Communication base station operation management system based on industrial internet of things
Through the communication base station operation management system based on the Industrial Internet of Things, combined with multimodal sensors, edge computing and intelligent operation and maintenance decisions, the insufficient monitoring and security problems of traditional base station management systems are solved, and efficient, safe and reliable operation and maintenance management of base station equipment is achieved.
Patent Information
- Application Number
- CN202510491493.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-18
- Publication Date
- 2025-07-08
AI Technical Summary
Traditional communication base station management systems cannot fully monitor the status of equipment and are vulnerable to network attacks. Operation and maintenance rely on manual experience, resulting in untimely and costly fault handling, making it difficult to meet the efficient management needs of modern communication networks.
The communication base station operation management system based on the industrial Internet of Things is adopted, including the perception layer, network layer and management layer. The multi-modal sensor fusion module, the edge computing node real-time analysis module and the intelligent operation and maintenance decision-making module are used to realize real-time monitoring, rapid diagnosis and early warning of base station equipment, combine quantum encryption and blockchain technology to ensure data security, and use LSTM algorithm to model and predict equipment status.
It realizes comprehensive real-time monitoring and rapid fault diagnosis of base station equipment, improves operation and maintenance efficiency, reduces failure rate and operation and maintenance costs, ensures the continuity and stability of communication services, and realizes the transformation from passive maintenance to active maintenance.
Smart Images
Figure CN120282189A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of communication management, and specifically refers to a communication base station operation management system based on the industrial Internet of Things. Background Art
[0002] With the rapid development of communication technologies, communication base stations, as the key infrastructure of wireless communication networks, are increasing in number and becoming more widely distributed. However, the traditional methods for operating and managing communication base stations face many challenges.
[0003] Early management systems mainly relied on a small number of common sensors, such as temperature and humidity sensors, power sensors, etc., and could only obtain some operating parameters of the base station equipment. It was difficult to monitor information related to potential equipment failures, such as vibrations caused by loose equipment components, leakage of harmful gases, etc., and it was impossible to comprehensively grasp the operating status of the base station equipment; during data transmission, conventional communication networks and encryption methods were mostly used in the past, which were vulnerable to hacker attacks and network eavesdropping, resulting in data leakage or tampering, seriously affecting the safe operation of communication base stations and the security of user information; traditional operation and maintenance decisions mostly relied on manual experience, lacking effective analysis and utilization of the massive base station operation data. When faced with complex equipment failures and abnormal operations, it was impossible to predict and judge in a timely and accurate manner, and often only dealt with the problems after the failures occurred, resulting in communication service interruptions, reducing communication quality and user experience, and at the same time increasing operation and maintenance costs; the manual inspection and maintenance methods were not only inefficient but also unable to achieve real-time monitoring and remote control of base station equipment. For widely distributed base stations, operation and maintenance personnel needed to spend a lot of time and effort on on-site inspections and operations, making it difficult to meet the requirements of modern communication networks for efficient management of base stations. Summary of the Invention
[0004] In view of the above situation, to overcome the defects of the prior art, the present invention provides a communication base station operation management system based on the industrial Internet of Things to solve the above-mentioned technical defects.
[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A communication base station operation management system based on the industrial Internet of Things includes a perception layer, a network layer, and a management layer, and the perception layer is connected to the management layer through the network layer;
[0006] Among them, the network layer constructs a hybrid communication network with 5G as the backbone and Wi-Fi and wired networks as supplements;
[0007] The perception layer includes:
[0008] The multimodal sensor fusion module deploys various types of sensors inside the base station. Using multimodal sensor fusion technology, it fuses and preprocesses the data collected by various sensors. The various types of sensors include vibration sensors, gas sensors, and image sensors;
[0009] The edge computing node real-time analysis module deploys edge computing nodes at the perception layer to perform preliminary processing and analysis on the data collected by sensors. By running lightweight machine learning algorithms, it realizes rapid diagnosis and early warning of equipment failures at the edge;
[0010] The management layer includes:
[0011] The intelligent operation and maintenance decision-making module builds an intelligent operation and maintenance decision-making platform based on big data analysis and artificial intelligence technology. According to the preset remaining service life of the equipment combined with the preset maintenance threshold, it arranges the maintenance plan in advance;
[0012] The warning terminal is used to perform corresponding information pop-up warning operations on the received abnormal signals.
