A wireless interference detection method, system, electronic device and storage medium

Abstract

The application discloses a wireless interference detection method and system, electronic equipment and storage medium, and relates to the technical field of communication network detection. The method comprises the following steps: acquiring an uplink interference level time sequence of a target cell in a first detection window at a target time point, wherein the end point of the first detection window is located before the target time point; performing mutation point detection on the uplink interference level time sequence, and when the detected target mutation point corresponds to an upward mutation direction and a mutation amplitude greater than a first amplitude threshold, determining the date corresponding to the target mutation point as a candidate abnormal occurrence day; acquiring uplink interference level data in a verification period, and when the data of the consecutive preset number of days in the verification period all satisfy the continuous abnormal condition, determining that there is real interference and generating an alarm information. The application can effectively filter false alarms caused by temporary interference fluctuations, and through long-period trend analysis, the detection blind area of slow interference deterioration is filled, and precise and efficient automatic wireless interference sensing is realized.

Description

Technical Field

[0001] This invention relates to the field of communication network detection technology, and in particular to a wireless interference detection method, system, electronic device, and storage medium. Background Technology

[0002] With the large-scale deployment and commercial use of 4G and 5G mobile communication networks, wireless interference has become a core factor affecting network operation quality and reducing user communication experience. Interference sources include various types such as co-channel and adjacent-channel interference within the system, interference from external illegal devices, and environmental multipath interference. Network operators need to promptly detect and address various types of wireless interference to ensure stable network operation. However, the traditional model of passively discovering problems through manual inspections and user complaints suffers from slow response and low efficiency. Therefore, how to achieve proactive and automated detection of wireless interference has become a common problem urgently needing to be solved in the field of mobile communication operation and maintenance.

[0003] Currently, the industry commonly uses fixed threshold alarms for wireless interference monitoring. By setting a uniform threshold for uplink interference levels, an alarm is automatically triggered when the monitored indicators exceed the threshold. This method, to some extent, replaces manual periodic inspections and initially achieves proactive detection of interference problems. However, this single judgment mechanism has significant shortcomings. It not only fails to adapt to the service differences of different cells, easily generating false alarms and missed alarms, but also cannot effectively distinguish between temporary signal fluctuations and long-term persistent faults. It is completely unable to identify gradual interference degradation caused by loose antennas, aging components, etc. At the same time, the entire interference handling process is fragmented, lacking the end-to-end closed-loop capability of intelligent root cause analysis, automatic dispatching, and verification of governance effects, making it difficult to meet the high-precision, all-scenario, and automated interference governance needs of modern mobile communication networks. Summary of the Invention

[0004] This invention provides a wireless interference detection method, system, electronic device, and storage medium, which can solve the problems of high false alarm rate, inability to identify slow deterioration, and lack of continuous confirmation mechanism in existing interference detection methods, and achieve accurate interference perception by combining statistical detection and service confirmation.

[0005] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions: A first aspect provides a wireless interference detection method, the method comprising: acquiring an uplink interference level time series of a target cell within a first detection window at a target time, wherein the end point of the first detection window is located before the target time; performing abrupt change point detection on the uplink interference level time series, wherein when the direction of the abrupt change corresponding to the detected target abrupt change point is upward and the abrupt change amplitude is greater than a first amplitude threshold, determining the date corresponding to the target abrupt change point as a candidate anomaly occurrence date; acquiring uplink interference level data of the target cell within a verification period, wherein the verification period is from the candidate anomaly occurrence date to the end point of the first detection window; and determining that the target cell has real interference and generating alarm information when the uplink interference level data for a consecutive preset number of days within the verification period all meet the continuous anomaly condition.

[0006] The method provided by this invention sets the end point of the detection window before the target time, and completes mutation detection and persistence confirmation in one go based on the available historical data, so as to realize alarm output on the same day without waiting for future data. Through the series judgment mechanism of mutation detection and persistence confirmation, an alarm is triggered only when the statistical significance, amplitude threshold and persistence conditions are met at the same time, which effectively filters false alarms caused by short-term interference fluctuations and ensures that alarms are only generated for interference that is still ongoing as of the end point of the detection window.

[0007] In one possible implementation of the first aspect, the step of detecting abrupt change points in the uplink interference level time series includes: applying a Pettitt nonparametric test to the uplink interference level time series, calculating the U statistic corresponding to each segment point, and selecting the segment point with the largest absolute value of the U statistic as a candidate abrupt change point; calculating the significance p-value corresponding to the candidate abrupt change point, and determining the candidate abrupt change point as the target abrupt change point when the significance p-value is less than a first significance threshold; and calculating the difference between the mean of the latter segment and the mean of the former segment of the target abrupt change point as the abrupt change amplitude.

[0008] The method provided by this invention uses Pettt's nonparametric test to detect mutation points. It does not require assumptions about the data distribution pattern and is insensitive to outliers. It can accurately identify sudden abnormal changes in uplink interference levels. Through dual determination of significance p-value and amplitude threshold, it ensures that it only responds to mutations that are statistically significant and whose amplitude reaches the level of engineering concern.

[0009] In one possible implementation of the first aspect, the first significance threshold is 0.001, the first amplitude threshold is 10dB, and the first detection window is the 30 working days prior to the target time.

[0010] The method provided by this invention achieves a balance between sensitivity and specificity by setting the p-value threshold to 0.001 and the amplitude threshold to 10dB. It has been verified by a large number of cells in the current network. The amplitude threshold of 10dB exceeds the daily fluctuation range of normal uplink interference levels (usually within 5dB), which can effectively filter normal fluctuations and capture real interference rises.

[0011] In one possible implementation of the first aspect, determining that the target cell has real interference when the uplink interference level data for a consecutive preset number of days within the verification period all meet the continuous abnormality condition includes: calculating the average uplink interference level for a first preset number of days before the candidate abnormality occurrence date as the service baseline; calculating the length of the verification period, the length being the difference in the number of days between the end of the first detection window and the candidate abnormality occurrence date; when the length is greater than or equal to the preset number of days, extracting the uplink interference level data corresponding to the last preset number of days within the verification period; and determining that the target cell has real interference when the uplink interference level data for the last preset number of days all meet the continuous abnormality condition relative to the service baseline.

[0012] The method provided by this invention calculates the average value of several working days before the mutation point as the business baseline, and compares and verifies the data of several consecutive days at the end of the verification period with the baseline, ensuring the timeliness of the continuous judgment - only when the continuous data closest to the end of the detection window meets the continuous abnormality condition is the real interference confirmed, and if the interference has been recovered within the verification period, no alarm is generated.

[0013] In one possible implementation of the first aspect, the first preset number of days is 14 working days, and the preset number of days is 3 days; the continuous abnormal condition is that the uplink interference level is greater than or equal to the service baseline plus a second amplitude threshold, and the second amplitude threshold is 10dB.

