Touch data transmission optimization system and method based on USB interface, and storage medium

Abstract

The invention discloses a touch data transmission optimization system and method based on a USB interface and a storage medium, relates to the technical field of data transmission, and solves the technical problem that when an OPS box transmits touch data through the USB interface, the touch feedback efficiency is low due to the fact that the OPS box may be affected by delay and errors. A transmission interface is selected according to touch content and USB interface data; obtaining a feedback prediction label through the feedback prediction model; dynamically adjusting the current prediction cache region according to the feedback prediction label and the previous prediction cache region; searching corresponding feedback data according to the touch content; according to the technical scheme, the feedback interface is selected according to the feedback data and the USB interface data, the touch data touched at the same time is subjected to priority division, transmission and feedback are conducted according to the priority, a dynamic prediction cache region is arranged, frequent data transmission with a database is avoided, the response speed of the feedback data is increased, and then the touch feedback efficiency is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of data transmission, and specifically relates to a touch data transmission optimization system and method and a storage medium based on a USB interface. Background Art

[0002] Data transmission refers to the process of transmitting information from one place to another through a specific channel or medium. In modern electronic devices, touch screens have become the main way to interact with users. However, touch data transmission systems still have problems such as slow data transmission speed, response delay, and complex connections. These problems are particularly prominent when processing large amounts of data or complex operations. Once the efficiency and stability of data transmission are poor, it will directly affect the user's interactive experience.

[0003] OPS is a computing module plug-in format that can be used to add computing power to flat panel displays. The OPS box is an intelligent terminal device designed based on this specification, supports USB interface, and can be used with a variety of display devices to obtain a better user experience. However, due to the limitations of the USB interface, transmission protocol, and system resource usage, there may be situations where too much system resources are occupied and the amount of data is too large, causing the touch data to be affected by delays and errors, resulting in low touch feedback efficiency; therefore, the touch data transmission system based on the USB interface still needs to be further improved. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a touch data transmission optimization system and method based on a USB interface, and a storage medium, which are used to solve the technical problem that when the OPS box transmits touch data through a USB interface, it may be affected by delays and errors, resulting in low touch feedback efficiency.

[0005] To achieve the above-mentioned purpose, the first aspect of the present application provides a touch data transmission optimization system based on a USB interface, comprising: a data acquisition module, a data analysis module, an early warning module, a display module and a database;

[0006] The data acquisition module is used to acquire touch data and USB interface data through a data acquisition device;

[0007] The data analysis module: obtains touch content according to touch data; selects a transmission interface according to touch content and USB interface data; obtains a feedback prediction label through a feedback prediction model; dynamically adjusts the current prediction buffer area according to the feedback prediction label and the previous prediction buffer area; searches for corresponding feedback data in the prediction buffer area and the database according to the touch content; and selects a feedback interface according to the feedback data and USB interface data;

[0008] The early warning module: issues a prompt according to the alarm signal and contacts the management personnel;

[0009] The display module is used for user operation and displaying feedback data.

[0010] The present application selects a transmission interface according to touch content and USB interface data; obtains a feedback prediction tag through a feedback prediction model; dynamically adjusts the current prediction cache area according to the feedback prediction tag and the previous prediction cache area; searches for corresponding feedback data in the prediction cache area and the database according to the touch content; selects a feedback interface according to the feedback data and USB interface data, prioritizes the touch data of simultaneous touches, transmits and feedbacks according to the priority, and sets a dynamic prediction cache area, thereby avoiding frequent data transmission with the database, improving the response speed of feedback data, and thereby improving the efficiency of touch feedback.

[0011] Further, obtaining touch content according to touch data includes:

[0012] Acquire contact data and display content in the current contact display screen; the touch data includes touch position, touch type and touch time;

[0013] Find the touch content corresponding to the contact position in the current display.

[0014] Further, selecting a transmission interface according to the touch content and USB interface data includes:

[0015] Get USB interface data, touch content and its corresponding touch time;

[0016] Determine whether there is touch content with consistent touch time, if yes, generate touch priority according to the touch content and its corresponding historical touch times; select a transmission interface according to the touch priority and the data size corresponding to the touch content;

[0017] No, select the transmission interface based on the data size corresponding to the touch content.

[0018] Furthermore, the touch content and its corresponding historical touch times generate touch priorities, including:

[0019] Obtain a number of touch contents and their corresponding historical touch counts LCCi; search for their corresponding content levels NRDi through a touch content level table;

[0020] The touch priority corresponding to the touch content is calculated by the formula CYi=1-exp(-(α1×LCCi / DC+α2×NRDi / DD)); wherein i represents the number corresponding to the simultaneously touched content, i=1, 2, ..., M; M represents the total number of simultaneously touched contents; α1 and α2 are weight coefficients, α1 and α2∈(0, 1).

[0021] The present application prioritizes touch data occurring simultaneously, calculates the content level corresponding to the historical touch count and the touch position, and preferentially transmits the touch data according to the touch priority, thereby improving the transmission efficiency.

[0022] Further, the selecting of the transmission interface according to the touch priority and the data size corresponding to the touch content includes:

[0023] Obtain the device performance, touch content corresponding to the maximum value of the touch priority, content type, and data size in turn;

[0024] