Offset monitoring method and system for base station antenna

Abstract

The invention discloses an offset monitoring method and system for a base station antenna, and the method comprises the steps: obtaining an antenna state parameter and an environment state parameter in a historical time period, and constructing a deformation prediction model; acquiring current antenna state parameters and environment state parameters in real time and inputting the current antenna state parameters and the environment state parameters into the deformation prediction model to obtain predicted values of the displacement angle, the inclination angle and the strain degree; when a certain value exceeds a preset deformation early warning threshold value, based on a preset fuzzy rule, reasoning to obtain a deformation level of the base station antenna; when the deformation grade reaches or is higher than a preset deformation early warning grade, estimating the offset degree of the base station antenna; and when the estimated offset degree is greater than a preset offset alarm range, giving a maintenance alarm. Based on the offset monitoring method, the offset monitoring cost of the base station antenna is reduced, the false alarm rate is reduced, and the prediction and judgment precision is improved.

Description

Technical Field

[0001] This invention relates to the field of base station antenna offset detection technology, and in particular to a method and system for monitoring the offset of a base station antenna. Background Technology

[0002] In wireless communication networks, the position and orientation of base station antennas are crucial for ensuring communication quality. However, in actual deployments, the position of base station antennas is not always fixed; their position and angle can be slightly offset by internal and external factors.

[0003] Such offsets mainly stem from two types of factors: firstly, natural environmental factors, such as strong winds and extreme temperature changes; and secondly, human activities, such as vibrations from nearby construction, both of which can cause base station antenna offsets. If these subtle changes are not detected and corrected in a timely manner, they will inevitably affect the signal coverage and thus reduce communication quality.

[0004] To address the offset problem, current mainstream methods generally rely on high-precision sensors and complex mathematical models. While these methods provide relatively accurate quantitative data, they also have several significant drawbacks: First, the cost is high. High-precision sensors are expensive, and the need for complex computational models further increases the overall cost of the solution, hindering large-scale adoption and application. Second, environmental interference easily leads to a high false alarm rate. These methods are sensitive to environmental factors, especially when relying on satellite positioning (such as BeiDou / GPS), where signal obstruction or multipath effects can introduce errors. More importantly, existing methods often lack full utilization of contextual data, simply relying on thresholds for judgment without considering historical data on environmental changes. Furthermore, sensor data itself is susceptible to noise from natural factors such as wind and temperature fluctuations. If only fixed static thresholds are used for anomaly detection, this noise can easily trigger false alarms, generating a large number of false alarms and increasing unnecessary maintenance burdens. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for monitoring the offset of base station antennas that combines hardware and software and can make full use of the base station antenna's own structure and historical environmental data to achieve quantitative prediction, qualitative evaluation and decision-making.

[0006] To achieve the above objectives, this invention discloses a method for monitoring the offset of a base station antenna, comprising: Obtain antenna state parameters related to the offset of the base station antenna within a historical time period and environmental state parameters related to the external environment in which the base station antenna is located. Based on the antenna state parameters, environmental state parameters, and support vector regression model framework of the historical time period, a deformation prediction model is constructed. The antenna state parameters and environmental state parameters of the current base station antenna are collected in real time and input into the deformation prediction model to obtain the predicted values ​​of the displacement, tilt angle and strain of the current base station antenna. When any one of the predicted values ​​of displacement, tilt angle and strain exceeds a preset deformation warning threshold, the deformation level of the base station antenna is inferred based on preset fuzzy rules. When the deformation level of the base station antenna reaches or exceeds the preset deformation warning level, the antenna state parameters and environmental state parameters of the current base station antenna are compared with the antenna state parameters and environmental state parameters of the historical time period to estimate the degree of offset of the base station antenna. When the estimated offset of the base station antenna exceeds the preset offset alarm range, a maintenance alarm is issued.

[0007] Specifically, the base station antenna includes a bracket and a top antenna, and the steps for obtaining the antenna state parameters and the environmental state parameters include: The acceleration at the bottom and middle of the bracket is detected by an accelerometer; The angular velocity and magnetic field strength of the head and sides of the top antenna are detected by the gyroscope and the magnetometer. The acceleration and stress changes at the connection point between the top antenna and the bracket are detected using accelerometers and strain sensors. The ambient temperature and humidity of the external environment where the base station antenna is located are detected by temperature and humidity sensors. The antenna state parameters include the acceleration at the bottom and middle of the bracket, the angular velocity and magnetic field strength at the head and sides of the top antenna, the acceleration and stress change at the connection point between the top antenna and the bracket, and the environmental state parameters include the ambient temperature and humidity of the external environment where the base station antenna is located.