[0014] The method provided by this invention covers different business models through a 14-working-day baseline calculation cycle, and the 3-day continuous verification window can effectively distinguish between continuous interference that requires operation and maintenance intervention and short-term fluctuations that can recover on their own. The 10dB second amplitude threshold is consistent with the amplitude threshold of mutation detection.

[0015] In one possible implementation of the first aspect, the method further includes: acquiring the uplink interference level time series of the target cell within a second detection window, wherein the length of the second detection window is greater than the length of the first detection window, and the end point of the second detection window is located before the target time; performing trend analysis on the uplink interference level time series within the second detection window, and determining that the target cell has real interference and generating alarm information when a statistically significant upward trend is detected, the deterioration rate exceeds a rate threshold, the cumulative deterioration exceeds a cumulative threshold, and the recent persistence verification is passed.

[0016] The method provided by this invention compensates for the deficiency of mutation detection in being insensitive to slow and gradual interference degradation (such as interference drift caused by antenna aging) by introducing a second detection window independent of mutation detection for long-term trend analysis. Through four levels of evaluation, namely statistical significance, degradation rate, cumulative degradation amount and recent persistence, it can provide early warning of gradual problems before the user's perception of degradation declines.

[0017] In one possible implementation of the first aspect, the trend analysis of the uplink interference level time series within the second detection window includes: applying the Mann-Kendall trend test to the uplink interference level time series within the second detection window, calculating the trend statistic Z value, and determining that there is a statistically significant upward trend when the Z value is positive and the corresponding p value is less than the second significance threshold; calculating the deterioration rate per unit time using Sen's slope estimation, and determining that the deterioration rate meets the standard when the deterioration rate exceeds the rate threshold; calculating the cumulative deterioration amount within the second detection window, and determining that the cumulative deterioration meets the standard when the cumulative deterioration amount exceeds the cumulative threshold.

[0018] The method provided by this invention uses a combination of Mann-Kendall trend test and Sen's slope estimation, which can both qualitatively determine the statistical significance of the trend and quantitatively estimate the rate of deterioration and the degree of cumulative deterioration, thus achieving a comprehensive quantitative assessment of the deterioration of slow disturbances.

[0019] In one possible implementation of the first aspect, the second detection window is 90 days prior to the target time, the second significance threshold is 0.05, the speed threshold is 0.1 dB per day, and the cumulative threshold is 5 dB; the recent persistence verification is to check whether the average uplink interference voltage over the most recent 7 days is higher than the third amplitude threshold of the historical average value excluding the most recent 7 days within the second detection window, where the third amplitude threshold is 3 dB.

[0020] The method provided by this invention can capture extremely small daily trends by using a long period window of 90 days and a daily deterioration rate threshold of 0.1 dB. The cumulative threshold of 5 dB ensures that intervention is carried out before the user perceives a significant decline. The recent continuous verification of 3 dB effectively confirms that the deterioration trend is still ongoing at the time of detection.

[0021] In one possible implementation of the first aspect, the method further includes: when it is determined that there is real interference in the target cell, extracting the interference feature vector of the target cell from three dimensions: time domain, frequency domain, and configuration domain; matching the interference feature vector with a preset interference mode feature library, and outputting the interference type and confidence level; generating a work order based on the interference type and the confidence level, wherein the work order includes a problem description, analysis evidence, judgment results, and handling suggestions.

[0022] The method provided by this invention automatically extracts multi-dimensional features and performs pattern matching after interference is confirmed, transforming expert experience into automatically executable judgment logic. This achieves an intelligent extension from anomaly detection to root cause analysis. Items with high confidence are directly adopted, those with medium confidence are provided for reference, and those with low confidence are marked for manual analysis, thus balancing automation efficiency and analytical accuracy.

[0023] In one possible implementation of the first aspect, the time-domain features include abrupt change amplitude and duration, the frequency-domain features include absolute value of interference level and spectral distribution shape, and the configuration-domain features include physical cell identifier (PCI) conflicts and neighbor cell interference correlations.

[0024] In one possible implementation of the first aspect, the method further includes: after the work order is processed, tracking the uplink interference level data of the target cell, determining the recovery status based on the recovery status of the processed uplink interference level, wherein the recovery status includes complete recovery, partial recovery, no recovery, and bounce; and executing a corresponding processing decision based on the recovery status.

[0025] The method provided by this invention transforms governance effect tracking into an automated closed-loop process by defining four recovery states and corresponding processing decisions, thus solving the problems of objective evaluation of processing effects and recurring similar problems in traditional operation and maintenance.

[0026] Secondly, the present invention provides a wireless interference detection system, the system comprising: a mutation detection module, configured to acquire an uplink interference level time sequence of a target cell within a first detection window at a target time, wherein the end point of the first detection window is located before the target time; to detect mutation points in the uplink interference level time sequence, wherein when the mutation direction corresponding to the detected target mutation point is upward and the mutation amplitude is greater than a first amplitude threshold, the date corresponding to the target mutation point is determined as a candidate anomaly occurrence date; and a persistence confirmation module, configured to acquire uplink interference level data of the target cell within a verification period, wherein the verification period is from the candidate anomaly occurrence date to the end point of the first detection window; and when the uplink interference level data for a consecutive preset number of days within the verification period all meet the persistent anomaly condition, the system determines that the target cell has real interference and generates alarm information.

[0027] In one possible implementation of the second aspect, the system further includes: a trend analysis module, used to acquire the uplink interference level time series of the target cell within a second detection window, wherein the length of the second detection window is greater than the length of the first detection window, and the end point of the second detection window is located before the target time; performing trend analysis on the uplink interference level time series within the second detection window, and determining that the target cell has real interference and generating alarm information when a statistically significant upward trend is detected, the deterioration rate exceeds a rate threshold, the cumulative deterioration exceeds a cumulative threshold, and the recent persistence verification is passed.

[0028] In one possible implementation of the second aspect, the system further includes: a feature analysis module, used to extract interference feature vectors of the target cell from three dimensions—time domain, frequency domain, and configuration domain—when it is determined that there is real interference in the target cell, match the interference feature vectors with a preset interference mode feature library, and output the interference type and confidence level; and a work order generation module, used to generate a work order based on the interference type and the confidence level, wherein the work order includes a problem description, analysis evidence, judgment results, and handling suggestions.

[0029] Thirdly, an electronic device is provided, the electronic device including a memory and one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the method as described in any implementation of the first aspect.

[0030] Fourthly, a computer-readable storage medium is provided, including computer instructions that, when executed on an electronic device, cause the electronic device to perform a method as described in any implementation of the first aspect.

[0031] Understandably, the beneficial effects of the system of the second aspect, the electronic device of the third aspect, and the computer-readable storage medium of the fourth aspect provided above can be referenced to the beneficial effects of the first aspect and any of its possible design embodiments, which will not be repeated here. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention; Figure 2 A flowchart of a wireless interference detection method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a wireless interference detection system provided in an embodiment of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the present invention, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. The "or" in the present invention is merely a description of the relationship between the related objects, indicating that three relationships can exist. For example, A or B can represent: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Furthermore, in the description of the present invention, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items.

