting function mentioned above can be represented by the following formula 3:

[0105] f(mos) = f1 (packet loss rate) + f2 (jitter) + f3 (delay) + f4 (SINR) Formula 3

[0106] Where f(mos) is a function associated with the measured MOS, f1(packet loss rate) is a function associated with the packet loss rate, f2(jitter) is a function associated with jitter, f3(delay) is a function associated with the delay, and f4(SINR) is a function associated with SINR.

[0107] S208. The electronic device determines the determination coefficients of the initial fitting function based on the reference MOS of multiple sample data and the measurement MOS of multiple sample data.

[0108] Specifically, the electronic device inputs the reference MOS of multiple sample data and the measurement MOS of multiple sample data into the determination function to determine the determination coefficient of the initial fitting function.

[0109] The above-mentioned determination function can be represented by the following formula four:

[0110]

[0111] Where R is the coefficient of determination, SS RES Sum of squared residuals, SS tot For the total sum of squares, SS reg For regression sum of squares, n is the total number of sample data, y i For the measurement MOS of the i-th sample data, is the reference MOS for the i-th sample data.

[0112] S209. The electronic device determines whether the coefficient of determination of the initial fitting function is greater than the second preset threshold.

[0113] S2010. In response to the determination coefficient of the initial fitting function being greater than the second preset threshold, the electronic device determines the initial fitting function as the target fitting function.

[0114] As one possible approach, the electronic device determines the initial fitting function as the target fitting function after determining that the coefficient of determination of the initial fitting function is greater than a second preset threshold.

[0115] For example, the second preset threshold is 0.8.

[0116] It should be noted that the second preset threshold is stored in the electronic device in advance by the maintenance personnel.

[0117] In another scenario, after determining that the coefficient of determination of the initial fitting function is less than or equal to a second preset threshold, the electronic device acquires more sample recorded speech data to correct the initial fitting function.

[0118] For example, the electronic device initially acquires 1000 sample data points. It then uses a regression equation to fit a trend line to the service parameters and measured MOS of these 1000 sample data points, obtaining an initial fitting function. Further, the electronic device determines the determination coefficient of the initial fitting function based on the reference MOS and measured MOS of the 1000 sample data points. If the determination coefficient of the initial fitting function is less than or equal to a second preset threshold, the electronic device acquires another 1000 sample data points. This process is repeated, increasing the number of sample data points, until the determination coefficient of the initial fitting function is greater than the second preset threshold.

[0119] In one design, to determine the mutation region, such as Figure 8 As shown, the method for determining abnormal data provided in this embodiment of the invention further includes the following steps S2011-S2013.

[0120] S2011. The electronic device inputs the business parameters of each sample data in multiple sample data into the target fitting function to determine the reference MOS of each sample data.

[0121] In some embodiments, the electronic device inputs the speech packet loss rate, speech jitter, speech delay, RSRP, and SINR of each sample data into the first initial fitting function to determine the reference MOS of each sample data.

[0122] In other embodiments, the electronic device inputs the speech packet loss rate, speech jitter, speech delay, and SINR of each sample data into the second initial fitting function to determine the reference MOS of each sample data.

[0123] S2012. The electronic device determines the difference between the reference MOS and the measurement MOS of each sample data, and obtains multiple differences.

[0124] Each difference corresponds to one sample data.

[0125] S2013. The electronic device determines the preset mutation interval based on the distribution of the difference.

[0126] One possible approach is for the electronic device to determine the similarity of each sample data point by comparing the reference MOS with the measured MOS of each sample data point. This results in a similarity score for multiple sample data points. Furthermore, based on the similarity scores of the multiple sample data points, the electronic device determines a confidence interval for the measured MOS of the multiple sample data points, and then determines abrupt change intervals based on these confidence intervals.

[0127] Specifically, the electronic device inputs the reference MOS and the measured MOS of each sample data into the second similarity formula to determine the similarity of the measured MOS of each sample data.

[0128] The second similarity formula mentioned above can be expressed as shown in Formula 5 below:

[0129]

[0130] Where, d i Let y be the similarity of the measured MOS for the i-th sample data. i For the measurement MOS of the i-th sample data, is the reference MOS for the i-th sample data.

[0131] Furthermore, the electronic device inputs the similarity of the measured MOS of multiple sample data into the mean function to obtain the mean of the measured MOS of multiple sample data.

[0132] The above mean function can be represented by the following formula six:

[0133]

[0134] Where μ is the mean of the measured MOS of multiple sample data, and d i Let be the similarity of the measured MOS for the i-th sample data, and n be the total number of sample data.

[0135] Subsequently, the electronic device inputs the similarity of the measured MOS of multiple sample data and the mean of multiple sample data into the standard deviation function to obtain the standard deviation of the measured MOS of multiple sample data.

