Tablet Computer Intelligent Management System and Method Based on 5G Transmission

Abstract

The present invention relates to the technical field of tablet computers, specifically to a tablet computer intelligent management system and method based on 5G transmission. The system includes a data collection and screening module, an intelligent fault analysis module, a response and configuration module, and a maintenance evaluation and optimization module. The data collection and screening module, based on the 5G network, performs parameter collection and collects data transmission rate and network latency information uploaded by the tablet computer. In the present invention, by using the random forest and decision tree algorithms, it is possible to effectively identify and classify the normal and abnormal behavior patterns of the device, achieve precise positioning of key fault points, not only shorten the fault diagnosis time, but also predict potential problems and respond quickly, significantly reducing the device downtime. The automated response strategy reduces the user's operation burden, and through continuous performance optimization, extends the device service life and improves the overall user experience. The system ensures service continuity and device stability through continuous maintenance efficiency evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of tablet computers, and particularly to an intelligent management system and method for tablet computers based on 5G transmission. Background Art

[0002] Tablet computers emphasize portability, touch interfaces, and efficient wireless communication capabilities. The devices usually come with large screens, a thin and light design, and integrate various sensors such as accelerometers, gyroscopes, and GPS, enabling them to adapt to various interactions and environments. Tablet computers support a wide range of applications, from office automation to entertainment, from education to remote control, and can also achieve high-speed data transmission through wireless technologies such as Wi-Fi, Bluetooth, and 4G / 5G.

[0003] Among them, the intelligent management system for tablet computers aims to improve the efficiency and intelligence level of device management. By integrating various management tools and applications, users can better monitor and control various functions and resources of the tablet computer. Its uses include application management, data security protection, device performance monitoring, and automatic updates. Usually, data analysis and machine learning technologies are utilized to predict device problems and optimize performance to provide a personalized usage experience.

[0004] Existing systems only start to react after a failure occurs, rather than performing preventive management, resulting in device crashes for users at critical moments, affecting work continuity and productivity. Existing technologies fail to effectively utilize the advantages of high-speed data transmission, and there will be delays when processing complex data sets, limiting their ability to quickly adjust resource allocation in emergency situations. The lack of real-time optimization and predictive maintenance also makes the device prone to performance degradation during long-term operation, increasing the device's failure and usage risks. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an intelligent management system and method for tablet computers based on 5G transmission.

[0006] To achieve the above objective, the present invention adopts the following technical solution: An intelligent management system for tablet computers based on 5G transmission, the system includes:

[0007] The data acquisition and screening module is based on the 5G network, performs parameter acquisition, collects the data transmission rate and network latency information uploaded by the tablet computer, summarizes the battery power and application crash logs, screens the key information in the data set, and generates a preliminary diagnosis data set;

[0008] The intelligent fault analysis module is based on the preliminary diagnosis data set, integrates associated time series data, classifies the data using the random forest algorithm, distinguishes normal and abnormal behavior patterns, identifies the key fault points of the tablet computer, and obtains fault identification indicators;

[0009] Based on the fault identification metrics, the response and configuration module analyzes the urgency of the fault using a decision tree algorithm, determines the response process, and configures the resource allocation of the tablet computer, including automatically restarting applications or adjusting performance allocation, to generate detailed response strategy rules.

[0010] Based on the detailed response strategy rules, the maintenance evaluation and optimization module records the effects of maintenance activities, sends the records to the maintenance database via the 5G network, and evaluates the success rate and efficiency of maintenance activities to generate maintenance effectiveness records.

[0011] The improvement of the present invention is that the specific steps for collecting the data transmission rate and network latency information are as follows:

[0012] Based on the 5G network, collect the data transmission rate and network latency of the tablet computer, and use the formula:

[0013]

[0014] and

[0015] ND t =L t

[0016] Obtain the data transmission rate DTR per second t and the network latency ND t , where S t represents the amount of data uploaded in the t-th second, T t represents the time required to upload S t , and L t represents the network latency time;

[0017] Integrate the data transmission rate and network latency, and smooth the integrated data, using the formula:

[0018]

[0019] and

[0020]

[0021] Obtain the smoothed data transmission rate and the network latency where w i represents the weight at the i-th second, DTR i and ND i represent the data transmission rate and network latency at the t-th second, and n is the size of the time window.

[0022] The improvement of the present invention is that the specific steps for screening the key information are as follows:

[0023] Summarize the battery power data and application crash logs, calculate the average battery power and the occurrence frequency of application crashes, using the formula:

[0024]

[0025] and

[0026]

[0027] to obtain the average battery power E avg and the application crash frequency F crash , where E i represents the i-th battery power reading, N is the total number of readings, T is the observation time interval, and C total is the total number of crashes during the observation period;

[0028] Based on the average battery power and the application crash frequency, perform data cleaning, remove invalid or incorrect data, and filter key information in the dataset to obtain a preliminary diagnosis dataset.

[0029] The improvement of the present invention is that the step of classifying the data is specifically as follows:

[0030] Based on the preliminary diagnosis dataset, collect and organize time series data, battery power, and network latency information, using the formula:

[0031]

[0032] to obtain the standardized feature matrix Xd norm , where Xd represents the original data point, mean(Xd) is the average value of the data, and std(Xd) is the standard deviation of the data;

[0033] Based on the standardized feature matrix, use the random forest algorithm for data classification to optimize the accuracy of the classification result, using the formula:

[0034]

[0035] to obtain the classification result YX, where N d is the number of decision trees, Xt i is the classification function of the i-th tree, and Xd norm is the standardized feature matrix.

[0036] The improvement of the present invention is that the step of obtaining the fault identification index is specifically as follows:

[0037] Analyze the generated abnormal category data, including extracting key features and patterns, and calculate the contribution rate and correlation of each feature, using the formula:

[0038]

[0039] Obtain a feature set IScore associated with abnormal behavior, where F i represents a single feature, and Gain(F i ) represents the contribution of feature F i to the improvement of the model performance when constructing a decision tree. TGain is the total contribution of the feature, and N s is the total number of features;

[0040] Based on the feature set associated with abnormal behavior, set a threshold to distinguish normal and abnormal states, using the formula:

[0041] Threshold = μ index + k s ·σ index

[0042] Obtain a fault identification index, where μ index is the historical average of the selected fault index, σ index is the historical standard deviation, k s is a coefficient used to adjust the threshold sensitivity, and Threshold is the threshold used to distinguish normal and abnormal states.

[0043] The improvement of the present invention is that the analysis step of the urgency of the fault is specifically:

[0044] Based on the fault identification index, initialize a decision tree model, distinguish faults of different types, and calculate the urgency score of each type, using the formula:

[0045] U initial = a c ·F frequency + b c ·F impact

[0046] Obtain a preliminary urgency score U initial , where F frequency and F impact represent the fault occurrence frequency and impact degree respectively, and a c and b c are parameter influence weights;

[0047] Based on the preliminary urgency score, combine the fault historical data and potential development trend, using the formula:

[0048] U final = U initial + c f ·H history + d f ·P potential

[0049] Obtain the emergency level score U of the fault final , where H history and P potential represent historical fault data and potential impacts respectively, and c f and d f are the corresponding adjustment coefficients, and U initial is the preliminary emergency score.

[0050] The improvement of the present invention is that the obtaining step of the response strategy details is specifically as follows:

[0051] Evaluate the matching degree of the resource demand according to the current total resource amount and the emergency fault, using the formula:

[0052]

[0053] Obtain the resource matching degree result R match , where C r represents the current CPU resource amount, M r represents the current memory resource amount, D r represents the CPU demand at the time of the fault, N r represents the memory demand at the time of the fault, and α g and β g are adjustment coefficients for adjusting the resource demand according to the current load;

[0054] Based on the resource matching degree result, determine the resource adjustment configuration, using the judgment formula:

[0055]

[0056] Obtain the response strategy details Adjust Type , where R match is the resource matching degree result, and γ o , δ o and ζ o are adjustment factors.

[0057] The improvement of the present invention is that the obtaining step of the maintenance efficiency record is specifically as follows:

[0058] Record the effect of the maintenance activity based on the response strategy details, using the formula:

[0059]

[0060] Obtain the maintenance efficiency score RJ, where lc i is the completion degree of activity i, le i is the effect score of activity i, lv i is the weight of activity i, and ln is the total number of maintenance activities;

[0061] Based on the maintenance efficiency score, it is sent to the maintenance database through the 5G network, and the rate and integrity of data transmission are verified to obtain a record of successful data reception;

[0062] Based on the record of successful data reception, calculate and analyze the success rate of maintenance activities, using the formula:

[0063]

[0064] Obtain the success rate and efficiency SR of maintenance activities, where Nl s is the number of successful maintenance activities, and Nl t is the total number of maintenance activities.

[0065] A tablet intelligent management method based on 5G transmission, the tablet intelligent management method based on 5G transmission is executed based on the above-mentioned tablet intelligent management system based on 5G transmission, and includes the following steps:

[0066] S1: Based on the 5G network, receive the data transmission rate and network latency information of the tablet, summarize the battery power and application crash logs, and screen the key information in the status data to generate a preliminary diagnosis dataset;

[0067] S2: Based on the preliminary diagnosis dataset, integrate the associated time series data, identify normal and abnormal behavior patterns, locate the key fault points of the tablet, and use the fault mode with index classification differences to obtain fault identification indicators;

[0068] S3: According to the fault identification indicators, analyze the urgency and scope of influence of the fault, configure response processes for high-risk fault points, perform delayed processing on low-risk faults, and configure the resource allocation of the tablet to generate response strategy details;

[0069] S4: Based on the response strategy details, record the effect of response execution, upload the data of maintenance activities to the maintenance database through the 5G network, and extract key performance data to generate a maintenance efficiency record;

[0070] S5: Use the maintenance efficiency record to evaluate the effect of maintenance activities, analyze the success rate and efficiency of various tablet maintenance activities, identify potential areas for optimization, and generate identification optimization results.

[0071] Compared with the prior art, the advantages and positive effects of the present invention are:

[0072] In the present invention, by utilizing the random forest and decision tree algorithms, the normal and abnormal behavior patterns of the device can be effectively identified and classified, achieving the precise positioning of key fault points. This not only shortens the fault diagnosis time but also predicts potential problems and responds promptly, significantly reducing the device downtime. Moreover, the automated response strategy reduces the user's operation burden, and through continuous performance optimization, the device service life is extended, enhancing the overall user experience. The system ensures service continuity and device stability through continuous maintenance efficiency evaluation and self-optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 FIG. is a module diagram of the intelligent management system for a tablet computer based on 5G transmission proposed by the present invention;

[0074] Figure 2 FIG. is a flowchart for collecting data transmission rate and network delay information in the present invention;

[0075] Figure 3 FIG. is a flowchart for screening key information in the present invention;

[0076] Figure 4 FIG. is a flowchart for classifying and processing data in the present invention;

[0077] Figure 5 FIG. is a flowchart for obtaining fault identification indicators in the present invention;

[0078] Figure 6 FIG. is a flowchart for analyzing the urgency of faults in the present invention;

[0079] Figure 7 FIG. is a flowchart for obtaining the response strategy details in the present invention;

[0080] Figure 8 FIG. is a flowchart for obtaining maintenance efficiency records in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0081] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0082] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the present invention. In addition, in the description of the present invention, the meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0083] Embodiment

[0084] Please refer to Figure 1 , the present invention provides a technical solution: The intelligent management system for tablets based on 5G transmission includes:

[0085] The data acquisition and screening module, based on the 5G network, performs parameter acquisition, collects the data transmission rate and network latency information uploaded by the tablet, summarizes the battery power and application crash logs, screens the key information in the dataset, and generates a preliminary diagnosis dataset;

[0086] The intelligent fault analysis module, based on the preliminary diagnosis dataset, integrates the associated time series data, classifies the data using the random forest algorithm, distinguishes normal and abnormal behavior patterns, identifies the key fault points of the tablet, and obtains fault identification indicators;

[0087] The response and configuration module, based on the fault identification indicators, analyzes the urgency of the fault using the decision tree algorithm, determines the response process, and configures the resource allocation of the tablet, including automatically restarting applications or adjusting performance allocation, and generates detailed response strategy rules;

[0088] The maintenance evaluation and optimization module, based on the detailed response strategy rules, records the effects of maintenance activities, sends the records to the maintenance database through the 5G network, and evaluates the success rate and efficiency of maintenance activities, and generates maintenance effectiveness records.

[0089] The preliminary diagnosis dataset includes data transmission efficiency, network stability indicators, battery status logs, and application stability analysis logs. The fault identification indicators include abnormal behavior types, key fault locations, and abnormal degree levels. The detailed response strategy rules include emergency response levels, resource reallocation plans, and performance optimization plans. The maintenance effectiveness records include maintenance operation success rates, maintenance efficiency evaluation results, and communication efficiency records.

[0090] Please refer to Figure 2 , the specific steps for collecting the data transmission rate and network latency information are as follows:

[0091] Based on the 5G network, collect the data transfer rate and network latency of the tablet computer, and use the formula:

[0092]

[0093] and

[0094] ND t = L t

[0095] Obtain the data transfer rate DTR per second t and network latency ND t , where S t represents the amount of data uploaded in the t-th second, and T t represents the time required to upload S t , and L t represents the network latency time;

[0096] Integrate the data transfer rate and network latency, and smooth the integrated data, using the formula:

[0097]

[0098] and

[0099]

[0100] Obtain the smoothed data transfer rate and network latency , where w i represents the weight at the i-th second, DTR i and ND i represent the data transfer rate and network latency at the t-th second, and n is the size of the time window.

[0101] In one measurement, the tablet computer uploaded 2048 bytes of data in 1 second, the upload time was 0.8 seconds, and the measured network latency was 35 milliseconds.

[0102] Calculation of data transfer rate:

[0103]

[0104] Recording of network latency:

[0105] ND t = 35ms

[0106] 2560 bytes / s represents the amount of data uploaded per second;

[0107] 35 ms is the network latency time at the measurement moment.

[0108] Calculate the weighted average over the past three seconds using the aforementioned weights and data transfer rates. The collected DTR values are 2500, 2600, and 2700 bytes / s respectively, and the weights are 0.2, 0.3, and 0.5 respectively.

[0109] Calculation of the weighted average data transfer rate:

[0110]

[0111] Calculate the weighted average network latency. The collected ND values are 33 ms, 36 ms, and 35 ms respectively:

[0112]

[0113] 2630 bytes / s represents the weighted average data transfer rate over the past three seconds;

[0114] 34.9 ms represents the weighted average network latency over the past three seconds.

[0115] Please refer to Figure 3 , and the specific steps for screening key information are as follows:

[0116] Summarize the battery power data and application crash logs, calculate the average battery power and the occurrence frequency of application crashes, using the formulas:

[0117]

[0118] and

[0119]

[0120] Obtain the average battery power E avg and the application crash frequency F crash , where E i represents the i-th battery power reading, N is the total number of readings, T is the observation time interval, and C total is the total number of crashes during the observation period;

[0121] Based on the average battery power and the application crash frequency, perform data cleaning, remove invalid or incorrect data, and screen the key information in the dataset to obtain the preliminary diagnosis dataset.

[0122] There were a total of 5 power measurements in a day, and the values were: 80%, 70%, 75%, 65%, and 85% respectively;

[0123] The steps for calculating the average power are as follows:

[0124] Sum the power percentages:

[0125] 80 + 70 + 75 + 65 + 85 = 375%

[0126] Calculate the average value:

[0127]

[0128] The obtained 75% is the average value of the battery power within this day.

[0129] T is 1 day;

[0130] The application crashed 10 times within one day, and the number of days is 1, then:

[0131] Calculate the frequency:

[0132]

[0133] The obtained result indicates that the frequency of the application crashing is 10 times per day.

[0134] Two data points are monitored, the power readings are 105% and 95%, and the integrity of the crash logs is incomplete and complete respectively. Let the judgment threshold be 100%;

[0135] For the first data point, 105% is greater than 100%. Therefore, the first data point is invalid due to the power value exceeding the range and the incomplete log;

[0136] For the second data point, 95% is less than 100%. Therefore, the second data point is valid because it meets all conditions.

[0137] Please refer to Figure 4 , the steps for classifying the data are specifically as follows:

[0138] Based on the preliminary diagnosis data set, collect and organize time series data, battery power, and network latency information, and use the formula:

[0139]

[0140] Obtain the standardized feature matrix Xd norm , where Xd represents the original data point, mean(Xd) is the average value of the data, and std(Xd) is the standard deviation of the data;

[0141] Based on the standardized feature matrix, use the random forest algorithm for data classification to optimize the accuracy of the classification results, and use the formula:

[0142]

[0143] Obtain the classification result YX, where N d is the number of decision trees, Xt i is the classification function of the i-th tree, and Xd norm is the standardized feature matrix.

[0144] The battery power data at 5 measurement points during a day are [80%, 85%, 75%, 70%, 65%];

[0145] mean(Xd) is calculated as

[0146] std(Xd) is calculated as

[0147] Calculate the difference between each data point and the average value [5%, 10%, 0%, -5%, -10%];

[0148] Calculate the variance:

[0149]

[0150] Calculate the standard deviation:

[0151]

[0152] The data after normalization is more suitable for processing.

[0153] Normalize each data point and etc.

[0154] Let N d be 100, and the classification result of each tree for the same data point is 1 (abnormal) or 0 (normal);

[0155] Suppose among 100 trees, 60 trees classify a certain data point as abnormal, then the classification result of this point is:

[0156]

[0157] It indicates that this data point tends to be classified as abnormal.

[0158] Please refer to Figure 5 , and the specific steps for obtaining the fault identification index are as follows:

[0159] Analyze the generated abnormal category data, including extracting key features and patterns, and calculating the contribution rate and relevance of each feature. Use the formula:

[0160]

[0161] Obtain the feature set IScore associated with abnormal behavior. Among them, F i represents a single feature, Gain(F i ) represents the contribution of feature F i to the improvement of the model performance when constructing the decision tree. TGain is the total contribution of the feature, and N sis the total number of features;

[0162] Based on the feature set associated with abnormal behavior, a threshold is set to distinguish normal and abnormal states, using the formula:

[0163] Threshold = μ index + k s ·σ index

[0164] to obtain the fault identification index, where μ index is the historical average of the selected fault index, σ index is the historical standard deviation, k s is the coefficient used to adjust the threshold sensitivity, and Threshold is the threshold used to distinguish normal and abnormal states.

[0165] There are three features F1, F2, and F3, and their improvement contributions in the random forest are 30, 20, and 10 respectively.

[0166] Calculate the total contribution: TGain = 30 + 20 + 10 = 60

[0167]

[0168] The score indicates that F1 is the most important feature, making the greatest contribution to the improvement of the model, while the contribution of F3 is the smallest.

[0169] Assume that the historical average μ index of the fault index is 0.2, the standard deviation σ index is 0.05, and k s is 3;

[0170] Calculate the threshold:

[0171] Threshold = 0.2 + 3 × 0.05 = 0.35;

[0172] The threshold is used for fault detection. When the real-time monitored fault index exceeds 0.35, the system will trigger an alarm, indicating that the abnormal behavior exceeds three standard deviations of the historical range, which highly likely indicates an abnormal system state.

[0173] Please refer to Figure 6 , and the analysis steps for the urgency of the fault are specifically as follows:

[0174] Based on the fault identification index, initialize the decision tree model, distinguish different types of faults, and calculate the urgency score for each type, using the formula:

[0175] U initial = a c ·F frequency + b c ·Fimpact

[0176] Obtain the preliminary emergency score U initial , where F frequency and F impact represent the failure occurrence frequency and impact degree respectively, and a c and b c are the parameter influence weights;

[0177] Based on the preliminary emergency score, combined with the failure historical data and potential development trend, use the formula:

[0178] U final = U initial + c f ·H history + d f ·P potential

[0179] Obtain the emergency degree score U of the failure final , where H history and P potential represent the historical failure data and potential impact respectively, c f and d f are the corresponding adjustment coefficients, and U initial is the preliminary emergency score.

[0180] It is monitored that F frequency is 0.3 times / hour, and F impact is evaluated as 0.7;

[0181] Set the weights a c = 0.6 and b c = 0.4;

[0182] Calculate the preliminary emergency degree score:

[0183] U initial = 0.6·0.3 + 0.4·0.7 = 0.18 + 0.28 = 0.46

[0184] The score of 0.46 indicates that the failure has a medium emergency degree, and this score will be used for further evaluation in subsequent analysis.

[0185] It is collected that H history = 0.2 and P potential = 0.5;

[0186] Set the weights c f = 0.5 and d f = 0.5.

[0187] Refine the emergency degree score of the failure:

[0188] U final= 0.46 + 0.5·0.2 + 0.5·0.5 = 0.46 + 0.1 + 0.25 = 0.81

[0189] The calculated 0.81 indicates that this fault has a high degree of urgency and needs to be processed preferentially.

[0190] Please refer to Figure 7 , and the specific steps for obtaining the response policy details are as follows:

[0191] Based on the current total resources and urgent faults, evaluate the matching degree of the resource requirements, using the formula:

[0192]

[0193] Obtain the resource matching degree result R match , where C r represents the current CPU resource amount, M r represents the current memory resource amount, D r represents the CPU demand at the time of the fault, N r represents the memory demand at the time of the fault, α g and β g are adjustment coefficients used to adjust the resource requirements according to the current load;

[0194] Based on the resource matching degree result, determine the resource adjustment configuration, using the judgment formula:

[0195]

[0196] Obtain the response policy details Adjust Type , where R match is the resource matching degree result, γ o , δ o and ζ o are adjustment factors used to determine the strategy and intensity of resource adjustment.

[0197] The current tablet has a 4-core CPU (C r = 4) and 8GB RAM (M r = 8);

[0198] The fault handling requires 2-core CPU (D r = 2) and 4GB RAM (N r = 4);

[0199] Let the adjustment coefficient α g = 1.2 and β g = 0.8;

[0200] Calculate the resource matching degree:

[0201]

[0202] The result value of 261.54% indicates that the current resources far exceed the fault handling requirements, suggesting that the resources are over-allocated. Let γ o = 0.5, δ o = 1, and ζ o = 0.2; Based on the above results, since 261.54 is greater than 120, thus:

[0203] ζ o ·R match = 0.2 × 261.54 = 52.31

[0204] 52.31% means that the resource allocation needs to be reduced to about half to improve system efficiency;

[0205] When it is less than 80%, the amount of resource classification needs to be increased, and when it is between 80% - 120%, it remains unchanged.

[0206] Please refer to Figure 8 , and the specific steps for obtaining the maintenance effectiveness record are as follows:

[0207] Based on the response policy details, record the effects of maintenance activities, using the formula:

[0208]

[0209] Obtain the maintenance effectiveness score RJ, where lc i is the completion degree of activity i, le i is the effect score of activity i, lv i is the weight of activity i, and ln is the total number of maintenance activities;

[0210] Based on the maintenance effectiveness score, send it to the maintenance database through the 5G network, and verify the rate and integrity of data transmission to obtain a record of successful data reception;

[0211] Based on the record of successful data reception, calculate and analyze the success rate of maintenance activities, using the formula:

[0212]

[0213] Obtain the success rate and efficiency SR of maintenance activities, where Nl s is the number of successful maintenance activities, and Nl t is the total number of maintenance activities.

[0214] Suppose there are three maintenance activities:

[0215] Activity 1: The completion degree is 1, the effect score is 8, and the weight is 0.3;

[0216] Activity 2: Completion rate is 0.8, effectiveness score is 7, and weight is 0.4;

[0217] Activity 3: Completion rate is 0.9, effectiveness score is 9, and weight is 0.3.

[0218] Calculation process:

[0219] RJ = (1×8×0.3) + (0.8×7×0.4) + (0.9×9×0.3)

[0220] RJ = 2.4 + 2.24 + 2.43

[0221] RJ = 7.07

[0222] The obtained maintenance efficiency score is 7.07, and the score represents a comprehensive evaluation of the completion quality and effectiveness of the overall activity.

[0223] A total of 20 maintenance activities were monitored, and 18 of them were successful.

[0224] Calculation process:

[0225]

[0226] SR = 0.9×100%

[0227] SR = 90%

[0228] The obtained success rate of maintenance activities is 90%, indicating that most maintenance activities have met the expected success criteria.

[0229] A tablet intelligent management method based on 5G transmission includes the following steps:

[0230] S1: Based on the 5G network, receive the data transmission rate and network latency information of the tablet, summarize the battery power and application crash logs, and screen the key information in the status data to generate a preliminary diagnosis dataset;

[0231] S2: Based on the preliminary diagnosis dataset, integrate the associated time series data, identify normal and abnormal behavior patterns, locate the key fault points of the tablet, and use the fault modes with index classification differences to obtain fault identification indicators;

[0232] S3: According to the fault identification indicators, analyze the urgency and impact scope of the faults, configure response processes for high-risk fault points, perform deferred processing on low-risk faults, and configure the resource allocation of the tablet to generate response strategy details;

[0233] S4: Based on the response policy rules, record the effects of response execution, upload the data of the maintenance activities to the maintenance database through the 5G network, and extract the key performance data to generate a maintenance efficiency record;

[0234] S5: Utilize the maintenance efficiency record to evaluate the effects of the maintenance activities, analyze the success rate and efficiency of various tablet computer maintenance activities, identify potential areas for optimization, and generate the identification of optimization results.

[0235] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A tablet computer intelligent management system based on 5G transmission, characterized in that, The system includes: The data collection and screening module, based on the 5G network, performs parameter collection, collects the data transmission rate and network latency information uploaded by the tablet computer, summarizes the battery power and application crash logs, screens the key information in the dataset, and generates a preliminary diagnosis dataset; The intelligent fault analysis module, based on the preliminary diagnosis dataset, integrates the associated time series data, and classifies the data using the random forest algorithm to distinguish normal and abnormal behavior patterns, identify the key fault points of the tablet computer, and obtain the fault identification index; The specific steps for obtaining the fault identification index are as follows: Analyze the generated abnormal category data, including extracting key features and patterns, and calculating the contribution rate and correlation of each feature, using the formula: Obtain a feature set associated with abnormal behavior , where represents a single feature represents the feature 's contribution to the improvement of the model performance when constructing a decision tree is the total contribution of the feature is the total number of features; Based on the feature set associated with the abnormal behavior, set a threshold to distinguish normal and abnormal states, using the formula: Obtain a fault identification index, where is the historical average value of the selected fault index, is the historical standard deviation, is a coefficient used to adjust the threshold sensitivity, is the threshold for distinguishing normal and abnormal states; The response and configuration module, based on the fault identification index, analyzes the urgency of the fault using the decision tree algorithm, determines the response process, and configures the resource allocation of the tablet computer, including automatically restarting applications or adjusting performance allocation, and generates the response strategy details; The specific steps for analyzing the urgency of the fault are as follows: Based on the fault identification index, initialize the decision tree model, distinguish different types of faults, and calculate the urgency score for each type, using the formula: Obtain a preliminary emergency score , where and represent the failure occurrence frequency and impact degree respectively, and are the parameter influence weights; Based on the preliminary urgency score, combine the fault history data and potential development trends, using the formula: Obtain the emergency level score of the fault , where and represent historical fault data and potential impacts respectively and are the corresponding adjustment coefficients is the preliminary emergency score The maintenance evaluation and optimization module, based on the response strategy details, records the effect of the maintenance activities, sends the record to the maintenance database through the 5G network, and evaluates the success rate and efficiency of the maintenance activities, and generates the maintenance efficiency record.