[0013] Further, in the multimodal sensor fusion module, the vibration sensor works based on the piezoelectric effect and converts the vibration signal into an electrical signal V(t). The specific process is as follows: When the equipment vibrates, the electric charge Q generated by the piezoelectric material is related to the vibration acceleration a. Through the formula Q = k·a (k is the piezoelectric constant), it is equivalent to a circuit composed of a capacitor C and a resistor R inside the vibration sensor. According to the capacitance definition U = Q / C, the change in electric charge is converted into a change in voltage, and the electrical signal V(t) is obtained.
[0014] Further, in the multimodal sensor fusion module, the relationship between the detection resistance value Rg of the gas sensor and the different gas concentrations C_env is Rg = R0(1 + αC) (R0 is the initial resistance, and α is a constant related to the gas and the sensor material). By detecting the change in the resistance value Rg, the gas concentration C(t) in the current environment is calculated using the formula C = (Rg - R0) / (R0·α).
[0015] Further, in the multimodal sensor fusion module, the image data collected by the image sensor is represented by the matrix I(x, y), where x and y are the coordinates of the image pixels. The method for obtaining its feature data F(t) includes:
[0016] Calculate the gradient values Gx and Gy of the matrix I(x, y) in the x direction and y direction, and calculate the gradient amplitude according to the formula. The gradient amplitude is used as the image edge feature data;
[0017] Calculate the partial derivatives Ix and Iy of the matrix I(x, y) in the x and y directions. For each pixel point (x, y) in the matrix I(x, y), construct a local autocorrelation matrix M. According to the eigenvalues of the autocorrelation matrix M, calculate the corner response function R. If the corner response function R of a pixel point is greater than a preset threshold, mark this point as a corner, and use the corner information as the image corner feature data;
[0018] The feature data F(t) of the image sensor is jointly composed of the image edge feature data and the image corner feature data.
[0019] Furthermore, the multimodal sensor fusion module uses a weighted fusion algorithm to fuse the data collected by different sensors at the same time t. Let the weight of the vibration sensor data be w1, the weight of the gas sensor data be w2, and the weight of the feature data extracted after processing the image sensor data be w3, and w1 + w2 + w3 = 1. The calculation formula for the fused data D(t) is D(t) = w1V(t) + w2C(t) + w3F(t).
[0020] Furthermore, in the edge computing node real-time analysis module, let the efficiency of the edge computing node in processing data be, the processor performance parameter be P, the memory capacity be M, and the complexity of the software algorithm be Ca.
[0021] Furthermore, in the edge computing node real-time analysis module, set the abnormal operation threshold T of the communication base station. If D(t) ≥ T, it means that the communication base station is operating abnormally, otherwise it is normal; within the time period [t1, t2], if the amount of data Q that the edge computing node can process and the amount of data Dtotal generated by D(t) satisfy D total ≤ η × (t2 - t1), then the edge computing node has sufficient processing capacity and generates an efficient processing signal. If D total > η × (t2 - t1), then a delayed processing signal is generated, and the efficient processing signal and the delayed processing signal are sent to the management layer through the network layer. The management layer performs a pop-up prompt alarm processing on the delayed processing signal.
[0022] Furthermore, in the intelligent operation and maintenance decision-making module, collect the historical fault data of the base station equipment. The data format includes the specific time of the fault occurrence, the fault type, the snapshot of the equipment operation state parameters at the time of the fault occurrence, the fault handling process and results, and real-time collect the equipment operation parameters, including the equipment voltage, current, and signal strength.
[0023] Furthermore, the intelligent operation and maintenance decision-making module uses the Long Short-Term Memory (LSTM) algorithm in deep learning to model and predict the operating status of base station equipment. It integrates historical fault data and equipment operating parameters according to the time series to form an input sequence. The LSTM model learns and processes the input sequence through its internal memory units and gating structures, mines the time series features and potential rules in the data, and realizes the modeling of the equipment operating status and the prediction of the remaining useful life of the equipment.
[0024] Furthermore, the intelligent operation and maintenance decision-making module optimizes the network structure and training parameters of the LSTM algorithm by analyzing the dimension and complexity of the data. In terms of the network structure, according to the characteristics of base station equipment data and prediction requirements, it adjusts the number of layers of the LSTM network, the number of neurons in each layer, and the connection method between neurons in each layer; in terms of training parameters, it optimizes the learning rate and the number of iterations to improve the accuracy and reliability of the prediction of the remaining useful life of the equipment.
[0025] Advantages of the present invention:
[0026] 1. Through the multi-modal sensor fusion module in the perception layer, a variety of types of sensors are deployed to achieve comprehensive real-time monitoring of the operating status of base station equipment and environmental parameters. It can not only monitor common parameters such as temperature, humidity, and power, but also obtain information such as equipment vibration, harmful gas concentration, and equipment appearance through vibration sensors, gas sensors, and image sensors, providing rich data support for accurately grasping the operating conditions of the base station and timely discovering potential fault hazards.
[0027] 2. In the present invention, the edge computing node real-time analysis module is used to preliminarily process and analyze sensor data at the edge, and runs lightweight machine learning algorithms, which can quickly diagnose equipment faults and give early warnings in a timely manner. Compared with the traditional centralized data processing method, it reduces data transmission delay, improves the fault response speed, enables operation and maintenance personnel to take measures before or at the initial stage of the fault, and reduces the impact of the fault on communication services.
[0028] 3. In the present invention, the intelligent operation and maintenance decision-making module of the management layer is based on big data analysis and artificial intelligence technology. It uses the LSTM algorithm to model and predict the operating status of base station equipment, and improves the accuracy and reliability of predicting the remaining service life of the equipment by optimizing the network structure and training parameters of the algorithm. According to the prediction results, the maintenance plan is arranged in advance to realize the transformation from passive maintenance to active maintenance, reduce the equipment failure rate, reduce the operation and maintenance cost, and improve the operation efficiency and reliability of the communication base station; the system can perform operation and maintenance processing on the abnormal operation of the communication base station in a timely manner according to the processing results of the edge computing node. For the situation of data processing delay, an alarm is prompted through a pop-up window at the management layer, enabling the operation and maintenance personnel to respond quickly and avoiding the expansion of communication failures caused by untimely handling of abnormal situations, thus ensuring the continuity and stability of communication services. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The present invention will be further described below with reference to the accompanying drawings.
[0030] Figure 1 It is a principle block diagram of the perception layer, network layer and management layer of the embodiment of the present invention;
[0031] Figure 2 It is a principle block diagram of the operation management of the communication base station based on the industrial Internet of Things in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0032] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited by the exemplary embodiments described herein.
[0033] Embodiment 1
[0034] Please refer to Figure 1 and Figure 2 As shown, the operation management system of the communication base station based on the industrial Internet of Things includes: a perception layer, a network layer and a management layer. The perception layer is connected to the management layer through the network layer, and the network layer constructs a hybrid communication network with 5G as the backbone and Wi-Fi and wired networks as supplements.
[0035] It should be noted that the network layer constructs a hybrid communication network with 5G as the backbone and Wi-Fi and wired networks as supplements, and uses quantum encryption technology and blockchain technology to ensure the high-speed, stable and secure data transmission. The 5G network meets the high-real-time data transmission requirements, Wi-Fi and wired networks provide flexible access methods and stable backups, quantum encryption technology prevents data from being stolen or tampered with, and blockchain technology ensures the integrity and traceability of data, effectively protecting the data security of the communication base station.
[0036] Further, it should be noted that the perception layer includes a multi-modal sensor fusion module and a real-time analysis module for edge computing nodes.
[0037] The multi-modal sensor fusion module deploys various types of sensors inside the base station. Using multi-modal sensor fusion technology, it fuses and preprocesses the data collected by various sensors.
