Communication control method and system based on dual-mode communication module

Abstract

The invention relates to the technical field of communication control, and discloses a communication control method and system based on a dual-mode communication module, the system integrates two communication modes of 4G and 5G, and flexible and reliable real-time data transmission is realized. The core is that a communication quality evaluation module is introduced, and 4G and 5G communication quality is scored by analyzing data such as signal strength, data transmission rate and packet loss rate. The system constructs a fault prediction module, and predicts the type and probability of future communication faults by using historical communication quality scores and equipment operation state data. And the maintenance early warning module carries out early warning on the maintenance demand of the communication equipment in combination with the evaluation and prediction results. And the maintenance scheme generation module automatically generates a maintenance scheme according to the early warning result, triggers an emergency notification when the maintenance demand is high and provides detailed steps. According to the invention, the stability and reliability of communication service are improved, and the maintenance process of communication equipment is optimized.

Description

Technical Field

[0001] The present invention relates to the field of communication control technology, and in particular to a communication control method and system based on a dual-mode communication module. Background Art

[0002] With the rapid development of communication technology, 4G and 5G communication modes have become the mainstream technologies in the current mobile communication field. 4G technology occupies an important position in the mobile communication market with its stable performance and wide coverage; while 5G technology, with its advantages of ultra-high speed, low latency and large number of connections, is gradually pushing mobile communications to a higher level. In practical applications, due to the influence of various factors such as network environment, device status, user behavior, etc., a single communication mode is often difficult to meet the communication needs in all scenarios. Therefore, the dual-mode communication module came into being, which can support both 4G and 5G communication modes at the same time to achieve more flexible and reliable communication services.

[0003] In the existing communication control system, only the data transmission function of the communication module is focused on, but there is a lack of comprehensive evaluation of communication quality and the ability to predict faults. As a result, when problems such as weak signals, slow data transmission rates, or high packet loss rates occur during the communication process, the system is unable to timely and accurately evaluate the communication quality and take corresponding measures to optimize it. The system also lacks effective prediction methods for possible faults in communication equipment, and cannot provide early warning and maintenance before the fault occurs, thus affecting the stability and reliability of communication services. Summary of the invention

[0004] The object of the present invention is to provide a communication control method and system based on a dual-mode communication module to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a communication control system based on a dual-mode communication module, the system comprising:

[0006] Dual-mode communication module, used to support both 4G and 5G communication modes and transmit data in real time;

[0007] A communication quality assessment module is used to construct a communication quality assessment model. The input data of the communication quality assessment model is the signal strength data, data transmission rate data and packet loss rate data during the transmission process of the dual-mode communication module. The output data is the communication quality score, including the 4G communication quality score and the 5G communication quality score. The model training uses a data set containing historical communication data and corresponding quality scores, and the corresponding labels are various communication quality scores.

[0008] A fault prediction module builds a fault prediction model, the input data of which are historical communication quality scores and equipment operation status data, and the output data are the predicted communication fault types and their occurrence probabilities that may occur in the future. Model training uses a time series data set containing historical communication quality scores and equipment operation status, and the corresponding labels are future communication fault types and probabilities.

[0009] A maintenance warning module is used to construct a maintenance warning model. The input data of the maintenance warning model are the evaluation results of the communication quality evaluation model and the prediction results of the fault prediction model. The maintenance demand of the communication equipment is warned by combining the information of the two. The model training uses a combined data set containing the communication quality evaluation results and the fault prediction results, and the corresponding label is the maintenance demand level of the communication equipment.

[0010] The maintenance plan generation module is connected to the maintenance warning module and generates a corresponding communication equipment maintenance plan based on the warning result. If the warning result is a high maintenance demand, an emergency maintenance notification is triggered and detailed maintenance steps are provided.

[0011] Preferably, it is characterized in that the communication quality assessment model is constructed by using a decision tree algorithm, and the characteristics of the input data are divided by the nodes of the decision tree, and its network structure includes:

[0012] Input layer: receives preprocessed communication data, and the data format is unified into a multi-dimensional feature vector form;

[0013] Decision node layer: According to the preset division criteria, the various features of the input data are judged and divided to form different branch paths;

[0014] Leaf node layer: Each leaf node corresponds to a communication quality score interval, and the input data is finally divided into the corresponding leaf node to determine its communication quality score;

[0015] Output layer: Output 4G communication quality score and 5G communication quality score.