[0034] Furthermore, to facilitate a clear description of the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" are not necessarily different.

[0035] In this embodiment of the invention, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this embodiment of the invention should not be construed as superior or more advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

[0036] With the large-scale deployment and commercial use of 4G and 5G mobile communication networks, wireless interference has become a core factor affecting network operation quality and reducing user communication experience. Interference sources include various types such as co-channel and adjacent-channel interference within the system, interference from external illegal devices, and environmental multipath interference. Network operators need to promptly detect and address various types of wireless interference to ensure stable network operation. However, the traditional model of passively discovering problems through manual inspections and user complaints suffers from slow response and low efficiency. Therefore, how to achieve proactive and automated detection of wireless interference has become a common problem urgently needing to be solved in the field of mobile communication operation and maintenance.

[0037] Currently, the industry commonly uses fixed threshold alarms for wireless interference monitoring. By setting a uniform threshold for uplink interference levels, an alarm is automatically triggered when the monitored indicators exceed the threshold. This method, to some extent, replaces manual periodic inspections and initially achieves proactive detection of interference problems. However, this single judgment mechanism has significant shortcomings. It not only fails to adapt to the service differences of different cells, easily generating false alarms and missed alarms, but also cannot effectively distinguish between temporary signal fluctuations and long-term persistent faults. It is completely unable to identify gradual interference degradation caused by loose antennas, aging components, etc. At the same time, the entire interference handling process is fragmented, lacking the end-to-end closed-loop capability of intelligent root cause analysis, automatic dispatching, and verification of governance effects, making it difficult to meet the high-precision, all-scenario, and automated interference governance needs of modern mobile communication networks.

[0038] In view of this, embodiments of the present invention provide a wireless interference detection method and system. The method includes: acquiring an uplink interference level time sequence of a target cell within a first detection window at a target time, wherein the end point of the first detection window is located before the target time; performing abrupt change detection on the uplink interference level time sequence, wherein when the direction of the detected abrupt change is upward and the magnitude of the abrupt change is greater than a first magnitude threshold, the date corresponding to the abrupt change is determined as a candidate anomaly occurrence date; acquiring uplink interference level data of the target cell within a verification period, wherein the verification period is from the date of the candidate anomaly occurrence to the end point of the first detection window; and determining that the target cell has real interference and generating alarm information when the uplink interference level data for a consecutive preset number of days within the verification period all meet the continuous anomaly condition.

[0039] The method provided by this invention sets the detection window endpoint before the target time, and completes mutation detection and continuity confirmation in one go based on the available historical data, so as to realize alarm output on the same day without waiting for future data.

[0040] In some embodiments, the wireless interference detection method provided by this invention can be executed by a wireless interference detection system. As an example, the detection system can be any electronic device 200 with data processing capabilities, such as a server, general-purpose computer, or cloud platform; the specific implementation of the detection system is not limited here.

[0041] Figure 1 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention is shown. The electronic device 200 includes a processor 210, a memory 220, and a communication interface 230.

[0042] Processor 210 may include one or more processing cores. Processor 210 connects to various parts within electronic device 200 using various interfaces and lines, and performs various functions and processes data of electronic device 200 by running or executing instructions, programs, code sets, or instruction sets stored in memory 220, and by calling data stored in memory 220. Optionally, processor 210 may be implemented using at least one of the following hardware forms: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA).

[0043] The memory 220 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 220 may include a non-transitory computer-readable storage medium. The memory 220 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 220 may include a program storage area. This program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the various method embodiments described above, etc.

[0044] Communication interface 230 is used to communicate with other devices, equipment or communication networks, such as data storage devices, image processing devices or Ethernet, wireless access network (RAN), wireless local area network (WLAN), etc.

[0045] In terms of physical implementation, the aforementioned devices (such as processor 210, memory 220, and communication interface 230) can each be devices within the same device (such as a laptop computer). Alternatively, at least two of these devices can be located within the same device, i.e., as different devices within the same device, similar to the deployment of devices or components in a distributed system.

[0046] It is understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device 200. In other embodiments of the present invention, the electronic device 200 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0047] The following description, in conjunction with the accompanying drawings, illustrates a wireless interference detection method provided by an embodiment of the present invention.

[0048] Figure 2 This is a flowchart illustrating a wireless interference detection method provided in an embodiment of the present invention. Optionally, this method can be... Figure 1 The illustrated electronic device 200 performs this method, which includes the following steps: S1. At the target time, obtain the uplink interference level time sequence of the target cell within the first detection window, wherein the end point of the first detection window is located before the target time.

[0049] Specifically, the target time is the trigger time for the system to perform detection. The system automatically starts a detection task once a day at the preset target time (e.g., 2:00 AM), performing interference detection and analysis on all target cells one by one. The target time can be flexibly set according to the actual needs of network operation and maintenance.

[0050] It should be noted that uplink interference level is a core performance indicator reflecting the uplink frequency band interference power received by a cell, and its unit is dBm. A higher uplink interference level indicates greater interference; for example, an uplink interference level of -90dBm is more severe than -110dBm. A sudden increase in uplink interference level is a typical characteristic of external interference.

[0051] Before constructing the time series, in one possible implementation, the method provided in this embodiment of the invention further includes a data preprocessing step: collecting raw uplink interference level data of the target cell from the network management system, and cleaning and formatting the raw data. Specifically, the preprocessing operations include: removing invalid values ​​(such as data points with values ​​of -999 or null values, which are usually caused by base station sensor failures or data transmission interruptions), correcting outliers (replacing values ​​exceeding 3 times the standard deviation with boundary values ​​to avoid interference from extreme outliers to statistical tests), filling missing values ​​(using the average of the two days before and after for linear interpolation to ensure the continuity of the time series), and standardizing the format (unifying units and field naming conventions). The preprocessed data effectively avoids interference from invalid and outlier values ​​to statistical analysis, fills the information gaps caused by missing data, and allows subsequent mutation detection and trend analysis to be carried out based on real, continuous, and regular data.

[0052] Furthermore, the system identifies specific business scenarios by using keywords in the cell names (such as "school," "venue," and "shopping mall"), and establishes separate baseline models for these scenarios. For low-value cells with recently low business volume, a more lenient detection threshold or a reduced detection frequency is used to optimize the allocation of computing resources.

[0053] In one possible implementation, the first detection window is the 30 working days prior to the target time. All data is aggregated at the daily granularity to form a daily uplink interference level sequence for each cell, taking the average of the daily busy hours. Each date is also labeled with a service tag: weekday, weekend, or holiday, for subsequent scenario-specific baseline calculations.

[0054] It should be understood that the reason for choosing 30 working days as the first detection window is that 30 working days cover a business cycle of about one calendar month, which includes the business characteristics of different stages at the beginning, middle and end of the month, and avoids the baseline from containing outdated information that no longer reflects the current network status due to an excessively long window.