[0038] Specifically, the method of fusing and preprocessing the data collected by various sensors is as follows:
[0039] Denote the vibration signal output by the vibration sensor as V(t), where t represents time. It works based on the piezoelectric effect. When the device vibrates, the electric charge Q generated by the piezoelectric material is related to the vibration acceleration a. Through the formula Q = k·a, where k represents the piezoelectric constant. Utilizing the characteristics of the capacitor, the change in electric charge is converted into a change in voltage, thereby obtaining the electrical signal output V(t). It should be noted that after the electric charge Q is generated, there is a circuit structure inside the vibration sensor, which can generally be equivalent to a circuit composed of a capacitor C and a resistor R. According to the definition of the capacitor, the voltage U across the capacitor is U = Q / C. Here, the voltage U is equivalent to the output electrical signal V(t). By loading the generated electric charge Q onto such a circuit structure and utilizing the characteristics of the capacitor, the change in electric charge is converted into a change in voltage, thereby obtaining the electrical signal output V(t).
[0040] Denote the detected resistance value of the gas sensor as Rg. In an environment with different gas concentrations C, the resistance value follows the relationship Rg = R0(1 + αC), where R0 is the initial resistance and α is a constant related to the gas and the sensor material. By detecting the change in the resistance value Rg, the gas concentration C(t) in the current environment is calculated through the formula C = (Rg - R0) / (R0·α).
[0041] Represent the image data collected by the image sensor with the matrix I(x, y), where both x and y represent the coordinates of the image pixels, and each pixel point corresponds to a grayscale value. Calculate the gradient values Gx and Gy of the matrix I(x, y) in the x direction and the y direction respectively. According to the formula Calculate the gradient magnitude, and use the gradient magnitude as the image edge feature data. Calculate the partial derivatives Ix and Iy of the matrix I(x, y) in the x direction and the y direction respectively. For each pixel point (x, y) in the matrix I(x, y), construct a local autocorrelation matrix M, and its calculation formula is where "*" represents the convolution operation, ω is a Gaussian window function used to weight the local area. According to the eigenvalues λ1 and λ2 of the autocorrelation matrix M, calculate the corner response function R, and its calculation formula is R = det(M) - k(tr(M)) 2, where det(M) = λ1λ2 is the determinant of matrix M, tr(M) = λ1 + λ2 is the trace of matrix M, k is an empirical constant with a value range of 0.04 - 0.06; if the corner response function R of a pixel point is greater than a preset threshold, then this point is marked as a corner, and the corner information is used as the image corner feature data; the feature data F(t) of the image sensor is jointly composed of the image edge feature data and the image corner feature data.
[0042] The data collected by different sensors at the same moment t is processed using a weighted fusion algorithm. Let the weight of the vibration sensor data be w1, the weight of the gas sensor data be w2, and the weight of the feature data extracted from the processed image sensor data be w3, and w1 + w2 + w3 = 1. The fused data is denoted as D(t), and its calculation formula is D(t) = w1V(t) + w2C(t) + w3F(t).
[0043] In a specific embodiment, in the present invention, through the multi-modal sensor fusion module of the perception layer, multiple types of sensors are deployed to achieve comprehensive real-time monitoring of the operating status of base station equipment and environmental parameters. It can not only monitor common parameters such as temperature, humidity, and power, but also obtain information such as equipment vibration, harmful gas concentration, and equipment appearance through vibration sensors, gas sensors, and image sensors, providing rich data support for accurately grasping the operating conditions of the base station and promptly discovering potential fault hazards.
[0044] The edge computing node real-time analysis module, by deploying edge computing nodes in the perception layer, preliminarily processes and analyzes the data collected by sensors. Through running lightweight machine learning algorithms, it realizes rapid diagnosis and early warning of equipment failures at the edge.
[0045] The edge computing node real-time analysis and processing module is as follows:
[0046] Let the data processing efficiency of the edge computing node be η, denote the processor performance parameter as P, the memory capacity as M, and the complexity of the software algorithm as Ca. According to the formula Calculate the data processing efficiency η of the edge computing node, where k1 is a constant related to the hardware architecture and algorithm optimization.
[0047] Set the abnormal operation threshold T of the communication base station. If D(t) ≥ T, it indicates that the communication base station is operating abnormally; otherwise, it indicates that the communication base station is operating normally.