[0016] Preferably, the fault prediction model is constructed using a recurrent neural network (RNN) algorithm, and the temporal features in the time series data are captured by RNN units. The network structure includes:

[0017] Input layer: receives historical communication quality scores and equipment operation status data;

[0018] RNN layer: contains multiple RNN units, each unit processes a time step in the sequence data and passes the hidden state to the next unit;

[0019] Fully connected layer: The hidden state output by the RNN layer is mapped to the predicted communication failure type and its probability of occurrence through the fully connected layer;

[0020] Output layer: Output the types of communication failures that may occur in the future and their probability of occurrence.

[0021] Preferably, the maintenance warning model separates data of different maintenance demand levels by constructing a decision boundary, and the construction steps include:

[0022] A1: The communication quality assessment results and fault prediction results are used as input features, and the communication equipment maintenance demand level is used as the target variable;

[0023] A2: Select the polynomial kernel function;

[0024] A3: Use the support vector machine algorithm to train the model and find the optimal decision boundary;

[0025] A4: For new input data, determine its maintenance requirement level based on its position in the feature space relative to the decision boundary.

[0026] Preferably, the optimization step of the maintenance warning model includes:

[0027] B1: Adjust the parameters of the support vector machine, including the penalty parameter C and the kernel function parameters;

[0028] B2: Use grid search combined with cross-validation to evaluate model performance and select the optimal parameter combination;

[0029] B3: Use the validation set to validate the optimized model and evaluate the accuracy and recall of the model;

[0030] B4: When the model performance reaches the preset standard, the model parameters are saved to obtain the trained maintenance warning model.

[0031] Preferably, the implementation method of the maintenance plan generation module includes:

[0032] Predefine a variety of communication equipment maintenance plans, each plan corresponds to different maintenance demand levels and fault types. The plan content covers the level of maintenance notification, deployment of maintenance personnel and specific maintenance operation steps;

[0033] Receive the warning results from the maintenance warning module, and match the corresponding maintenance plan in the plan library according to the maintenance requirement level and fault type in the warning results;

[0034] According to the maintenance plan matched by the decision engine, specific communication equipment maintenance operations are performed, including replacing communication modules, optimizing network configuration, performing equipment debugging and implementing system upgrades.

[0035] Preferably, the specific method for constructing the solution library is:

[0036] A hash table is set as a data structure for storing maintenance plans, where each key contains a unique identifier, which is used to represent a specific combination of maintenance requirement level and fault type; the value part contains a text string, which represents a specific maintenance plan; the hash table uses an open address method or a chain address method to handle conflicts; the hash table structure is defined as: H = {(k,v)|k∈{maintenance requirement level_fault type},v∈{maintenance plan}}; where k represents the combination of maintenance requirement level and fault type, and v represents the corresponding maintenance plan.

[0037] Preferably, the maintenance plan generation module also includes a key matcher, which has a built-in hash function; using the hash function, the key matcher quickly calculates the hash value of the input maintenance requirement level and fault type combination, and then searches the hash table for a maintenance plan that matches the hash value.

[0038] Preferably, the maintenance plan generation module also includes a plan editor, which provides an editing function for the user to add, modify or delete keys and values ​​in the hash table according to actual equipment conditions and technical updates.

[0039] Preferably, a communication control method based on a dual-mode communication module comprises:

[0040] S1: Supports both 4G and 5G communication modes through dual-mode communication module, and transmits data in real time;

[0041] S2: Construct a communication quality assessment model, whose input data is the signal strength data, data transmission rate data and packet loss rate data during the transmission of the dual-mode communication module, and the output data is the communication quality score, including the 4G communication quality score and the 5G communication quality score; use a data set containing historical communication data and corresponding quality scores to train the model, and the corresponding labels are various communication quality scores;

[0042] S3: Build a fault prediction model, whose input data is historical communication quality scores and equipment operation status data, and whose output data is the predicted communication fault types and their occurrence probabilities in the future. Use a time series dataset containing historical communication quality scores and equipment operation status to train the model, and the corresponding labels are future communication fault types and probabilities.