[0136] The standard deviation function described above can be represented by the following formula (Equation 7):

[0137]

[0138] Where σ is the standard deviation of the measured MOS of multiple sample data, μ is the mean of the measured MOS of multiple sample data, and d i Let be the similarity of the measured MOS for the i-th sample data, and n be the total number of sample data.

[0139] Finally, the electronic device determines the confidence interval of the measured MOS of multiple sample data based on the standard deviation and mean of the measured MOS of multiple sample data, and determines the mutation interval of the measured MOS of multiple sample data based on the confidence interval of the measured MOS of multiple sample data.

[0140] Specifically, the electronic device uses n-sigma to determine the confidence interval of the measured MOS of multiple sample data. This embodiment of the invention uses 3-sigma as an example. Based on the 3-sigma algorithm, the confidence interval of the measured MOS of multiple sample data is (μ-3σ, μ+3σ). Further, the electronic device determines the anomalous interval of the measured MOS of multiple sample data [d: d > μ+3σ] and the anomalous interval of the measured MOS of multiple sample data [d: d > μ-3σ], and uses the anomalous intervals of ...

[0141] It should be noted that the present invention can also determine the abrupt change range of the measured MOS of multiple sample data using sigma or 2-sigma.

[0142] Currently, abnormal data measurement MOS caused by non-network reasons leads to test results that cannot accurately reflect the perceived status of voice services in the network. Therefore, in order to improve the accuracy of test results, an embodiment of the present invention provides a method for determining abnormal data, which further includes the following step S2014.

[0143] S2014. The electronic device determines the reference MOS of the target abnormal data as the measurement MOS of the target abnormal data.

[0144] One possible approach is for the electronic device to acquire a reference MOS for the target anomaly data and replace the measurement MOS for the target anomaly data with the reference MOS for the target anomaly data to update the measurement MOS for the target anomaly data.

[0145] Understandably, when it is determined that the target anomaly data is not caused by network issues, the measured MOS determined by the target anomaly data can no longer accurately reflect the test results of voice services in the network. To improve the accuracy of the test results, this invention determines the reference MOS that the voice service data should be measured under this network environment based on network environment parameters and a fitting function, and replaces the measurement MOS with the larger error with this reference MOS, thereby improving the accuracy of the test results.

[0146] In one scenario, the electronic device deletes the Measurement MOS (Measuring Object) for abnormal target data. This way, the electronic device removes measurement MOS that cannot accurately reflect the test results of voice services in the network, retaining only those that truly reflect the test results of voice services in the network, thereby improving the accuracy of the test results.

[0147] In another scenario, the electronic device retests the original data corresponding to the target anomaly data and obtains updated sample data. Furthermore, the electronic device compares the updated sample data with the original data to obtain the updated MOS, and replaces the measured MOS of the target anomaly data with the updated MOS.

[0148] In this way, by calibrating the MOS of the target abnormal data, the calibrated MOS is obtained, thereby improving the accuracy of the test results.

[0149] The foregoing mainly describes the solutions provided by the embodiments of the present invention from a methodological perspective. To achieve the above functions, the electronic device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of the present invention.

[0150] In this embodiment of the invention, the device can be divided into functional modules according to the above method example. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated modules can be implemented in hardware or as software functional modules. Optionally, the module division in this embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0151] Figure 9 This is a schematic diagram of a device for determining abnormal data provided in an embodiment of the present invention. Figure 9 As shown, the determining device 30 may be located in the aforementioned electronic device. The determining device 30 includes an acquisition unit 301 and a determining unit 302.

[0152] Acquisition unit 301 is used to acquire candidate anomaly data for the target service; the measured MOS of the candidate anomaly data is less than a first preset threshold. For example, combined with Figure 4 The acquisition unit 301 can be used to execute S201.

[0153] The determining unit 302 is used to determine the reference MOS of the candidate abnormal data based on the service parameters of the candidate abnormal data; the service parameters are used to indicate the service quality of the target service, and the reference MOS is positively correlated with the service quality. For example, combined with... Figure 4 Unit 302 can be used to execute S202.

[0154] The determining unit 302 is also used to determine the target similarity between the measured MOS of the candidate anomalous data and the reference MOS of the candidate anomalous data. For example, combined with Figure 4 Unit 302 can be used to execute S203.

[0155] The determining unit 302 is also used to determine, when the target similarity is within a preset mutation range, the candidate abnormal data as target abnormal data caused by non-network reasons. For example, combined with Figure 4 Unit 302 can be used to execute S205.

[0156] Optionally, the service parameters include at least one of packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR). The determination unit 302 is specifically used to: input the service parameters of the candidate abnormal data into the target fitting function and output the reference MOS of the candidate abnormal data; the target fitting function is obtained by fitting the service parameters and the measured MOS of the multiple sample data of the target service.