[0048] Obtain the data volume Q that the edge computing node can process within the time period [t1, t2]. According to the definition of efficiency, we get Q = η×(t2 - t1). Obtain the data volume Dtotal generated by D(t) within the time period [t1, t2]. If Dtotal ≤ Q, that is, D total≤η×(t2 - t1), indicating that the edge computing node has sufficient processing power to process the data generated by D(t) in a timely manner, can quickly analyze D(t) and determine whether it meets the abnormal condition D(t)≥T, and generate an efficient processing signal.
[0049] On the contrary, if D total >η×(t2 - t1), indicating that the speed of data generation exceeds the processing capacity of the edge computing node, and the computing node cannot process the data generated by D(t) in a timely and complete manner, affecting the timeliness of judging the abnormal situation of the communication base station and generating a delayed processing signal.
[0050] Send the generated efficient processing signal and delayed processing signal to the management layer through the network layer. Through the management layer, perform a pop-up prompt alarm processing on the delayed processing signal, and perform operation and maintenance processing on the abnormal operation of the communication base station in a timely manner.
[0051] In a specific embodiment, in the present invention, the edge computing node real-time analysis module is used to perform preliminary processing and analysis on sensor data at the edge, run lightweight machine learning algorithms, and can quickly diagnose equipment failures and give early warnings. Compared with the traditional centralized data processing method, it reduces data transmission delay, improves the fault response speed, enables operation and maintenance personnel to take measures before or at the initial stage of the fault, and reduces the impact of the fault on communication services.
[0052] The management layer includes an intelligent operation and maintenance decision-making module and a warning terminal. The intelligent operation and maintenance decision-making module arranges a maintenance plan in advance according to the preset remaining service life of the equipment and the preset maintenance threshold; the warning terminal is used to perform corresponding information pop-up warning operations on the received abnormal signals.
[0053] The specific analysis process of the intelligent operation and maintenance decision-making module is as follows:
[0054] Build an intelligent operation and maintenance decision-making platform based on big data analysis and artificial intelligence technology, collect historical fault data of base station equipment, and the data format includes the specific time of fault occurrence, fault type, snapshot of the operating state parameters of the equipment at the time of fault, fault handling process and results, and collect equipment operating parameters in real time, including equipment voltage, current and signal strength.
[0055] Use the long short-term memory network (LSTM) algorithm in the deep learning algorithm to model and predict the operating state of the base station equipment, and integrate and process the historical fault data and equipment operating parameters according to the time series to form an input sequence; specifically: arrange parameters such as equipment voltage, current, and signal strength in the past period of time (such as the past 1 hour, 1 day, etc., which can be adjusted according to actual needs) and the fault information that occurred during this period in chronological order to form a multi-dimensional data vector as the input of the LSTM model.
[0056] The LSTM model contains multiple memory units and gating structures inside. Through these structures, the input sequence is learned and processed to mine the time series features and potential rules in the data. It should be noted that during the operation of the model, the weight matrix and bias vector will be continuously adjusted according to the input data to adapt to the changes and characteristics of the data, so as to realize the modeling of the device operation state and the prediction of the remaining service life of the device.
[0057] Optimize the network structure and training parameters of the LSTM algorithm by analyzing the dimension and complexity of the data. It should be noted that in terms of the network structure, according to the characteristics and prediction requirements of the base station device data, the number of layers of the LSTM network and the number of neurons in each layer are adjusted and optimized. For example, the number of network layers is increased or decreased, and the connection method between neurons in each layer is adjusted to improve the learning ability and expression ability of the model. In terms of training parameters, parameters such as the learning rate and the number of iterations are optimized. Through experiments and data analysis, the optimal learning rate is found, so that the model can converge to the optimal solution faster during the training process. At the same time, the number of iterations is reasonably set to avoid overfitting or underfitting of the model, thereby improving the accuracy and reliability of the prediction of the remaining service life of the device.