[0055] S2. Detect abrupt change in the uplink interference level time series. When the direction of the detected target abrupt change is upward and the magnitude of the abrupt change is greater than the first magnitude threshold, determine the date corresponding to the target abrupt change as the candidate abrupt change date.

[0056] Specifically, the system automatically identifies locations where significant abrupt changes in uplink interference levels occur within the time series of the first detection window. A sudden increase in uplink interference levels is a typical statistical characteristic of external interference or equipment malfunction; detecting these abrupt changes allows for rapid location of the starting point of the anomaly.

[0057] In one possible implementation, mutation point detection employs the Pettitt nonparametric test. The Pettitt test is a nonparametric statistical test used to identify locations in a time series where the distribution has undergone significant changes. Its advantages include not requiring assumptions about the data distribution shape (such as a normal distribution), being insensitive to a few outliers, and being well-suited for non-normal, noisy time series such as network performance data. The specific steps are as follows: S201. For the uplink interference level time series (x1, x2, ..., xN), where N is the number of data points within the first detection window (N=30 when the first detection window is 30 working days), calculate the U statistic corresponding to each segment point t (t ranges from 1 to N-1). The U statistic is used to measure the degree of difference in the distribution of the two data segments before and after the segment point, and its recursive formula is: U(1,N)=0; U(t,N)=U(t-1,N)+Σsgn(xt-xj), where j ranges from 1 to N.

[0058] The sgn function is defined as follows: it takes the value 1 when (xt-xj) is greater than 0, 0 when (xt-xj) is equal to 0, and -1 when (xt-xj) is less than 0. The recursive formula reduces the computational complexity of the U statistic from O(N²) to O(N), meeting the efficiency requirements of real-time detection of large-scale cells.

[0059] S202. Calculate the absolute value of all U(t,N) (t=1,2,...,N-1), and take the split point corresponding to the maximum absolute value as the candidate mutation point. The date corresponding to this point is denoted as Tm.

[0060] It should be noted that the Pettitt test cannot mathematically determine the last data point of the sequence as a mutation point. This is an inherent property of the algorithm. This scheme adapts to this characteristic by setting the end point of the first detection window before the target time, without affecting the effectiveness of the detection.

[0061] The method for determining the direction of the abrupt change is as follows: if U(Tm,N) is positive, it means that the overall uplink interference level in the second half of Tm is higher than that in the first half, that is, the interference increases abruptly, which is the main abnormal direction to be concerned about; if U(Tm,N) is negative, it means that the overall second half is lower than that in the first half, that is, the interference decreases abruptly.

[0062] S203. Calculate the significance p-value corresponding to the candidate mutation point. The p-value is used to measure the statistical significance of the mutation point; the smaller the p-value, the less likely the mutation is caused by random fluctuations. The approximate formula for calculating the p-value is: p = 2 × exp(-6 × (U_max) 2 ) / (N 3 +N 2 )) Where U_max is the maximum absolute value of the U statistic.

[0063] In one possible implementation, when the p-value is less than the first significance threshold of 0.001, the candidate mutation point is determined to be a statistically significant target mutation point. The rationale for choosing 0.001 as the first significance threshold is that, after verification with thousands of cells on the live network, this value strikes a balance between detection sensitivity (able to detect real interference) and specificity (not falsely reporting normal fluctuations). If the threshold is set larger (e.g., 0.01), the false alarm rate will increase significantly; if it is set smaller (e.g., 0.0001), some real interference may be missed.

[0064] S204. Calculate the difference between the mean of the latter segment and the mean of the former segment of the target mutation point as the mutation amplitude.

[0065] Specifically, the time series is divided into a first segment (x1 to xTm) and a second segment (xTm+1 to xN) with the target mutation point Tm as the boundary. The mean of the first segment and the mean of the second segment are calculated respectively. The difference between the mean of the second segment and the mean of the first segment is the mutation amplitude (unit: dB).

[0066] In one possible implementation, when the mutation direction is upward and the mutation amplitude is greater than the first amplitude threshold of 10dB, the date Tm corresponding to the target mutation point is determined as the candidate anomaly occurrence date. The rationale for choosing 10dB as the first amplitude threshold is that the daily fluctuation of uplink interference levels in a normal cell is usually within 5dB; a 10dB increase significantly exceeds the normal range and falls within the interference level requiring attention. Mutations below 10dB are considered normal fluctuations and do not generate candidate anomalies, thereby reducing the false alarm rate.

[0067] It should be noted that if the p-value does not meet the first significance threshold, or the mutation magnitude does not meet the first magnitude threshold, it means that no statistically significant mutation with the required magnitude was found within the first detection window. In this round of mutation detection, no candidate anomalies will be output, and the system will wait for the next detection cycle.

[0068] S3. Obtain the uplink interference level data of the target cell during the verification period, wherein the verification period is from the date of occurrence of the candidate anomaly to the end of the first detection window.

[0069] It should be understood that statistically significant differences do not necessarily equate to persistent faults requiring maintenance intervention. For example, a community might experience a sudden spike in uplink interference levels due to a temporary external interference source (such as temporary welding equipment at a nearby construction site), but the interference returns to normal the following day after the source disappears. Such brief fluctuations do not require on-site intervention; generating alarms for them would waste maintenance resources. This step obtains continuous data after the sudden change, providing a data foundation for subsequent persistent fault assessment.

[0070] Specifically, let the target time for the current detection be day T (the date the system starts detecting), and the end of the first detection window be day T-1 (the day before the target time). The verification period begins on the day after the candidate anomaly occurs (Tm+1) and ends at the end of the first detection window (day T-1). Each day within the verification period corresponds to one uplink interference level data point.

[0071] With this design, only when Tm ≤ T-4 will the validation period include at least 3 days of data for continuous assessment. If Tm > T-4 (i.e., the mutation occurred within the last 3 days), the validation data is insufficient, and the system temporarily stores the candidate anomaly, waiting for subsequent data accumulation before making a judgment. This design ensures the rigor of continuous assessment, preventing incomplete judgments due to insufficient data.

[0072] S4. When the uplink interference level data for consecutive preset days within the verification period all meet the persistent abnormality condition, it is determined that a real interference exists in the target cell, and alarm information is generated.

[0073] In a possible implementation, the specific steps of persistence confirmation are as follows: S401. Calculate the average value of uplink interference levels over a first preset number of days before the candidate abnormality occurrence date Tm as a service baseline. In a possible implementation, the first preset number of days is 14 working days. The reason for selecting 14 working days (about two weeks) is that two weeks cover two complete weekly working day cycles, which can effectively smooth the service differences on different dates within a week, and at the same time prevent the baseline cycle from being too long to include outdated historical information.

[0074] For example, if Tm is June 5, the average value of uplink interference levels over the 14 working days before Tm (the working days between May 16 and June 4) is taken as the baseline. Assume the baseline value is -110dBm.