[0043] S4: construct a maintenance warning model, whose input data is the evaluation results of the communication quality evaluation model and the prediction results of the fault prediction model, and combine the information of the two to warn the maintenance needs of the communication equipment; use the combined data set containing the communication quality evaluation results and the fault prediction results to train the model, and the corresponding label is the maintenance demand level of the communication equipment;

[0044] S5: Connect the maintenance warning module and the maintenance plan generation module, generate a corresponding communication equipment maintenance plan according to the warning result of the maintenance warning module, and if the warning result is a high maintenance demand, trigger an emergency maintenance notification and provide detailed maintenance steps.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] A communication quality assessment module is built to collect key data such as signal strength, data transmission rate, and packet loss rate of the dual-mode communication module during transmission in real time, and the communication quality assessment model is used to conduct a comprehensive analysis of these data to obtain the quality scores of the 4G and 5G communication modes. This comprehensive assessment method can more accurately reflect the actual communication quality and provide a strong basis for subsequent optimization and adjustment. By building a fault prediction model, combined with historical communication quality scores and equipment operating status data, the types of communication faults that may occur in the future and their probability of occurrence are predicted. This prediction capability enables the system to provide early warning and maintenance before a fault occurs, effectively avoiding interruptions or degradation of communication services due to faults, and improving the stability and reliability of communication services.

[0047] The maintenance warning module combines the communication quality assessment results and fault prediction results to warn of the maintenance needs of communication equipment. This warning mechanism enables maintenance personnel to understand the maintenance needs of the equipment in a timely manner, so as to formulate maintenance plans in advance, reduce the occurrence of sudden failures, and improve maintenance efficiency. The maintenance plan generation module can generate corresponding communication equipment maintenance plans based on the warning results, provide maintenance personnel with detailed maintenance steps, and further improve the accuracy and efficiency of maintenance. The dual-mode communication module supports both 4G and 5G communication modes, realizing more flexible and reliable communication services. In different scenarios, the system can select the optimal communication mode according to the actual situation to ensure the continuity and stability of communication services. Through comprehensive communication quality assessment and fault prediction, the system can detect and deal with potential problems in a timely manner, further enhancing the reliability and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a working principle diagram of the communication control system of the present invention;

[0049] Figure 2 Schematic diagram of the working principle of the method for maintenance early warning model construction and optimization;

[0050] Figure 3 Diagram of the steps to train an intent recognition model. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1-3 The present invention provides a technical solution: a communication control system based on a dual-mode communication module, the system comprising:

[0053] Dual-mode communication module: It is the core component of this system. It is designed to support both 4G and 5G communication modes to achieve real-time data transmission. The module can automatically select the optimal communication mode for data transmission according to the current network environment and device status to ensure the stability and reliability of communication services.

[0054] Communication quality assessment module: responsible for building a communication quality assessment model to evaluate the communication quality of the dual-mode communication module during the transmission process. The input data of the model includes signal strength data, data transmission rate data, and packet loss rate data during the transmission process of the dual-mode communication module. Through comprehensive analysis of these data, the model can output 4G communication quality scores and 5G communication quality scores, thereby fully reflecting the current communication quality status. In order to train the communication quality assessment model, a data set containing historical communication data and corresponding quality scores is used. Each record in the data set contains signal strength, data transmission rate, packet loss rate, and corresponding communication quality score. By training these data with a machine learning algorithm, the model can learn the mapping relationship between signal strength, data transmission rate, and packet loss rate and communication quality score, thereby achieving accurate evaluation of communication quality.