[0008] Specifically, the steps for constructing the deformation prediction model include: The antenna state parameters and the environmental state parameters are filtered and preprocessed to obtain a feature dataset; The support vector regression model framework is initialized based on the feature dataset; The feature dataset is input into the initialized support vector regression model framework for training, thereby obtaining the deformation prediction model.

[0009] Furthermore, the filtering preprocessing steps include: Use a low-pass filter to remove high-frequency noise from the antenna state parameters and the environmental state parameters; The antenna state parameters and the environmental state parameters are smoothed using the moving average method.

[0010] Specifically, the initialization steps of the support vector regression model framework include: The kernel function of the support vector regression model framework is selected based on the distribution of the feature dataset, and the hyperparameters of the kernel function are adjusted by grid search and cross-validation methods.

[0011] Specifically, before inputting the antenna state parameters and environmental state parameters of the current base station antenna collected in real time into the deformation prediction model, they need to be filtered and preprocessed. The filtering preprocessing steps include: Use a low-pass filter to remove high-frequency noise from the antenna state parameters and the environmental state parameters; The antenna state parameters and the environmental state parameters are smoothed using the moving average method.

[0012] Furthermore, the step of reasoning to obtain the deformation level of the base station antenna based on preset fuzzy rules includes: Obtain the antenna state parameters and environmental state parameters of the current base station antenna after filtering and preprocessing, and calculate the standard deviation of the antenna state parameters; For the predicted displacement degree, tilt angle, and strain degree, the standard deviation of the antenna state parameters of the current base station antenna after filtering preprocessing, and the environmental state parameters, a membership function is configured to obtain a fuzzy set; Based on the fuzzy rules and the fuzzy set, fuzzy reasoning is performed and defuzzification is applied to obtain the deformation level of the base station antenna.

[0013] The present invention also discloses a base station antenna offset monitoring system, which operates based on the base station antenna offset monitoring method described above.

[0014] This invention also discloses a base station antenna offset monitoring system, which includes: One or more processors; Memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including instructions for performing the base station antenna offset monitoring method as described above.

[0015] The present invention also discloses a computer-readable storage medium comprising a computer program that can be executed by a processor to perform the base station antenna offset monitoring method as described above.

[0016] Compared with existing technologies, the base station antenna offset monitoring method provided by the above-mentioned technical solution of this invention, by simultaneously collecting historical antenna body parameters and environmental state parameters, constitutes a complete dataset from "environment" to "state," enabling the model to learn the relationship between environmental changes and base station antenna changes, directly addressing the root cause of base station antenna changes. Through the deformation prediction model, online and continuous quantitative evaluation of the base station antenna can be performed. Simultaneously, based on the introduced fuzzy rules, this monitoring method can perform deformation level inference for the base station antenna, effectively filtering false alarms caused by noise in the collected data and reducing the false alarm rate. This monitoring method only compares and estimates the deformation level with historical data when it reaches a level requiring intervention. Based on the estimated offset degree, it then compares it with a predetermined offset alarm range. This layered comparison and judgment increases the reliability of alarms, avoids false alarms, and greatly saves operation and maintenance resources. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the offset detection method for base station antennas in an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram illustrating the steps for obtaining antenna state parameters and environmental state parameters in an embodiment of the present invention.

[0019] Figure 3 This is a schematic diagram illustrating the steps involved in constructing a deformation prediction model in an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of the fuzzy inference steps for the deformation level of a base station antenna in an embodiment of the present invention. Detailed Implementation

[0021] To illustrate the technical content, structural features, objectives, and effects of this invention in detail, the following description, in conjunction with the embodiments and accompanying drawings, provides a comprehensive explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to be exemplary and should not be construed as limiting the scope of protection of this application.