[0075] S402. Calculate the length K of the verification period, where K is the difference in days between the end point of the first detection window (day T-1) and the candidate abnormality occurrence date (Tm), that is, K=(T-1)-Tm. The value of K indicates how many calendar days of uplink interference level data are available from the day after the mutation occurs to the end point of the window.

[0076] S403. When K≥N (N is a preset number of days, and in a possible implementation, N=3 days), extract the uplink interference level data of the last N days within the verification period. The reason for extracting the last N consecutive days (the N consecutive days closest to the end point of the first detection window) is that the data of these N days are closest to the current moment and can best reflect the latest state of the interference at the time of detection.

[0077] An example is given for illustration: if Tm is June 5 and T is June 9, then T-1 is June 8, K=3 days, the verification period is a total of 3 days including June 6, June 7 and June 8, and all 3 days are extracted. If Tm is June 5 and T is June 10, then T-1 is June 9, K=4 days, the verification period is a total of 4 days from June 6 to June 9, and the last 3 days, namely June 7, June 8 and June 9, are extracted.

[0078] If K<N, it indicates that the verification data is insufficient, and the system does not process the candidate abnormality for the time being, and waits for subsequent data accumulation to re-judge in the next detection cycle.

[0079] S404. When the uplink interference level data of the last N days all meet the persistent abnormality condition relative to the service baseline, it is determined that real interference exists in the target cell.

[0080] In one possible implementation, the persistent abnormal condition is that the uplink interference level is greater than or equal to the service baseline plus a second amplitude threshold, where the second amplitude threshold is 10dB.

[0081] For example, if the baseline value is -110dBm, the anomaly detection condition is an uplink interference level ≥ -100dBm. For anomalies with rising uplink interference levels, this condition must be met on every day of the last 3 days of the verification period.

[0082] When the above conditions are met, it indicates that the abnormal rise in uplink interference level has lasted for at least 3 days and has not recovered by the end of the first detection window (yesterday). This constitutes persistent interference requiring maintenance intervention, and the system confirms it as genuine interference and generates an alarm message. If the conditions are not met (i.e., any day in the last N days does not meet the persistent abnormality condition), it indicates that the interference has naturally recovered during the verification period, and the candidate anomaly is discarded without generating any alarm.

[0083] The method provided by this invention constructs a complete mutation detection chain through the progressive coordination of three stages: detection and localization, mutation determination, and persistence verification. The Pettitt nonparametric test, through recursive calculation of the U statistic and significance determination of the p-value, can accurately identify sudden abnormal changes in uplink interference levels; a 10dB amplitude threshold filters out normal service fluctuations; and 3-day persistence verification effectively distinguishes between continuous interference requiring maintenance intervention and short-term fluctuations that can recover on their own. These stages significantly reduce the false alarm rate while ensuring detection sensitivity, avoiding ineffective investment of maintenance resources.

[0084] In some embodiments, the method provided by the present invention further includes: S5. Obtain the uplink interference level time series of the target cell within the second detection window, perform trend analysis on the uplink interference level time series within the second detection window, and when a statistically significant upward trend is detected, the deterioration rate exceeds the rate threshold, the cumulative deterioration exceeds the cumulative threshold, and the recent persistence verification is passed, determine that the target cell has real interference and generate alarm information.

[0085] It should be noted that S5 above is a parallel detection path independent of mutation detection (S2-S4), used to detect slow, gradual interference degradation problems that mutation detection cannot capture. These types of problems (such as gradually deteriorating coverage due to slow antenna loosening, or daily increases in interference due to aging feeders) have extremely small daily variations (possibly only 0.1-0.2 dB per day) and cannot be identified by Petitt's mutation detection. However, long-term accumulation can lead to severe network quality degradation. The method provided in this embodiment fills this detection blind spot through long-term trend analysis.

[0086] Specifically, the endpoint of the second detection window is also located before the target time.

[0087] In a possible implementation, the second detection window is 90 days before the target moment, covering an operation period of about one quarter, and the data integrity is required to reach more than 90% (i.e., at least 81 valid data points). A long period of 90 days can fully exhibit the slowly changing trend characteristics.

[0088] In a possible implementation, the Mann-Kendall trend test method is adopted for trend analysis. The Mann-Kendall test is a non-parametric method for detecting the monotonic trend of a time series, which has no requirements on the data distribution and is not interfered by a few outliers.

[0089] The specific steps are as follows: S501. For the uplink interference level time series (x1, x2, ..., xn) with a length of n in the second detection window, count all data pairs (xi, xj) where i<j. The calculation formula of the S statistic is: S=Σ(i=1 to n-1)Σ(j=i+1 to n)sgn(xj-xi) A positive S statistic indicates that the overall series shows an upward trend, and a negative S statistic indicates a downward trend.

[0090] S502. Calculate the variance Var(S) of the S statistic. When there are identical values in the series, the variance needs to be corrected. The calculation formula of the variance is: Var(S)=[n×(n-1)×(2n+5)-Σ(p=1 to g)tp×(tp-1)×(2tp+5)] / 18 Where g is the number of identical value groups in the series, and tp is the number of identical values in the p-th group. The correction term ensures that the variance estimation is still accurate when duplicate values exist.

[0091] S503. Calculate the trend statistic Z value: If S>0, then Z=(S-1) / √Var(S); if S=0, then Z=0; if S<0, then Z=(S+1) / √Var(S).

[0092] A positive Z value indicates that the series shows an upward trend (interference increases), and a negative Z value indicates that the series shows a downward trend (interference decreases). The present invention mainly focuses on the upward trend of interference.

[0093] In a possible implementation, when the Z value is positive and the corresponding p value (which can be obtained by looking up the standard normal distribution table) is less than the second significance threshold of 0.05, it is determined that there is a statistically significant upward trend. 0.05 is a commonly used significance level in statistics, indicating that the probability that the trend is caused by random fluctuations is less than 5%.

[0094] S504. Quantify the deterioration rate using Sen's slope estimation. Calculate the slope (xj-xi) / (j-i) for all data pairs (i<j), and take the median of all slopes as the Sen's slope estimate, which represents the average variation of the uplink interference level per unit time (per day). The advantage of median estimation is that it is not affected by extreme values, and the estimation result is more robust than the mean value.

[0095] In a possible implementation, when the deterioration rate exceeds a rate threshold (0.1 dB per day), it is determined that the deterioration rate meets the standard. The rate threshold of 0.1 dB per day means that the interference rises at a rate of 0.1 dB per day. Although the daily change is negligible, the cumulative increase can reach 9 dB in 90 days, which is a trend that requires attention.

[0096] S505. Calculate the cumulative deterioration amount in the second detection window. The cumulative deterioration amount is the Sen's slope estimate multiplied by the window length (90 days). In a possible implementation, when the cumulative deterioration amount exceeds the cumulative threshold of 5 dB, it is determined that the cumulative deterioration meets the standard. The cumulative threshold of 5 dB exceeds the normal daily fluctuation range of the uplink interference level (usually within 5 dB), and intervention and processing should be performed before the user perception deteriorates significantly.