[0157] Optionally, the acquisition unit 301 is also used to acquire the service parameters of multiple sample data and the measurement MOS of the multiple sample data. For example, combined with Figure 7 The acquisition unit 301 can be used to execute S206.

[0158] The determining unit 302 is also used to input the business parameters of each sample data into a preset initial fitting function, determine the reference MOS of each sample data, and obtain the reference MOS of multiple sample data. For example, combined with Figure 7 Unit 302 can be used to execute S207.

[0159] The determining unit 302 is also used to determine the determination coefficients of the initial fitting function based on the reference MOS of multiple sample data and the measured MOS of multiple sample data. For example, combined with Figure 7 Unit 302 can be used to execute S208.

[0160] The determining unit 302 is further configured to determine the initial fitting function as the target fitting function in response to the initial fitting function having a determination coefficient greater than a second preset threshold. For example, combined with Figure 7 Unit 302 can be used to execute S2010.

[0161] Optionally, the determining unit 302 is further configured to: input the business parameters of each sample data in multiple sample data into the target fitting function to determine the reference MOS of each sample data; determine the difference between the reference MOS of each sample data and the measured MOS of each sample data to obtain multiple differences; one difference corresponds to one sample data. Based on the distribution of the differences, determine a preset mutation interval. For example, combined with... Figure 8 Unit 302 can be used to execute S2011-S2013.

[0162] Optionally, the determining unit 302 is also used to: determine the reference MOS of the target abnormal data as the measurement MOS of the target abnormal data.

[0163] In the case of implementing the functions of the integrated modules described above in hardware, this embodiment of the invention provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 10 As shown, the electronic device 40 includes a processor 401, a memory 402, and a bus 403. The processor 401 and the memory 402 can be connected via the bus 403.

[0164] Processor 401 is the control center of the communication device. It can be a single processor or a collective term for multiple processing elements. For example, processor 401 can be a general-purpose central processing unit (CPU) or other general-purpose processors. Among them, the general-purpose processor can be a microprocessor or any conventional processor.

[0165] As one embodiment, processor 401 may include one or more CPUs, for example Figure 10 CPU 0 and CPU 1 are shown in the diagram.

[0166] The memory 402 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0167] In one possible implementation, the memory 402 can exist independently of the processor 401. The memory 402 can be connected to the processor 401 via a bus 403 and is used to store instructions or program code. When the processor 401 calls and executes the instructions or program code stored in the memory 402, it can implement the sensor determination method provided in this embodiment of the invention.

[0168] In another possible implementation, the memory 402 can also be integrated with the processor 401.

[0169] Bus 403 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0170] It should be pointed out that, Figure 10 The structure shown does not constitute a limitation on the electronic device 40. Except... Figure 10 In addition to the components shown, the electronic device 40 may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0171] As an example, combined Figure 9 The function implemented by the determining unit 302 in the determining device 30 is the same as Figure 10 The processor 401 in it has the same function.

[0172] Optionally, the electronic device 40 provided in this embodiment of the invention may further include a communication interface 404.

[0173] Communication interface 404 is used to connect with other devices via a communication network. This communication network can be Ethernet, a wireless access network, a wireless local area network (WLAN), etc. Communication interface 404 may include a receiving unit for receiving data and a transmitting unit for transmitting data.

[0174] In one design, the communication interface in the electronic device provided by the embodiments of the present invention can also be integrated into the processor.

[0175] In another hardware structure of the electronic device provided in this embodiment of the invention, the electronic device may include a processor and a communication interface. The processor is coupled to the communication interface.

[0176] The functions of the processor can be found in the processor description above. In addition, the processor also has storage functions, which can be found in the memory function description above.

[0177] The communication interface is used to provide data to the processor. This communication interface can be an internal interface of the communication device or an external interface of the communication device.

[0178] It should be noted that the above-mentioned alternative hardware structure does not constitute a limitation on the electronic device. In addition to the above-mentioned alternative hardware component, the electronic device may include more or fewer components, or combine certain components, or have different component arrangements.

[0179] Through the above description of the embodiments, those skilled in the art will clearly understand that, for the sake of convenience and brevity, only the division of the above functional units is used as an example. In practical applications, the above functions can be assigned to different functional units as needed, that is, the internal structure of the device can be divided into different functional units to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0180] This invention also provides a computer-readable storage medium storing instructions. When a computer executes these instructions, the computer performs each step of the abnormal data determination method flow shown in the above-described method embodiments.

[0181] This invention also provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the method for determining abnormal data described in the above method embodiments.