[0022] This invention discloses a method for monitoring the offset of a base station antenna, enabling low-cost and high-accuracy monitoring of the degree of base station antenna offset. (See also...) Figure 1 As shown, the method for monitoring the offset of the base station antenna specifically includes the following steps: S1: Obtain antenna state parameters related to the offset of the base station antenna within a historical time period and environmental state parameters related to the external environment in which the base station antenna is located.

[0023] In this embodiment, the base station antenna includes a support frame and a top antenna. Antenna state parameters include the acceleration at the bottom and middle of the support frame, the angular velocity and magnetic field strength at the head and sides of the top antenna, and the acceleration and stress change at the connection point between the top antenna and the support frame. Environmental state parameters include the ambient temperature and humidity of the external environment where the base station antenna is located. Acceleration can be measured using an accelerometer, angular velocity using a gyroscope, magnetic field strength using a magnetometer, stress change using a strain sensor, and ambient temperature and humidity using a temperature sensor and a humidity sensor, respectively.

[0024] S2: Based on antenna state parameters, environmental state parameters, and support vector regression model framework from historical time periods, construct a deformation prediction model.

[0025] It should be noted that the "historical time period" refers to the antenna state parameters and environmental parameters collected at a certain time interval before the current moment (such as the previous week or month, etc. The training set of the deformation prediction model can be arbitrarily adjusted according to the characteristics of the base station antenna and the external environment, without specific restrictions).

[0026] S3: Real-time acquisition of antenna state parameters and environmental state parameters of the current base station antenna and input into the deformation prediction model to obtain predicted values ​​of the displacement, tilt angle and strain of the current base station antenna.

[0027] S4: Determine whether any one of the predicted displacement, tilt angle, and strain exceeds a preset deformation warning threshold. If yes, proceed to step S5; otherwise, return to step S3. The deformation warning threshold is related to the physical characteristics of the base station antenna and the external environmental conditions.

[0028] S5: Based on preset fuzzy rules, the deformation level of the base station antenna is inferred. The specific fuzzy rules should be defined by operation and maintenance personnel based on their experience and professional knowledge.

[0029] S6: Determine whether the deformation level of the base station antenna reaches or exceeds the preset deformation warning level. If yes, proceed to step S7; otherwise, return to step S3. In this embodiment, the deformation level can be set to three levels: slight, moderate, and severe. When the deformation level reaches moderate or severe, step S7 is executed.

[0030] S7: Compare the current antenna status parameters and environmental status parameters of the base station antenna with the antenna status parameters and environmental status parameters of historical time periods to estimate the degree of offset of the base station antenna.

[0031] It should be noted that the purpose of step S7 is to estimate the range of the base station antenna offset by comparing the differences between historical data and current data. This usually refers to the tilt angle of the base station antenna relative to its initial position. Alternatively, displacement or strain can be used as indicators of the offset.

[0032] S8: Determine whether the estimated offset of the base station antenna is greater than the preset offset alarm range. If yes, proceed to step S9; otherwise, return to step S3.

[0033] S9: Issue a maintenance alert.

[0034] Compared with existing technologies, the base station antenna offset monitoring method proposed in this invention improves upon existing technologies in terms of data foundation, judgment logic, and decision-making process, thereby reducing overall monitoring costs and enhancing monitoring accuracy, robustness, and engineering practicality. First, this monitoring method replaces some expensive hardware with algorithms (software), enabling high-value deformation monitoring through the use of low-cost sensor combinations.

[0035] Secondly, traditional methods typically monitor the parameters of the base station antenna itself in isolation, neglecting the structural impact of environmental changes. This invention, by simultaneously collecting and constructing an "environmental" [database]... The "state" associated dataset enables the model to learn the intrinsic relationship between environmental factors such as wind and temperature and base station antenna deformation, enhancing the model's ability to interpret the degree of offset from the causal level and laying a data foundation for accurate prediction.

[0036] During the real-time monitoring phase, this invention uses a deformation prediction model to make online and continuous predictions of the antenna displacement, tilt angle, and strain. Furthermore, by introducing fuzzy rules to infer the deformation level of the prediction results, it can effectively identify single-index anomalies caused by instantaneous noise (such as gusts of wind or temperature fluctuations), thereby significantly reducing the false alarm rate.