[0097] S506. Perform recent persistence verification. To trigger an alarm for a slow deterioration trend, the condition of "still continuing up to the target moment" needs to be satisfied.

[0098] In a possible implementation, the data of the latest 7 days (i.e., day T-7 to day T-1) is checked: the actual mean value of the uplink interference level in the latest 7 days is calculated; the mean value of the latest 7 days is compared with the historical mean value of the second detection window excluding the latest 7 days. When the mean value of the latest 7 days is more than 3 dB, the third amplitude threshold, higher than the historical mean value, it is confirmed that the deterioration trend indeed exists and is still ongoing. The verification threshold of 3 dB is larger than the normal measurement error range (usually ±2 dB), which can effectively confirm the persistence of the deterioration trend.

[0099] When the trend is statistically significant (p<0.05 and Z is a positive value), the deterioration rate meets the standard (>0.1 dB per day), the cumulative deterioration meets the standard (cumulative >5 dB in 90 days), and the recent persistence verification is passed (mean value of the latest 7 days > historical mean value +3 dB) are all satisfied, it is determined that the target cell has real interference and alarm information is generated.

[0100] The coordination rules for anomalies detected simultaneously by mutation detection and trend analysis are as follows: If Petitt detects a candidate mutation point with a mutation magnitude exceeding 10dB, the Petitt result is prioritized, with the Mann-Kendall trend serving as background reference information; if only Mann-Kendall detects a trend and recent persistence verification passes, while Petitt shows no significant mutation, the trend analysis result is adopted; if both detect anomalies and the mutation point is located in the early part of the window (within the first 15 days) and the trend spans the entire window, it is determined to be a stage of deterioration occurring against a long-term deterioration background, and the output is merged into a single alarm with correlation annotation to avoid duplicate alerts. The two detectors run independently and in parallel without blocking each other.

[0101] S6. When it is determined that there is real interference in the target cell, extract the interference feature vector of the target cell from three dimensions: time domain, frequency domain and configuration domain; match the interference feature vector with the preset interference mode feature library, and output the interference type and confidence level; generate a work order based on the interference type and confidence level.

[0102] Once the interference is confirmed to be real, the system enters the in-depth analysis phase to conduct an intelligent preliminary judgment of the root cause.

[0103] In one possible implementation, feature extraction is performed across three dimensions: time domain, frequency domain, and configuration domain. Time domain features include the magnitude of the abrupt change (mean interference during the abnormal period minus the baseline value, in dB) and the duration (number of days from Tm to T-1). Frequency domain features include the absolute value of the interference level (average uplink interference level during the abnormal period) and the shape of the spectral distribution (examining the distribution pattern of the interference level across different frequency domain resource blocks, such as flat spectrum, narrowband spikes, and broadband rise). Configuration domain features include Physical Cell Identifier (PCI) conflicts (checking for modulo-3 or modulo-6 conflicts with neighboring cells) and the correlation of interference between neighboring cells (calculating the correlation coefficient of uplink interference levels between the abnormal cell and surrounding neighboring cells).

[0104] The system has a built-in preset interference mode feature library, which includes the following typical interference modes: The characteristic combination of external radio frequency interference is: abrupt change amplitude ≥10dB, duration ≥3 days, no PCI conflict, and flat spectrum. When all characteristics are met, the confidence level is high, and it is recommended to investigate by scanning the frequency on-site to find the illegal signal source.

[0105] The characteristic combination of co-channel interference from neighboring cells is: increased interference during busy hours, presence of PCI mode 3 conflict, and high correlation with interference from neighboring cells. When 2-3 of these conditions are met, the confidence level is medium, and the recommended troubleshooting direction is to optimize PCI, adjust antenna tilt angle, or power.

[0106] The characteristic combination of antenna feeder system faults is: slowly increasing interference (cumulative ≥5dB over 90 days) and synchronously decreasing downlink traffic. When both conditions are met, the confidence level is high, and it is recommended to investigate the feeder and antenna hardware at the site.

[0107] The characteristic combination of hardware failures includes: sudden changes in interference, accompanied by other alarms such as VSWR. The confidence level is medium; the recommended troubleshooting direction is to check the device's alarm logs.

[0108] The extracted feature vectors are compared one by one with the conditional expressions in the feature library. Based on the matching results, the interference type and confidence level (high / medium / low) are output. Results with high confidence are directly adopted and automatically filled into the work order; results with medium confidence provide multiple possible reasons for reference and are marked as pending confirmation in the work order; results with low confidence or no match are marked as unknown type, but do not block the alarm process, and are marked as requiring manual analysis in the work order.

[0109] The work order includes the following structured content: problem description (abnormal cell ID, interference occurrence time, uplink interference level rise), analysis evidence (uplink interference level trend chart, interference feature summary), judgment results (suspected interference type and confidence level), and handling suggestions (specific investigation steps). Work orders are automatically prioritized according to the following rules: Emergency – affecting VIP areas (such as government offices, transportation hubs), or interference levels exceeding -90dBm; High – interference levels between -100dBm and -90dBm lasting more than 3 days; Medium – interference levels between -105dBm and -100dBm lasting more than 3 days; Low – other interference requiring attention.

[0110] In one possible implementation, the feature library supports closed-loop self-learning updates: when human intervention corrects the judgment results, the system automatically records the feature vector of the case and the finally confirmed root cause; monthly cluster analysis is performed on newly accumulated labeled samples, and candidate feature rules are generated when new recurring patterns are discovered, which are then added to the feature library after human confirmation; for unknown types of cases that cannot be matched continuously, the system regularly compiles reports for expert analysis and is used for subsequent expansion of the feature library.

[0111] S7. After the work order is processed, track the uplink interference level data of the target cell, determine the recovery status based on the recovery of the uplink interference level after processing, and execute the corresponding processing decision based on the recovery status.

[0112] In one possible implementation, the system defines four recovery states: Full Recovery: The average uplink interference voltage over three consecutive days recovers to within 5 dB of the baseline value, with no further deterioration trend. Partial Recovery: Interference improves (decreases by more than 5 dB), but remains above the baseline value plus 5 dB. No Recovery: Interference shows no significant improvement (change less than 3 dB), or continues to worsen. Rebound: Improves briefly after treatment, but deteriorates again to pre-treatment levels within 7 days.

[0113] Based on the recovery status, the corresponding processing decision is executed: when fully recovered, the work order is automatically closed and archived as a successful case; when partially recovered, a supplementary work order is automatically generated, suggesting further investigation of remaining issues; when not recovered, the work order priority is upgraded and pushed to human experts for processing; when a rebound occurs, a recurrence alarm is triggered, the original work order is reactivated and marked as recurring.