[0055] Fault prediction module: responsible for building a fault prediction model to predict the types of faults that may occur in communication equipment in the future and their probability of occurrence. The input data of the model includes historical communication quality scores and equipment operating status data. Through the time series analysis of these data, the model can output the types of communication faults that may occur in the future and their probability of occurrence, providing early warning for equipment maintenance. In order to train the fault prediction model, a time series data set containing historical communication quality scores and equipment operating status is used. Each record in the data set contains the communication quality score and equipment operating status at a specific time point, as well as the types of faults that will occur in the future and their probability of occurrence. By training these data with a time series analysis algorithm, the model can learn the time series relationship between the communication quality score and equipment operating status and the future fault types and their probability of occurrence, thereby achieving accurate prediction of faults.

[0056] Maintenance warning module: responsible for building a maintenance warning model to warn of maintenance needs of communication equipment by combining communication quality assessment results and fault prediction results. The input data of this model includes the assessment results of the communication quality assessment model and the prediction results of the fault prediction model. Through comprehensive analysis of these data, the model can output the maintenance requirement level of communication equipment and provide decision support for equipment maintenance. In order to train the maintenance warning model, a combined data set containing communication quality assessment results and fault prediction results is used. Each record in the data set contains the communication quality assessment results, fault prediction results, and the corresponding maintenance requirement level. By training these data with a machine learning algorithm, the model can learn the mapping relationship between the communication quality assessment results and fault prediction results and the maintenance requirement level, thereby achieving accurate warning of maintenance needs.

[0057] Maintenance plan generation module: connected to the maintenance warning module, responsible for generating corresponding communication equipment maintenance plans based on the warning results. If the warning result is a high maintenance demand, an emergency maintenance notification will be triggered, and detailed maintenance steps and operation guides will be provided so that maintenance personnel can respond quickly and handle the fault problem.

[0058] The present invention will be further described below in conjunction with Examples 1 to 4:

[0059] Embodiment 1:

[0060] The communication quality evaluation model is constructed by using a decision tree algorithm and is used to evaluate the communication quality of the dual-mode communication module during transmission. The network structure includes an input layer, a decision node layer, a leaf node layer and an output layer.

[0061] (1) Input layer: Receives preprocessed communication data in a unified format of multi-dimensional feature vectors. These feature vectors include key indicators such as signal strength, data transmission rate, and packet loss rate, which are input into the model after preprocessing.

[0062] (2) Decision node layer: According to the preset division criteria, the various features of the input data are judged and divided to form different branch paths. For example, for the feature of signal strength, a threshold can be set to divide the data with signal strength higher than the threshold into one branch, and the data with signal strength lower than the threshold into another branch. In this way, through the judgment and division of multiple decision nodes, the input data is gradually guided to different leaf nodes.

[0063] (3) Leaf node layer: Each leaf node corresponds to a communication quality score interval. When input data is divided into a leaf node, it means that its communication quality score falls within the score interval corresponding to the leaf node. In this way, the communication quality score of the input data can be determined by dividing the leaf nodes.

[0064] (4) Output layer: Output 4G communication quality score and 5G communication quality score. According to the leaf nodes to which the input data is divided, the model can output the corresponding 4G or 5G communication quality score, thereby comprehensively reflecting the current communication quality status.

[0065] Suppose there is a set of preprocessed communication data, whose feature vectors include signal strength, data transmission rate and packet loss rate. This set of data is input into the communication quality assessment model, and after judgment and division at the decision node layer, it is finally divided into a leaf node. Assuming that the communication quality score interval corresponding to the leaf node is [80,90], the model will output the 4G or 5G communication quality score of this set of data as 85 (or other values ​​within this interval, depending on the implementation details of the model).

[0066] By adopting the decision tree algorithm to construct a communication quality evaluation model, the present invention can achieve accurate evaluation of communication quality. The model can guide the data to the corresponding leaf node through the judgment and division of the decision node according to the feature vector of the input data, so as to determine its communication quality score. This solves the problem of lack of comprehensive evaluation of communication quality in the existing communication control system and improves the stability and reliability of communication services.

[0067] The fault prediction model is constructed using a recurrent neural network (RNN) algorithm to predict the types of faults that may occur in communication equipment in the future and their probability of occurrence. The network structure includes an input layer, an RNN layer, a fully connected layer, and an output layer.

[0068] (1) Input layer: Receives historical communication quality scores and equipment operating status data. These data are organized into time series and input into the model.