[0037] The decision-making mechanism of this method embodies a hierarchical verification approach. Only when the deformation level reaches a threshold requiring intervention is a comparative analysis with historical data initiated to estimate the specific degree of deviation. This estimated result must again exceed a preset deviation alarm range to trigger a final maintenance alarm. This "predictive" approach... Evaluate diagnosis The multi-layered, progressive decision-making structure enhances the credibility of alarms, avoids invalid alarms, and makes operation and maintenance responses more targeted, thereby saving valuable on-site maintenance resources.

[0038] Specifically, see Figure 2 As shown, in this embodiment, the steps for obtaining antenna state parameters and environmental state parameters include: S11: Accelerometers are used to detect the acceleration at the bottom and middle of the bracket. The accelerometers are used to detect the vibration and tilt of the bracket. The vibration and tilt of the antenna bracket directly affect the working state of the antenna. By installing accelerometers on the bracket, the stability of the bracket can be effectively monitored.

[0039] S12: The angular velocity and magnetic field strength of the top antenna's head and sides are detected using a gyroscope and magnetometer. The tilt angle and orientation of the top antenna's head directly affect signal coverage and quality. By installing a gyroscope and magnetometer on the head, the tilt of the base station antenna can be comprehensively monitored. S13: Accelerometers and strain sensors are used to detect the acceleration and stress changes at the connection point between the top antenna and the support frame. Stress changes at the connection point reflect whether the base station antenna is subjected to external forces, such as wind or hail. Strain sensors monitor stress changes at the connection point to prevent the antenna from loosening or detaching, thus indirectly affecting the antenna's displacement and tilt.

[0040] S14: The ambient temperature and humidity of the external environment where the base station antenna is located are detected by temperature and humidity sensors. Temperature changes affect the thermal expansion and contraction of the base station antenna materials, which may lead to displacement and tilting of the base station antenna. In high humidity environments, the materials may absorb water and expand, affecting the mechanical properties of the base station antenna, which may also lead to displacement and tilting.

[0041] Specifically, see Figure 3 As shown, in this embodiment, the steps for constructing the deformation prediction model include: S21: Perform filtering preprocessing on the antenna state parameters and environmental state parameters to obtain the feature dataset.

[0042] S22: Initialize the support vector regression model framework based on the feature dataset.

[0043] S23: Input the feature dataset into the initialized support vector regression model framework for training, thereby obtaining the deformation prediction model.

[0044] The filtering preprocessing steps include: (1) Use a low-pass filter to remove high-frequency noise from the antenna state parameters and environmental state parameters.

[0045] (2) The antenna state parameters and environmental state parameters are smoothed based on the moving average method.

[0046] The initialization steps of the support vector regression model framework include: The kernel function for the support vector regression model framework is selected based on the distribution of the feature dataset, and the hyperparameters of the kernel function are adjusted using grid search and cross-validation methods. The purpose of the initialization step is to select a suitable kernel function and dynamically adjust its hyperparameters according to the distribution of antenna state parameters and environmental state parameters of the base station antenna to adapt to the characteristics of the base station antenna data. Hyperparameters are a technical term understood by those skilled in the art; they vary depending on the kernel function and will not be elaborated upon here.

[0047] Specifically, before inputting the antenna state parameters and environmental state parameters of the current base station antenna, which are collected in real time, into the deformation prediction model, they need to be pre-processed by filtering to further reduce the impact of data noise. The steps of filtering pre-processing are as described above and will not be repeated here.

[0048] Furthermore, see Figure 4 As shown, the steps for inferring the deformation level of a base station antenna based on preset fuzzy rules include: S51: Obtain the antenna state parameters and environmental state parameters of the current base station antenna after filtering and preprocessing, and calculate the standard deviation of the antenna state parameters.

[0049] S52: To obtain a fuzzy set, configure the membership function for the standard deviation of the antenna state parameters of the current base station antenna after filtering and preprocessing, as well as the environmental state parameters, to obtain the predicted displacement, tilt angle, and strain.

[0050] S53: Perform fuzzy inference based on fuzzy rules and fuzzy sets, and then perform defuzzification processing to obtain the deformation level of the base station antenna.