[0058] Finally, the platform arranges the maintenance plan in advance according to the result of the remaining service life of the device predicted by the LSTM algorithm, combined with the pre-set maintenance threshold. The specific manifestation is as follows: when the remaining service life of the device is lower than a certain specific value, such as 1 month, 3 months, etc., it can be adjusted according to factors such as the importance of the device and the maintenance cost. For devices whose predicted remaining service life is close to or lower than the threshold, the system will automatically generate maintenance tasks, including information such as maintenance time, maintenance personnel arrangement, and preparation of required parts. The maintenance plan will be displayed to the operation and maintenance management personnel through the system interface. The management personnel can adjust and confirm according to the actual situation, and then perform maintenance operations according to the plan to ensure the stable operation of the device and reduce the probability of failures.
[0059] In a specific embodiment, the intelligent operation and maintenance decision-making module of the management layer in the present invention is based on big data analysis and artificial intelligence technology, uses the LSTM algorithm to model and predict the operation state of the base station device, and improves the accuracy and reliability of the prediction of the remaining service life of the device by optimizing the network structure and training parameters of the algorithm. According to the prediction results, the maintenance plan is arranged in advance to realize the transformation from passive maintenance to active maintenance, reduce the device failure rate, reduce the operation and maintenance cost, and improve the operation efficiency and reliability of the communication base station. The system can timely perform operation and maintenance processing on the abnormal operation situation of the communication base station according to the processing results of the edge computing nodes. For the situation of data processing delay, through the pop-up window prompt alarm of the management layer, the operation and maintenance personnel can respond quickly to avoid the expansion of communication failures caused by untimely handling of abnormal situations, and ensure the continuity and stability of communication services.
[0060] The above formulas are all dimensionless and only take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulations to get a formula closest to the actual situation. The magnitude of the coefficient is a specific value obtained by quantifying each parameter. Regarding the magnitude of the coefficient, as long as it does not affect the proportional relationship between the parameters and the quantified values, it is acceptable.
[0061] In addition, those skilled in the art can understand that various aspects of the present invention can be described and illustrated by several patentable types or situations, including any new and useful processes, machines, products, or combinations of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present invention can be executed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software can all be referred to as "data blocks", "modules", "engines", "units", "components", or "systems". In addition, various aspects of the present invention may be embodied as a computer product located in one or more computer-readable media, which includes computer-readable program code.
[0062] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those of ordinary skill in the art to which this invention belongs. It should also be understood that terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense, unless explicitly defined as such herein.
[0063] The above is the description of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, those skilled in the art will readily understand that many modifications can be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined by the claims. It should be understood that the above is the description of the present invention and should not be considered as limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.
Claims
1. A communication base station operation management system based on the industrial Internet of Things, characterized in that, It includes a perception layer, a network layer, and a management layer. The perception layer is connected to the management layer through the network layer; Among them, the network layer constructs a hybrid communication network with 5G as the backbone and Wi-Fi and wired networks as supplements; The perception layer includes: A multi-modal sensor fusion module. By deploying various types of sensors inside the base station and using multi-modal sensor fusion technology, the data collected by various sensors are fused and pre-processed. The various types of sensors include vibration sensors, gas sensors, and image sensors; An edge computing node real-time analysis module. By deploying edge computing nodes in the perception layer, the data collected by sensors are preliminarily processed and analyzed. By running lightweight machine learning algorithms, rapid diagnosis and early warning of equipment failures are achieved at the edge; The management layer includes: An intelligent operation and maintenance decision-making module. An intelligent operation and maintenance decision-making platform is built based on big data analysis and artificial intelligence technology. According to the preset remaining service life of the equipment and the preset maintenance threshold, the maintenance plan is arranged in advance; An early warning terminal, which is used to perform corresponding information pop-up warning operations on the received abnormal signals.
2. The communication base station operation management system based on the industrial Internet of Things according to claim 1, wherein In the multi-modal sensor fusion module, the vibration sensor works based on the piezoelectric effect and converts the vibration signal into an electrical signal V(t). The specific process is as follows: When the equipment vibrates, the electric charge Q generated by the piezoelectric material is related to the vibration acceleration a. Through the formula Q = k·a (k is the piezoelectric constant), it is equivalent to a circuit composed of a capacitor C and a resistor R inside the vibration sensor. According to the capacitance definition U = Q / C, the change in electric charge is converted into a change in voltage, and the electrical signal V(t) is output.
3. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that In the multi-modal sensor fusion module, the relationship between the detected resistance value Rg of the gas sensor and the concentration C of different gases in the environment is Rg = R0(1 + αC) (R0 is the initial resistance, and α is a constant related to the gas and the sensor material). By detecting the change in the resistance value Rg, the gas concentration C(t) in the current environment is calculated using the formula C = (Rg - R0) / (R0·α).
4. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that, In the multi-modal sensor fusion module, the image data collected by the image sensor is represented by the matrix I(x, y), where x and y are the coordinates of the image pixels, and the method for obtaining its feature data F(t) includes: Calculating the gradient values Gx and Gy of the matrix I(x, y) in the x direction and the y direction, and calculating the gradient amplitude according to the formula. The gradient amplitude is used as the image edge feature data; Calculating the partial derivatives Ix and Iy of the matrix I(x, y) in the x direction and the y direction. For each pixel point (x, y) in the matrix I(x, y), a local autocorrelation matrix M is constructed. According to the eigenvalues of the autocorrelation matrix M, the corner response function R is calculated. If the corner response function R of the pixel point is greater than the preset threshold, the point is marked as a corner, and the corner information is used as the image corner feature data; The feature data F(t) of the image sensor is jointly composed of the image edge feature data and the image corner feature data.
5. The communication base station operation management system based on the industrial Internet of Things according to claim 1, wherein The multi-modal sensor fusion module uses a weighted fusion algorithm to fuse the data collected by different sensors at the same moment t. Let the weight of the vibration sensor data be w1, the weight of the gas sensor data be w2, and the weight of the feature data extracted after processing the image sensor data be w3, and w1 + w2 + w3 = 1. The calculation formula for the fused data D(t) is D(t) = w1V(t) + w2C(t) + w3F(t).
6. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that In the real-time analysis module of the edge computing node, let the efficiency of the edge computing node in processing data be, the processor performance parameter be P, the memory capacity be M, and the complexity of the software algorithm be Ca.
7. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that In the real-time analysis module of the edge computing node, an abnormal operation threshold T of the communication base station is set. If D(t)≥T, it indicates that the communication base station has an abnormal operation; otherwise, it is normal. During the time period [t1, t2], if the data volume Q that the edge computing node can process and the data volume Dtotal generated by D(t) satisfy D total ≤η×(t2 - t1), the edge computing node has sufficient processing capacity and generates an efficient processing signal. If D total >η×(t2 - t1), a delayed processing signal is generated, and the efficient processing signal and the delayed processing signal are sent to the management layer through the network layer. The management layer performs a pop-up prompt alarm processing on the delayed processing signal.
8. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that In the intelligent operation and maintenance decision-making module, historical fault data of the base station equipment is collected. The data format includes the specific time of the fault occurrence, the fault type, the snapshot of the operating state parameters of the equipment at the time of the fault occurrence, the fault handling process and results, and the operating parameters of the equipment are collected in real time, including the equipment voltage, current, and signal strength.
9. The communication base station operation management system based on the industrial Internet of Things according to claim 1, characterized in that The intelligent operation and maintenance decision-making module uses the long short-term memory network (LSTM) algorithm in the deep learning algorithm to model and predict the operating state of the base station equipment. The historical fault data and the equipment operating parameters are integrated and processed according to the time series to form an input sequence. The LSTM model learns and processes the input sequence through its internal memory units and gating structures, mines the time series features and potential laws in the data, and realizes the modeling of the equipment operating state and the prediction of the remaining service life of the equipment.
10. The communication base station operation management system based on the industrial Internet of Things according to claim 9, characterized in that, The intelligent operation and maintenance decision-making module optimizes the network structure and training parameters of the LSTM algorithm by analyzing the dimension and complexity of the data. In terms of the network structure, according to the characteristics and prediction requirements of the base station equipment data, the number of layers of the LSTM network, the number of neurons in each layer, and the connection method between the neurons in each layer are adjusted; in terms of the training parameters, the learning rate and the number of iterations are optimized to improve the accuracy and reliability of the prediction of the remaining service life of the equipment.