[0069] (2) RNN layer: It contains multiple RNN units, each of which processes a time step in the sequence data. RNN units can capture the temporal features in the time series data and pass the hidden state to the next unit. In this way, through the processing of the RNN layer, the model can learn the temporal relationship between the historical communication quality score and the equipment operation status and the future fault type and its probability of occurrence.

[0070] (3) Fully connected layer: The hidden state output by the RNN layer is mapped to the predicted communication fault type and its probability of occurrence through the fully connected layer. The function of the fully connected layer is to map the high-dimensional feature vector output by the RNN layer to the low-dimensional fault type and probability space, thereby facilitating subsequent output and processing.

[0071] (4) Output layer: Output the types of communication failures that may occur in the future and their probability of occurrence. Based on the output of the fully connected layer, the model can predict the types of failures that may occur in communication equipment in the future and the probability of occurrence of each failure.

[0072] By using the RNN algorithm to construct a fault prediction model, the present invention can achieve accurate prediction of communication equipment faults. The model can predict the types of faults that may occur in the future and their probability of occurrence by learning the temporal relationship between historical communication quality scores and equipment operating status data and future fault types and their probability of occurrence. This solves the problem of lack of effective prediction of communication equipment faults in existing communication control systems and improves equipment maintenance efficiency and service quality.

[0073] Embodiment 2:

[0074] The steps of constructing the maintenance early warning model include:

[0075] Step A1: Select communication quality assessment results and fault prediction results as input features of the model. Communication quality assessment results may include but are not limited to indicators such as signal strength, signal-to-noise ratio, bit error rate, etc.; fault prediction results may come from historical fault records of the equipment, operating status monitoring data, etc., and extract features with predictive value for maintenance needs through preprocessing and feature engineering. The maintenance requirement level of the communication equipment is used as the target variable of the model. The maintenance requirement level can be divided according to the actual maintenance needs of the equipment and the degree of business impact, such as emergency maintenance, planned maintenance, routine inspection, etc.

[0076] Step A2: Considering the complexity and nonlinear characteristics of communication equipment maintenance data, a polynomial kernel function is selected as the kernel function of the support vector machine (SVM). The polynomial kernel function can capture high-order nonlinear relationships in the data and help improve the model's fitting ability and generalization performance.

[0077] Step A3: Use the support vector machine algorithm to train the prepared data. During the training process, the optimal decision boundary is found by solving the optimization problem, which can maximize the interval between data of different maintenance demand levels, so as to achieve accurate classification of the data.

[0078] Step A4: For new input data, first map it into the feature space, and then determine the maintenance requirement level it belongs to based on the relationship between its position in the feature space and the decision boundary. Specifically, the maintenance requirement level can be determined by calculating the distance from the data point to the decision boundary or using the value of the decision function.

[0079] The optimization methods for maintaining the early warning model include:

[0080] Step B1: Adjust the parameters of the support vector machine, including the penalty parameter C and the kernel function parameter. The penalty parameter C is used to control the degree of penalty for training errors. The larger the C value, the tighter the model fits the training data, but it may lead to overfitting. The kernel function parameter affects the mapping method of the feature space and the nonlinear ability of the model.

[0081] Step B2: Use grid search combined with cross-validation to evaluate model performance under different parameter combinations. Grid search traverses the preset parameter range to find the parameter combination that optimizes model performance; cross-validation is used to evaluate the performance of the model on independent data sets to ensure the generalization ability of the model.

[0082] Step B3: Use the validation set to validate the optimized model and evaluate the accuracy and recall of the model. The accuracy reflects the correctness of the model's judgment of the maintenance requirement level, while the recall reflects the model's ability to identify specific maintenance requirement level data. By comprehensively analyzing the accuracy and recall, the performance of the model can be fully evaluated.

[0083] Step B4: When the model performance reaches the preset standard, the model parameters are saved to obtain the trained maintenance warning model. The saved model parameters include the weight vector, bias term, kernel function parameters, etc. of the support vector machine, which constitute a complete description of the model and can be used for subsequent prediction and maintenance decision support.