[0114] As can be seen from S1-S7 above, the method provided by the embodiments of the present invention constructs a complete automated processing link from interference detection to governance effect evaluation through a three-layer detection architecture of mutation detection, trend analysis and persistence confirmation, combined with an intelligent judgment mechanism of multi-dimensional feature extraction and pattern matching, and a closed-loop verification system of recovery state tracking.

[0115] This invention can also be understood as follows: It employs a cascaded determination mechanism combining Pettitt nonparametric testing and business continuity confirmation, solving the problem of high false alarm rates in traditional fixed threshold methods; it uses a combination of Mann-Kendall trend testing and Sen's slope estimation to fill the detection blind spot of slow interference degradation; it utilizes three-dimensional feature extraction in the time domain, frequency domain, and configuration domain, along with interference pattern library matching, to achieve intelligent extension from anomaly detection to root cause analysis; and it employs four types of recovery state definitions and corresponding processing decision trees to achieve a systematic breakthrough from single-point detection to closed-loop governance.

[0116] To further illustrate the technical solution of the present invention, a specific embodiment is provided below.

[0117] In a residential 4G cell (Cell_A), the uplink interference level is consistently around -110dBm. The system collects the previous day's uplink interference level data for Cell_A at 2:00 AM daily, aggregates it into a daily average, and labels it as a weekday.

[0118] At 2:00 AM on June 9th, the system performed a Pettitt test on the uplink interference level sequence for 30 working days from May 10th to June 8th. Calculations showed that the maximum value of the U statistic occurred on June 5th, with p=0.0003 (less than 0.001). The mean before the abrupt change was -110dBm, and the mean after the abrupt change was -85dBm, an increase of 25dB exceeding the 10dB threshold. The system calculated K=(June 8th)-(June 5th)=3 days, satisfying K≥3. Uplink interference level data for the last 3 days of the validation period (June 6th, June 7th, and June 8th) were extracted: -82dBm, -83dBm, and -85dBm, all greater than or equal to -100dBm (-110dBm+10dB), satisfying the condition of persistent anomaly. This confirmed the presence of real interference in Cell_A.

[0119] The system enters the feature analysis phase: sudden change amplitude 25dB, duration ≥3 days, no PCI conflict, and flat spectrum. After matching the feature library, it outputs "suspected external RF interference, high confidence", generates an alarm work order, and sets the priority to high (interference level between -100dBm and -90dBm). On-site frequency scanning and troubleshooting are recommended.

[0120] After receiving the order, the engineer went to the site to investigate and found an illegal signal amplifier operating near the residential area. After shutting down the interference source, the uplink interference level recovered to -108dBm, -109dBm, and -111dBm from June 10th to June 12th. The system automatically assessed: the average value for the three consecutive days after processing was -109.3dBm, and the baseline value + 5dB = -105dBm. Since the average value was less than this value, it was judged as "completely restored," and the work order was automatically closed and archived as a successful case.

[0121] This case demonstrates a complete closed-loop process: the system completed mutation detection and persistence confirmation in one go in the early morning of June 9th, verifying the continuous interference from the mutation on June 5th to June 8th, a total of 3 days, and confirming that the interference was still ongoing as of June 8th, thus issuing an alarm. From detection to root cause analysis to effect verification, the entire process is completed automatically without human intervention.

[0122] The foregoing mainly describes the solutions of the embodiments of the present invention from a methodological perspective. It is understood that, in order to achieve the above-mentioned functions, the wireless interference detection system 100 includes at least one of the hardware structures and 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 in 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.

[0123] In this embodiment of the invention, the wireless interference detection system 100 can be divided into functional units according to the above method example. For example, the wireless interference detection system 100 can be divided into functional units corresponding to each function, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this embodiment of the invention is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0124] For example, Figure 3 A schematic diagram of the hardware structure of a wireless interference detection system according to an embodiment of the present invention is shown. The wireless interference detection system 100 includes: The mutation detection module 110 is used to acquire the uplink interference level time series of the target cell within a first detection window at a target time, wherein the end point of the first detection window is located before the target time; to detect mutation points in the uplink interference level time series; and when the mutation direction corresponding to the detected target mutation point is upward and the mutation amplitude is greater than a first amplitude threshold, to determine the date corresponding to the target mutation point as a candidate anomaly occurrence date. The continuous verification module 120 is used to acquire uplink interference level data of the target cell during the verification period, which is from the date of the candidate anomaly to the end of the first detection window; when the uplink interference level data for a consecutive preset number of days during the verification period all meet the continuous anomaly condition, it is determined that there is real interference in the target cell and an alarm message is generated.

[0125] Optionally, the system further includes: a trend analysis module 130, used to acquire the uplink interference level time series of the target cell within a second detection window, wherein the length of the second detection window is greater than the length of the first detection window, and the end point of the second detection window is located before the target time; performing trend analysis on the uplink interference level time series within the second detection window, and determining that the target cell has real interference and generating alarm information when a statistically significant upward trend is detected, the deterioration rate exceeds the rate threshold, the cumulative deterioration exceeds the cumulative threshold, and the recent persistence verification is passed.

[0126] Optionally, the system further includes: a feature analysis module 140, used to extract interference feature vectors of the target cell from three dimensions—time domain, frequency domain, and configuration domain—when it is determined that there is real interference in the target cell, match the interference feature vectors with a preset interference mode feature library, and output the interference type and confidence level; and a work order generation module 150, used to generate a work order based on the interference type and the confidence level, wherein the work order includes a problem description, analysis evidence, judgment results, and handling suggestions.

[0127] It should be understood that specific descriptions of the above-mentioned optional methods can be found in the foregoing method embodiments, and will not be repeated here. Furthermore, explanations of any of the wireless interference detection systems 100 provided above, as well as descriptions of their beneficial effects, can be found in the corresponding method embodiments described above, and will not be repeated here.

[0128] This invention also provides a computer-readable storage medium storing at least one computer instruction, which is loaded and executed by a processor to implement the methods of the various embodiments described above. Explanations of the relevant content and descriptions of the beneficial effects of any of the computer-readable storage media provided above can be found in the corresponding embodiments described above, and will not be repeated here.

[0129] This invention also provides a chip. This chip integrates a control circuit for implementing the functions of the aforementioned wireless interference detection system 100 and one or more ports. Optionally, the functions supported by this chip are as described above and will not be repeated here.

[0130] Those skilled in the art will understand that the program for implementing all or part of the steps of the above embodiments, which can be executed by a program instructing related hardware, can be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a random access memory, etc. The processing unit or processor mentioned above can be a central processing unit, a general-purpose processor, an application-specific integrated circuit (ASIC), a microprocessor (DSP), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0131] This invention also provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform any of the methods described in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0132] It should be noted that the devices for storing computer instructions or computer programs provided in the embodiments of the present invention, such as, but not limited to, the aforementioned memory, computer-readable storage medium, and communication chip, are all non-transitory. Those skilled in the art should recognize that the functions described in the embodiments of the present invention in one or more of the above examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable storage medium or transmitted as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium accessible to general-purpose or special-purpose computers.