[0051] The working principle of this monitoring method is described below using a specific implementation method, but this should not be considered as a limitation on the scope of protection of this monitoring method. The steps of this specific implementation method include: (I) Acquisition and processing of antenna state parameters and environmental state parameters First, multiple connection nodes are installed at key locations on the base station antenna to connect sensors including accelerometers, gyroscopes, magnetometers, strain sensors, temperature sensors, and humidity sensors. The specific installation locations are as follows: (1) For stents Accelerometers: Two triaxial accelerometers are installed at the bottom and middle of the antenna bracket to detect the vibration and tilt of the bracket.

[0052] (2) Regarding the top antenna Gyroscope: A three-axis gyroscope is installed on the head and two sides of the top antenna to detect the tilt angle of the top antenna.

[0053] Magnetometers: A triaxial magnetometer is installed on the head and two sides of the top antenna to detect the direction of the top antenna.

[0054] (3) Regarding the connection point Accelerometer: A triaxial accelerometer is installed at the connection between the top antenna and the bracket to detect vibrations at the connection point.

[0055] Strain sensor: A strain sensor is installed at the connection between the top antenna and the bracket to detect stress changes at the connection.

[0056] (4) Regarding the environment where the base station antenna is located Temperature sensor: Install a temperature sensor at a suitable location near the base station antenna to monitor the ambient temperature.

[0057] Humidity sensor: Install a humidity sensor at a suitable location near the base station antenna to monitor the ambient humidity.

[0058] Among them, accelerometers, gyroscopes, magnetometers, strain sensors, temperature sensors, and humidity sensors can be installed using the following methods: ① Adhesive installation: Use high-strength double-sided tape to stick the sensor node to the selected position, ensuring that the connection node is firmly fixed and will not fall off due to wind or vibration.

[0059] ② Clamp-type installation: Special clamps are used to fix the connection nodes to the antenna, ensuring that the connection nodes are in close contact with the antenna surface and reducing signal interference.

[0060] Each accelerometer, gyroscope, magnetometer, strain sensor, temperature sensor, and humidity sensor periodically collects data at a sampling frequency of 100Hz, forming data points (each data point includes a timestamp, acceleration, angular velocity, magnetic field strength, stress change value, temperature, and humidity). This data is transmitted to the central processing unit (CPU) via an NB-IoT wireless communication module. The CPU filters and smooths the raw data to reduce noise interference. The connection nodes have built-in rechargeable lithium batteries and are equipped with solar charging panels to ensure long-term operation.

[0061] (ii) Training and real-time monitoring of deformation prediction models Based on the antenna state parameters and environmental state parameters collected in the above steps, a deformation prediction model is trained, and the trained deformation prediction model is used to predict whether the base station antenna has shifted. If no shift has occurred (i.e., the displacement, tilt angle, and strain are within the preset threshold range), the base station antenna is considered to be in normal condition, and monitoring continues; if the prediction result indicates that the base station antenna has shifted (i.e., exceeded the preset threshold), the next step will be performed.

[0062] (1) Model training ① Data preparation Collect and prepare historical data for training the model. Specifically, collect data for a one-month historical period, including data on normal base station antenna operation and data on known base station antenna variations. Extract features from the historical data, including acceleration, angular velocity, magnetic field strength, stress changes, temperature, and humidity, to form a prepared feature dataset and corresponding labels.

[0063] ② Kernel function selection and hyperparameter tuning Within the vector regression model framework, a suitable kernel function is selected, and the kernel parameters are dynamically adjusted based on the distribution of antenna state parameters and environmental state parameters to adapt to the characteristics of the antenna data.

[0064] In this embodiment, the deformation prediction model uses a Gaussian kernel function (RBF kernel):

[0065] in, This is a kernel function used to compute the results of two input data points. and Similarity in a high-dimensional feature space. and The input data points specifically include acceleration, angular velocity, magnetic field strength, stress change, temperature, and humidity. These two data points represent different sample points and are used to calculate the similarity between them. These are kernel parameters, dynamically adjusted based on the distribution of antenna data, controlling the width of the kernel function. The larger the kernel size, the narrower the kernel width, and the more sensitive the model is to local features. The smaller the kernel size, the wider the kernel function, and the more sensitive the model is to global features.