[0084] Embodiment 3:

[0085] In order to efficiently and accurately generate a maintenance plan for a communication device, this embodiment describes in detail a method for implementing a maintenance plan generation module, and the specific steps are as follows:

[0086] 1. Pre-definition of maintenance plan

[0087] Solution library construction: Predefine and organize a variety of communication equipment maintenance solutions, each of which clearly corresponds to a specific maintenance requirement level and fault type. The maintenance requirement level can be divided based on factors such as the operating status of the equipment and the degree of business impact, such as emergency, high priority, medium priority, low priority, etc. The fault types cover various faults that may occur in communication equipment, such as hardware failure, software failure, network configuration errors, etc. The content of the solution is detailed and comprehensive, including but not limited to the level of maintenance notification (such as emergency notification, general notification), the deployment of maintenance personnel (such as required skills, number of people), and specific maintenance operation steps (such as specific steps for replacing communication modules, parameter settings for network configuration optimization, etc.).

[0088] Solution storage and management: All predefined maintenance solutions are stored in a solution library, which can be a database, file system or other suitable storage medium. An effective management mechanism is established to ensure the integrity and accessibility of the solution library.

[0089] 2. Receiving warning results and matching with solutions

[0090] Receive early warning results: The maintenance plan generation module is closely integrated with the maintenance early warning module, and receives early warning results issued by the early warning module in real time. The early warning results contain key information such as maintenance requirement level and fault type.

[0091] Solution matching: According to the maintenance requirement level and fault type in the early warning results, the corresponding maintenance solution is automatically matched in the solution library. The matching process can be based on predefined rules or algorithms to ensure that the most suitable maintenance solution is found quickly and accurately.

[0092] 3. Maintenance plan execution

[0093] Decision engine call: The maintenance plan generation module calls the decision engine and takes the matched maintenance plan as input. The decision engine is responsible for generating specific execution instructions based on the content of the maintenance plan.

[0094] Generation and issuance of execution instructions: The decision engine generates detailed execution instructions based on the content of the maintenance plan, including instructions for the deployment of maintenance personnel, the execution sequence of maintenance operation steps, a list of required tools or materials, etc. The execution instructions are issued to relevant maintenance personnel or departments through appropriate communication channels (such as email, SMS, internal system notifications, etc.).

[0095] Maintenance operation execution: Maintenance personnel perform maintenance operations on communication equipment according to the specific steps in the maintenance plan based on the received execution instructions. For example, when replacing a communication module, maintenance personnel need to disassemble, install and test the module according to the steps in the plan; when optimizing network configuration, specific network parameters need to be adjusted to improve network performance.

[0096] Maintenance result feedback and record: After the maintenance operation is completed, the maintenance personnel will feedback the maintenance results to the maintenance plan generation module. The maintenance results include whether the maintenance is successful, whether additional problems are encountered, whether further operations are required, etc. The maintenance plan generation module records the maintenance results for subsequent maintenance effect evaluation and plan optimization.

[0097] Embodiment 4:

[0098] This embodiment is used to describe in detail a specific implementation method using a hash table as a storage structure, and the functional design of a key matcher and a solution editor in a maintenance solution generation module.

[0099] Use hash table to build solution library:

[0100] Hash table structure design: Set the hash table H as the data structure for storing maintenance plans, where each element is a key-value pair (k, v). Key k: Contains a unique identifier, which is composed of a combination of the maintenance requirement level and the fault type. For example, key k can be "Emergency_Hardware Failure", "High Priority_Software Failure", etc., to uniquely represent a specific maintenance requirement level and fault type combination. Value v: Contains a text string that represents a specific maintenance plan. The maintenance plan describes in detail the level of maintenance notification, the deployment of maintenance personnel, specific maintenance operation steps, etc.

[0101] Hash function and conflict handling: The hash table uses a hash function to map the key k to a certain position in the hash table. The design of the hash function should ensure the uniform distribution of keys to reduce the occurrence of conflicts. When a conflict occurs, the hash table uses the open address method or the chain address method to handle it. The open address method stores the conflicting key-value pairs by looking for the next free position; the chain address method maintains a linked list in each hash table position and links the conflicting key-value pairs in the linked list.

[0102] Hash table structure definition: Hash table H is defined as: H = {(k,v)|k∈{maintenance requirement level_fault type},v∈{maintenance plan}}. Where k represents the combination of maintenance requirement level and fault type, and v represents the corresponding maintenance plan.

[0103] The maintenance plan generation module includes a key matcher, which has a built-in hash function. The hash function is the same as the hash function used when constructing the hash table to ensure that the hash value of the input maintenance requirement level and fault type combination can be correctly calculated.

[0104] When receiving the warning result from the maintenance warning module, the key matcher extracts the combination of maintenance requirement level and fault type in the warning result and calculates the hash value of the combination using the built-in hash function. Based on the calculated hash value, the key matcher searches the hash table for the maintenance plan that matches the hash value and provides it as output to the decision engine for subsequent processing.

[0105] The maintenance plan generation module also includes a plan editor, which provides an editing function for users to add, modify or delete keys and values ​​in the hash table according to actual equipment conditions and technical updates.

[0106] The user can enter a new maintenance requirement level and fault type combination as the key k and the corresponding maintenance plan as the value v through the plan editor and add it to the hash table.

[0107] The user can select an existing key k through the plan editor and modify its corresponding value v to update the content of the maintenance plan.

[0108] The user can select an existing key k through the scheme editor and delete it from the hash table to remove maintenance schemes that are no longer needed.

[0109] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0110] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A communication control system based on a dual-mode communication module, characterized in that: The system comprises: Dual-mode communication module, used to support both 4G and 5G communication modes and transmit data in real time; A communication quality assessment module is used to construct a communication quality assessment model. The input data of the communication quality assessment model is the signal strength data, data transmission rate data and packet loss rate data during the transmission process of the dual-mode communication module. The output data is the communication quality score, including the 4G communication quality score and the 5G communication quality score. The model training uses a data set containing historical communication data and corresponding quality scores, and the corresponding labels are various communication quality scores. A fault prediction module builds a fault prediction model, the input data of which are historical communication quality scores and equipment operation status data, and the output data are the predicted communication fault types and their occurrence probabilities that may occur in the future. Model training uses a time series data set containing historical communication quality scores and equipment operation status, and the corresponding labels are future communication fault types and probabilities. A maintenance warning module is used to construct a maintenance warning model. The input data of the maintenance warning model are the evaluation results of the communication quality evaluation model and the prediction results of the fault prediction model. The maintenance demand of the communication equipment is warned by combining the information of the two. The model training uses a combined data set containing the communication quality evaluation results and the fault prediction results, and the corresponding label is the maintenance demand level of the communication equipment. The maintenance plan generation module is connected to the maintenance warning module and generates a corresponding communication equipment maintenance plan based on the warning result. If the warning result is a high maintenance demand, an emergency maintenance notification is triggered and detailed maintenance steps are provided.

2. The communication control system based on the dual-mode communication module according to claim 1, characterized in that: The communication quality assessment model is constructed using a decision tree algorithm, and the characteristics of the input data are divided by the nodes of the decision tree. Its network structure includes: Input layer: receives preprocessed communication data, and the data format is unified into a multi-dimensional feature vector form; Decision node layer: According to the preset division criteria, the various features of the input data are judged and divided to form different branch paths; Leaf node layer: Each leaf node corresponds to a communication quality score interval, and the input data is finally divided into the corresponding leaf node to determine its communication quality score; Output layer: Output 4G communication quality score and 5G communication quality score.

3. The communication control system based on the dual-mode communication module according to claim 1, characterized in that: The fault prediction model is constructed using a recurrent neural network (RNN) algorithm, and captures the temporal features in time series data through RNN units. Its network structure includes: Input layer: receives historical communication quality scores and equipment operation status data; RNN layer: contains multiple RNN units, each unit processes a time step in the sequence data and passes the hidden state to the next unit; Fully connected layer: The hidden state output by the RNN layer is mapped to the predicted communication failure type and its probability of occurrence through the fully connected layer; Output layer: Output the types of communication failures that may occur in the future and their probability of occurrence.