[0133] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting wireless interference, characterized in that, include: At the target time, the uplink interference level time sequence of the target cell within the first detection window is obtained, and the end point of the first detection window is located before the target time. The uplink interference level time series is subjected to abrupt change detection. When the direction of the abrupt change corresponding to the detected target abrupt change is upward and the abrupt change amplitude is greater than the first amplitude threshold, the date corresponding to the target abrupt change is determined as the candidate abrupt change date. Acquire uplink interference level data of the target cell during the verification period, wherein the verification period is from the date of occurrence of the candidate anomaly to the end of the first detection window; When the uplink interference level data for a consecutive preset number of days during the verification period meets the continuous abnormal condition, it is determined that there is real interference in the target cell and an alarm message is generated.

2. The method according to claim 1, characterized in that, The step of detecting abrupt changes in the uplink interference level time series includes: The uplink interference level time series was subjected to the Pettitt nonparametric test to calculate the U statistic corresponding to each segmentation point, and the segmentation point with the largest absolute value of the U statistic was selected as the candidate mutation point. Calculate the significance p-value corresponding to the candidate mutation point. When the significance p-value is less than the first significance threshold, the candidate mutation point is determined to be the target mutation point. The difference between the mean value of the latter segment and the mean value of the former segment of the target mutation point is calculated as the mutation amplitude.

3. The method according to claim 2, characterized in that, The first significance threshold is 0.001, the first amplitude threshold is 10dB, and the first detection window is the 30 working days prior to the target time.

4. The method according to claim 1, characterized in that, The step of determining that the target cell has real interference when the uplink interference level data for a consecutive preset number of days within the verification period all meet the continuous abnormal condition includes: The uplink interference average value for the first preset number of days prior to the date of occurrence of the candidate anomaly is calculated as the service baseline; Calculate the length of the verification period, where the length is the difference in the number of days between the end of the first detection window and the date of occurrence of the candidate anomaly; When the length is greater than or equal to the preset number of days, extract the uplink interference level data corresponding to the last preset number of days in the verification period; When the uplink interference level data for the last preset number of days all meet the continuous abnormal condition relative to the service baseline, it is determined that there is real interference in the target cell.

5. The method according to claim 4, characterized in that, The first preset number of days is 14 working days, and the preset number of days is 3 days; the continuous abnormal condition is that the uplink interference level is greater than or equal to the service baseline plus the second amplitude threshold, and the second amplitude threshold is 10dB.

6. The method according to claim 1, characterized in that, The method further includes: The uplink interference level time series of the target cell within the second detection window is obtained. The length of the second detection window is greater than the length of the first detection window, and the end point of the second detection window is located before the target time. Trend analysis is performed on the uplink interference level time series within the second detection window. When a statistically significant upward trend is detected, the deterioration rate exceeds the rate threshold, the cumulative deterioration exceeds the cumulative threshold, and the recent persistence verification is passed, it is determined that there is real interference in the target cell and an alarm message is generated.

7. The method according to claim 6, characterized in that, The trend analysis of the uplink interference level time series within the second detection window includes: The Mann-Kendall trend test is applied to the uplink interference level time series within the second detection window to calculate the trend statistic Z value. When the Z value is positive and the corresponding p value is less than the second significance threshold, it is determined that there is a statistically significant upward trend. Sen's slope estimation is used to calculate the rate of deterioration per unit time. When the rate of deterioration exceeds the rate threshold, it is determined that the rate of deterioration meets the standard. Calculate the cumulative deterioration within the second detection window. When the cumulative deterioration exceeds the cumulative threshold, it is determined that the cumulative deterioration has met the standard.

8. The method according to claim 7, characterized in that, The second detection window is 90 days prior to the target time, the second significance threshold is 0.05, the speed threshold is 0.1 dB per day, and the cumulative threshold is 5 dB. The recent persistence verification is to check whether the average uplink interference voltage over the past 7 days is higher than the third amplitude threshold of the historical average value excluding the past 7 days within the second detection window, where the third amplitude threshold is 3dB.

9. The method according to claim 1, characterized in that, The method further includes: When it is determined that there is real interference in the target cell, the interference feature vector of the target cell is extracted from three dimensions: time domain, frequency domain, and configuration domain. The interference feature vector is matched with a preset interference pattern feature library to output the interference type and confidence level. A work order is generated based on the interference type and the confidence level. The work order includes a problem description, analysis evidence, judgment results, and handling suggestions.

10. The method according to claim 9, characterized in that, The time-domain features include abrupt change amplitude and duration; the frequency-domain features include absolute value of interference level and spectral distribution shape; and the configuration-domain features include physical cell identifier (PCI) conflicts and neighbor cell interference correlations.

11. The method according to claim 9, characterized in that, The method further includes: After the work order is processed, the uplink interference level data of the target cell is tracked, and the recovery status is determined based on the recovery of the processed uplink interference level. The recovery status includes complete recovery, partial recovery, no recovery, and bounce. Execute the corresponding processing decision based on the recovery status.

12. A wireless interference detection system, characterized in that, include: The mutation detection module is used to acquire the uplink interference level time series of the target cell within a first detection window at a target time, wherein the end point of the first detection window is located before the target time; to detect mutation points in the uplink interference level time series; and when the mutation direction corresponding to the detected target mutation point is upward and the mutation amplitude is greater than a first amplitude threshold, the date corresponding to the target mutation point is determined as a candidate anomaly occurrence date. The continuous verification module is used to acquire uplink interference level data of the target cell during the verification period, which is from the date of the candidate anomaly to the end of the first detection window. When the uplink interference level data for a consecutive preset number of days during the verification period meets the continuous anomaly condition, it is determined that there is real interference in the target cell and an alarm message is generated.

13. The system according to claim 12, characterized in that, The system also includes: The trend analysis module is used to obtain the uplink interference level time series of the target cell within a second detection window. The length of the second detection window is greater than the length of the first detection window, and the end point of the second detection window is located before the target time. The module performs trend analysis on the uplink interference level time series within the second detection window. When a statistically significant upward trend is detected, the deterioration rate exceeds the rate threshold, the cumulative deterioration exceeds the cumulative threshold, and the recent persistence verification is passed, the module determines that there is real interference in the target cell and generates an alarm message.

14. The system according to claim 12, characterized in that, The system also includes: The feature analysis module is used to extract the interference feature vector of the target cell from three dimensions: time domain, frequency domain and configuration domain when it is determined that there is real interference in the target cell, match the interference feature vector with the preset interference mode feature library, and output the interference type and confidence level. The work order generation module is used to generate work orders based on the interference type and the confidence level. The work order includes a problem description, analysis evidence, judgment results, and handling suggestions.

15. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the wireless interference detection method as described in any one of claims 1-11.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to enable the computer to implement the wireless interference detection method as described in any one of claims 1-11.

Images