[0066] Furthermore, the kernel parameters are dynamically adjusted based on the distribution of antenna data. Regularization parameters in the Support Vector Machine (SVM) algorithm Specifically, the optimal method is selected based on the statistical characteristics of the data (such as standard deviation) or through cross-validation. Value and The hyperparameters are adjusted to ensure optimal model performance on the training data while avoiding overfitting. During the process, grid search and cross-validation methods will be used to find the optimal solution. The value will be adjusted again during this process to ensure optimal model performance. value, and They influence each other.

[0067] Furthermore, the steps of grid search involve defining a hyperparameter grid, including different... and Value. The performance of each hyperparameter combination is evaluated using a 5-fold cross-validation method, and the combination with the best performance is selected, thus obtaining the optimal hyperparameters. and .

[0068]

[0069] in, It is a weight vector used to determine the slope of the decision boundary. It is a bias term used to determine the intercept of the decision boundary. It is a regularization parameter that controls the complexity of the model and prevents overfitting. It is a slack variable that allows some data points to be outside the decision boundary, used to handle indivisible cases. It is a feature mapping function that maps input data... (This represents a single input data point, usually a vector) mapped to a high-dimensional feature space. The target value (the label of a single sample, usually a scalar) specifically includes the antenna displacement, tilt angle, and strain. It refers to the number of samples.

[0070] Furthermore, multi-task learning can be introduced to simultaneously predict the antenna's displacement, tilt angle, and strain, thereby improving the model's prediction accuracy. Specifically, a multi-task SVR model is constructed to simultaneously predict the antenna's displacement, tilt angle, and strain.

[0071]

[0072] in, It is a weight matrix used to handle multiple outputs in multi-task learning, where each output corresponds to a task (such as displacement, tilt angle, and strain). It is a bias vector used to handle multiple outputs in multi-task learning, with each output corresponding to a task. It is a regularization parameter that controls the complexity of the model and prevents overfitting. It is a slack variable that allows some data points to be outside the decision boundary, used to handle indivisible cases. It is a feature mapping function that maps input data... (This represents the input data matrix, where each row is a sample, each sample has multiple outputs such as displacement, tilt angle, and strain, and each column is a feature) is mapped to a high-dimensional feature space. It is a multi-task label (representing a multi-task label matrix, where each row is a sample, each sample has multiple outputs such as displacement degree, tilt angle and strain degree, and each column is a label for a task), specifically including the antenna displacement degree, tilt angle and strain degree. It refers to the number of samples.

[0073] (2) Real-time monitoring The system receives sensor data in real time and uses a trained model to predict the displacement, tilt angle, and strain of the base station antenna, thus making a preliminary judgment on the status of the base station antenna.

[0074] ① Use a trained deformation prediction model to predict the displacement, tilt angle, and strain of the base station antenna. Specifically, input the filtered and smoothed data into the trained model, and the model outputs the predicted displacement, tilt angle, and strain.

[0075]

[0076] in, It is the predicted displacement, tilt angle, and strain. It is a weight matrix. It is a feature mapping function. It is the bias vector.

[0077] ②Based on the prediction results, a preliminary assessment of the base station antenna status is made. Set preset thresholds, for example, displacement degree: 5 degrees, tilt angle: 5 degrees, strain degree: 0.01%.

[0078] Normal state: If the predicted displacement, tilt angle and strain are all within the threshold range, the system considers the base station antenna to be in a normal state and continues to monitor it.

[0079] Abnormal state: If the predicted displacement, tilt angle or strain exceeds the threshold, the system assumes that the base station antenna has changed and proceeds to the next step.

[0080] (III) Fuzzy logic system processing, offset estimation and alarm handling (1) Fuzzy reasoning ① Input variable preparation Model prediction results: The predicted displacement, tilt angle, and strain are obtained from the deformation prediction model.

[0081] Sensor noise level: Calculate the standard deviation of the filtered acceleration data.

[0082] Acceleration data source: Acceleration data is obtained from accelerometers installed on the antenna bracket and at the connection point.

[0083] Preprocessing: Low-pass filter and moving average method are used to filter and smooth the acceleration data to reduce noise interference.

[0084] Calculate the standard deviation: Calculate the standard deviation of the filtered acceleration data as an indicator of the sensor noise level.

[0085] Environmental conditions: Acquire data from temperature and humidity sensors.

[0086] The prepared input variables include displacement, tilt angle, strain, sensor noise level, temperature, and humidity.

[0087] ② Blurring Define a suitable membership function for each variable to convert the variable into a fuzzy set.

[0088] Table 1

[0089] Table 2

[0090] ③ Establish fuzzy rules to determine the antenna deformation level based on the fuzzy set of input variables. Specifically, when defining fuzzy rules, a series of fuzzy rules are defined based on experience and professional knowledge, for example: Minor changes: If the displacement is "small", the strain is "small", the sensor noise level is "low", the ambient temperature is "suitable", and the ambient humidity is "suitable", then the deformation level is "slight".

[0091] Moderate changes: If the displacement level is "medium", the strain level is "medium", the sensor noise level is "medium", the ambient temperature is "suitable temperature" or "low temperature", and the ambient humidity is "suitable humidity" or "low humidity", then the deformation level is "medium".

[0092] Major changes: If the displacement is "large", the strain is "large", the sensor noise level is "high", the temperature is "high temperature" or "low temperature", and the humidity is "high humidity" or "low humidity", then the deformation level is "severe".

[0093] ④ Fuzzy reasoning Based on the above fuzzy rules, fuzzy rule matching, fuzzy set operations and aggregation, and defuzzification (such as the centroid method) are performed to convert the fuzzy set into specific deformation levels (such as slight, moderate, severe, etc.).

[0094] (2) Degree of offset estimation If fuzzy inference determines that the deformation level of the base station antenna is moderate or severe, the offset estimation module is activated. By comparing the difference between historical time period data (preferably, but not limited to, the past week) and current data, the offset degree of the antenna is estimated (usually referring to the tilt angle of the base station antenna relative to its initial position; displacement or strain can also be used as indicators of offset degree). Once a significant change in the base station antenna is detected, an alarm will be issued immediately, and detailed maintenance recommendations will be provided to maintenance personnel.

[0095] (3) Alarm handling If the antenna offset reaches a certain standard, an alarm will be issued immediately. That is, if fuzzy reasoning determines that the deformation level of the base station antenna is moderate or severe, and the offset degree is confirmed, such as the offset angle exceeding the preset standard (e.g., 10 degrees), a maintenance alarm will be issued immediately.

[0096] Alarm methods: Notify maintenance personnel via SMS, email, or mobile APP.

[0097] Fault cause and maintenance suggestion push: Based on the degree of change and environmental conditions, possible causes of failure are provided (such as excessive wind, temperature changes, etc.). At the same time, detailed maintenance suggestions (such as recalibrating the antenna, reinforcing the bracket, etc.) can be provided based on the degree of offset and environmental conditions.

[0098] Through the above steps, this application can not only monitor the status of base station antennas in real time and make a preliminary judgment on whether the antennas have changed, but also further estimate the degree of antenna offset when changes occur, issue alarms in a timely manner, and provide detailed maintenance suggestions, thereby achieving low-cost and high-precision antenna change monitoring.

[0099] The present invention also discloses a base station antenna offset monitoring system, which operates based on the base station antenna offset monitoring method described above.

[0100] This invention also discloses another base station antenna offset monitoring system, which includes one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors. The programs include instructions for performing the base station antenna offset monitoring method as described above. The processor may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, used to execute relevant programs to implement the functions required by the modules in the base station antenna offset monitoring system of this application embodiment, or to execute the base station antenna offset monitoring method of this application embodiment.

[0101] This invention also discloses a computer-readable storage medium comprising a computer program executable by a processor to perform the base station antenna offset monitoring method described above. 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 integrating one or more available media. The available medium can be read-only memory (ROM), random access memory (RAM), or magnetic media, such as floppy disks, hard disks, magnetic tapes, magnetic disks, or optical media, such as digital versatile discs (DVDs), or semiconductor media, such as solid-state disks (SSDs).