[0182] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing, or any other form of computer-readable storage medium in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In embodiments of the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0183] Since the electronic devices, computer-readable storage media, and computer program products in the embodiments of the present invention can be applied to the above methods, the technical effects they can achieve can also be referred to the above method embodiments. The embodiments of the present invention will not be repeated here.

[0184] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for determining abnormal data, characterized in that, The method includes: Acquire candidate anomaly data for the target business; the measured mean subjective opinion score (MOS) of the candidate anomaly data is less than a first preset threshold. Based on the service parameters of the candidate anomaly data, a reference MOS is determined for the candidate anomaly data; the service parameters are used to indicate the service quality of the target service, and the reference MOS is positively correlated with the service quality; Determine the target similarity between the measured MOS of the candidate anomaly data and the reference MOS of the candidate anomaly data; If the target similarity is within a preset mutation range, the candidate abnormal data is determined to be target abnormal data caused by non-network reasons; The service parameters include at least one of packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR). Determining the reference MOS of the candidate abnormal data based on the service parameters of the candidate abnormal data includes: The service parameters of the candidate anomaly data are input into the target fitting function, and the reference MOS of the candidate anomaly data is output; the target fitting function is obtained by fitting the service parameters and the measured MOS of the multiple sample data of the target service.

2. The determination method according to claim 1, characterized in that, The method further includes: Obtain the business parameters and the measured MOS of the multiple sample data; The business parameters of each sample data are input into a preset initial fitting function to determine the reference MOS of each sample data, thereby obtaining the reference MOS of the multiple sample data. The determination coefficients of the initial fitting function are determined based on the reference MOS and the measured MOS of the plurality of sample data. In response to the initial fitting function having a determination coefficient greater than a second preset threshold, the initial fitting function is determined to be the target fitting function.

3. The determination method according to claim 1, characterized in that, The method further includes: The business parameters of each sample data in the plurality of sample data are input into the target fitting function to determine the reference MOS of each sample data; The difference between the reference MOS and the measured MOS of each sample data is determined to obtain multiple differences; one difference corresponds to one sample data. The preset mutation interval is determined based on the distribution of the difference.

4. The determining method according to claim 2 or 3, characterized in that, The method further includes: The reference MOS of the target anomaly data is determined as the measurement MOS of the target anomaly data.

5. A device for determining abnormal data, characterized in that, The device includes: an acquisition unit and a determination unit; The acquisition unit is used to acquire candidate abnormal data of the target business; the measured average subjective opinion score (MOS) of the candidate abnormal data is less than a first preset threshold. The determining unit is configured to determine a reference MOS for the candidate abnormal data based on the service parameters of the candidate abnormal data; the service parameters are used to indicate the service quality of the target service, and the reference MOS is positively correlated with the service quality; The determining unit is further configured to determine the target similarity between the measured MOS of the candidate abnormal data and the reference MOS of the candidate abnormal data; The determining unit is further configured to determine, when the target similarity is within a preset mutation range, that the candidate abnormal data is target abnormal data caused by non-network reasons; The service parameters include at least one of packet loss rate, jitter, delay, reference signal received power (RSRP), and signal-to-interference-plus-noise ratio (SINR). The determining unit is specifically used for: The service parameters of the candidate anomaly data are input into the target fitting function, and the reference MOS of the candidate anomaly data is output; the target fitting function is obtained by fitting the service parameters and the measured MOS of the multiple sample data of the target service.

6. The determining device according to claim 5, characterized in that, The acquisition unit is further configured to acquire the service parameters of the plurality of sample data and the measurement MOS of the plurality of sample data; The determining unit is further configured to input the business parameters of each sample data into a preset initial fitting function, determine the reference MOS of each sample data, and obtain the reference MOS of the plurality of sample data; The determining unit is further configured to determine the determination coefficient of the initial fitting function based on the reference MOS of the plurality of sample data and the measured MOS of the plurality of sample data; The determining unit is further configured to determine the initial fitting function as the target fitting function in response to the initial fitting function having a determination coefficient greater than a second preset threshold.

7. The determining device according to claim 5, characterized in that, The determining unit is further configured to: The business parameters of each sample data in the plurality of sample data are input into the target fitting function to determine the reference MOS of each sample data; The difference between the reference MOS and the measured MOS of each sample data is determined to obtain multiple differences; one difference corresponds to one sample data. The preset mutation interval is determined based on the distribution of the difference.

8. The determining device according to claim 6 or 7, characterized in that, The determining unit is further configured to: determine the reference MOS of the target abnormal data as the measurement MOS of the target abnormal data.

9. An electronic device, characterized in that, include: Processor, memory, and communication interface; wherein the communication interface is used for communication between the communication device and other devices or networks; The memory is used to store one or more programs, which include computer-executable instructions. When the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the method for determining abnormal data as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, implement the method for determining abnormal data as described in any one of claims 1 to 4.

Images