2. The tablet computer intelligent management system based on 5G transmission according to claim 1, characterized in that, The specific steps for collecting the data transmission rate and network latency information are as follows: Based on the 5G network, collect the data transmission rate and network latency of the tablet computer, using the formula: and Obtain the data transfer rate per second and network latency , where represents the amount of data uploaded in the th second, represents the time required for uploading , and represents the network latency time; Integrate the data transmission rate and network latency, and smooth the integrated data, using the formula: and Obtain the data transmission rate after smoothing processing and network latency , where represents the weight of the th second, and represents the data transmission rate and network latency of the th second, is the time window size.

3. The tablet computer intelligent management system based on 5G transmission according to claim 1, characterized in that, The specific steps for screening the key information are as follows: Summarize the battery power data and application crash logs, calculate the average battery power and the occurrence frequency of application crashes, using the formula: and Obtain the average battery power and the application crash frequency , where represents the th battery power reading, is the total number of readings, is the observation time interval, is the total number of crashes during the observation period; Based on the average battery power and application crash frequency, perform data cleaning, remove invalid or incorrect data, and screen the key information in the dataset to obtain the preliminary diagnosis dataset.

4. The tablet computer intelligent management system based on 5G transmission according to claim 1, characterized in that, The specific steps for classifying the data are as follows: Based on the preliminary diagnosis dataset, collect and organize the time series data, battery power, and network latency information, using the formula: Obtain the standardized feature matrix , where represents the original data points, is the average value of the data, is the standard deviation of the data; Based on the standardized feature matrix, use the random forest algorithm for data classification to optimize the accuracy of the classification results, using the formula: Obtain the classification result , where is the number of decision trees is the classification function of the th tree is the standardized feature matrix 5. The intelligent management system for a tablet computer based on 5G transmission according to claim 1, wherein The specific steps for obtaining the response strategy details are as follows: Evaluate the matching degree of the resource requirements according to the current total resources and emergency faults, using the formula: Obtain the resource matching degree result , where represents the current CPU resource quantity, represents the current memory resource quantity, represents the CPU demand at the time of failure, represents the memory demand at the time of failure, and is an adjustment coefficient used to adjust the resource demand according to the current load; Based on the resource matching degree result, determine the resource adjustment configuration, using the judgment formula: Obtain response policy details , where is the resource matching degree result, , and are adjustment factors.

6. The tablet computer intelligent management system based on 5G transmission according to claim 1, characterized in that, The specific steps for obtaining the maintenance efficiency record are as follows: Based on the response strategy details, record the effect of the maintenance activities, using the formula: Obtain the maintenance efficiency score , where is the completion degree of activity , is the effect score of activity , is the weight of activity , is the total number of maintenance activities; Based on the maintenance efficiency score, it is sent to the maintenance database through the 5G network, and the rate and integrity of data transmission are verified to obtain a data successful reception record; Based on the data successful reception record, calculate and analyze the success rate of maintenance activities, using the formula: Obtain the success rate and efficiency of maintenance activities , where is the number of successful maintenance activities, is the total number of maintenance activities.

7. A smart management method for a tablet computer based on 5G transmission, characterized in that, Execute according to the tablet computer intelligent management system based on 5G transmission described in any one of claims 1-6, including the following steps: Based on the 5G network, receive the data transmission rate and network latency information of the tablet computer, summarize the battery power and application crash logs, and screen the key information in the status data to generate a preliminary diagnosis data set; Based on the preliminary diagnosis data set, integrate the associated time series data, identify normal and abnormal behavior patterns, locate the key fault points of the tablet computer, and use the fault modes with different indicator classifications to obtain fault identification indicators; According to the fault identification indicators, analyze the urgency and impact scope of the faults, configure response processes for high-risk fault points, perform deferred processing on low-risk faults, and configure the resource allocation of the tablet computer to generate response strategy details; Based on the response strategy details, record the effect of response execution, upload the data of maintenance activities to the maintenance database through the 5G network, and extract key performance data to generate a maintenance efficiency record; Using the maintenance efficiency record, evaluate the effect of maintenance activities, analyze the success rate and efficiency of various tablet computer maintenance activities, identify potential areas for optimization, and generate identification optimization results.

Images