[0102] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Although the embodiments have been described in the description and drawings of this application, this does not limit the scope of patent protection of this application. Any technical solutions resulting from equivalent structural or procedural substitutions or modifications made based on the essential concept of this application and utilizing the content described in the description and drawings of this application, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. A method for monitoring the offset of a base station antenna, characterized in that, include: Obtain antenna state parameters related to the offset of the base station antenna within a historical time period and environmental state parameters related to the external environment in which the base station antenna is located. Based on the antenna state parameters, environmental state parameters, and support vector regression model framework of the historical time period, a deformation prediction model is constructed. The antenna state parameters and environmental state parameters of the current base station antenna are collected in real time and input into the deformation prediction model to obtain the predicted values ​​of the displacement, tilt angle and strain of the current base station antenna. When any one of the predicted values ​​of displacement, tilt angle and strain exceeds a preset deformation warning threshold, the deformation level of the base station antenna is inferred based on preset fuzzy rules. When the deformation level of the base station antenna reaches or exceeds the preset deformation warning level, the antenna state parameters and environmental state parameters of the current base station antenna are compared with the antenna state parameters and environmental state parameters of the historical time period to estimate the degree of offset of the base station antenna. When the estimated offset of the base station antenna exceeds the preset offset alarm range, a maintenance alarm is issued.

2. The method for monitoring the offset of a base station antenna according to claim 1, characterized in that, The base station antenna includes a bracket and a top antenna. The steps for obtaining the antenna status parameters and the environmental status parameters include: The acceleration at the bottom and middle of the bracket is detected by an accelerometer; The angular velocity and magnetic field strength of the head and sides of the top antenna are detected by the gyroscope and the magnetometer. The acceleration and stress changes at the connection point between the top antenna and the bracket are detected using accelerometers and strain sensors. The ambient temperature and humidity of the external environment where the base station antenna is located are detected by temperature and humidity sensors. The antenna state parameters include the acceleration at the bottom and middle of the bracket, the angular velocity and magnetic field strength at the head and sides of the top antenna, the acceleration and stress change at the connection point between the top antenna and the bracket, and the environmental state parameters include the ambient temperature and humidity of the external environment where the base station antenna is located.

3. The method for monitoring the offset of a base station antenna according to claim 1, characterized in that, The steps for constructing the deformation prediction model include: The antenna state parameters and the environmental state parameters are filtered and preprocessed to obtain a feature dataset; The support vector regression model framework is initialized based on the feature dataset; The feature dataset is input into the initialized support vector regression model framework for training, thereby obtaining the deformation prediction model.

4. The method for monitoring the offset of a base station antenna according to claim 3, characterized in that, The filtering preprocessing steps include: Use a low-pass filter to remove high-frequency noise from the antenna state parameters and the environmental state parameters; The antenna state parameters and the environmental state parameters are smoothed using the moving average method.

5. The method for monitoring the offset of a base station antenna according to claim 3, characterized in that, The initialization steps of the support vector regression model framework include: The kernel function of the support vector regression model framework is selected based on the distribution of the feature dataset, and the hyperparameters of the kernel function are adjusted by grid search and cross-validation methods.

6. The method for monitoring the offset of a base station antenna according to claim 1, characterized in that, Before inputting the antenna state parameters and environmental state parameters of the current base station antenna collected in real time into the deformation prediction model, filtering preprocessing is required. The filtering preprocessing steps include: Use a low-pass filter to remove high-frequency noise from the antenna state parameters and the environmental state parameters; The antenna state parameters and the environmental state parameters are smoothed using the moving average method.

7. The method for monitoring the offset of a base station antenna according to claim 6, characterized in that, The step of reasoning the deformation level of the base station antenna based on preset fuzzy rules includes: Obtain the antenna state parameters and environmental state parameters of the current base station antenna after filtering and preprocessing, and calculate the standard deviation of the antenna state parameters; For the predicted displacement degree, tilt angle, and strain degree, the standard deviation of the antenna state parameters of the current base station antenna after filtering preprocessing, and the environmental state parameters, a membership function is configured to obtain a fuzzy set; Based on the fuzzy rules and the fuzzy set, fuzzy reasoning is performed and defuzzification is applied to obtain the deformation level of the base station antenna.

8. A base station antenna offset monitoring system, characterized in that, The base station antenna offset monitoring system operates based on the base station antenna offset monitoring method according to any one of claims 1 to 7.

9. A base station antenna offset monitoring system, characterized in that, include: One or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including instructions for performing the offset monitoring method for a base station antenna as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Includes a computer program, which can be executed by a processor to perform the base station antenna offset monitoring method as described in any one of claims 1 to 7.

Images