4. The communication control system based on the dual-mode communication module according to claim 1, characterized in that: The maintenance warning model separates data of different maintenance demand levels by constructing a decision boundary, and the construction steps include: A1: The communication quality assessment results and fault prediction results are used as input features, and the communication equipment maintenance demand level is used as the target variable; A2: Select the polynomial kernel function; A3: Use the support vector machine algorithm to train the model and find the optimal decision boundary; A4: For new input data, determine its maintenance requirement level based on its position in the feature space relative to the decision boundary.

5. The communication control system based on the dual-mode communication module according to claim 4, characterized in that: The optimization steps of the maintenance warning model include: B1: Adjust the parameters of the support vector machine, including the penalty parameter C and the kernel function parameters; B2: Use grid search combined with cross-validation to evaluate model performance and select the optimal parameter combination; B3: Use the validation set to validate the optimized model and evaluate the accuracy and recall of the model; B4: When the model performance reaches the preset standard, the model parameters are saved to obtain the trained maintenance warning model.

6. The communication control system based on the dual-mode communication module according to claim 1, characterized in that: The implementation method of the maintenance plan generation module includes: Predefine a variety of communication equipment maintenance plans, each plan corresponds to different maintenance demand levels and fault types. The plan content covers the level of maintenance notification, deployment of maintenance personnel and specific maintenance operation steps; Receive the warning results from the maintenance warning module, and match the corresponding maintenance plan in the plan library according to the maintenance requirement level and fault type in the warning results; According to the maintenance plan matched by the decision engine, specific communication equipment maintenance operations are performed, including replacing communication modules, optimizing network configuration, performing equipment debugging and implementing system upgrades.

7. The communication control system based on the dual-mode communication module according to claim 6, characterized in that: The specific method of building a solution library is: A hash table is set as a data structure for storing maintenance plans, where each key contains a unique identifier, which is used to represent a specific combination of maintenance requirement level and fault type; the value part contains a text string, which represents a specific maintenance plan; the hash table uses an open address method or a chain address method to handle conflicts; the hash table structure is defined as: H = {(k,v)|k∈{maintenance requirement level_fault type},v∈{maintenance plan}}; where k represents the combination of maintenance requirement level and fault type, and v represents the corresponding maintenance plan.

8. The communication control system based on the dual-mode communication module according to claim 7, characterized in that: The maintenance plan generation module also includes a key matcher, which has a built-in hash function. Using the hash function, the key matcher quickly calculates the hash value of the input maintenance requirement level and fault type combination, and then searches the hash table for a maintenance plan that matches the hash value.

9. The communication control system based on the dual-mode communication module according to claim 8, characterized in that: The maintenance plan generation module also includes a plan editor, which provides an editing function for users to add, modify or delete keys and values ​​in the hash table according to actual equipment conditions and technical updates.

10. A communication control method based on a dual-mode communication module, characterized in that: The method comprises: S1: Supports both 4G and 5G communication modes through dual-mode communication module, and transmits data in real time; S2: Construct a communication quality assessment model, whose input data is the signal strength data, data transmission rate data and packet loss rate data during the transmission of the dual-mode communication module, and the output data is the communication quality score, including the 4G communication quality score and the 5G communication quality score; use a data set containing historical communication data and corresponding quality scores to train the model, and the corresponding labels are various communication quality scores; S3: Build a fault prediction model, whose input data is historical communication quality scores and equipment operation status data, and whose output data is the predicted communication fault types and their occurrence probabilities in the future. Use a time series dataset containing historical communication quality scores and equipment operation status to train the model, and the corresponding labels are future communication fault types and probabilities. S4: construct a maintenance warning model, whose input data is the evaluation results of the communication quality evaluation model and the prediction results of the fault prediction model, and combine the information of the two to warn the maintenance needs of the communication equipment; use the combined data set containing the communication quality evaluation results and the fault prediction results to train the model, and the corresponding label is the maintenance demand level of the communication equipment; S5: Connect the maintenance warning module and the maintenance plan generation module, generate a corresponding communication equipment maintenance plan according to the warning result of the maintenance warning module, and if the warning result is a high maintenance demand, trigger an emergency maintenance notification and provide detailed maintenance steps.

Images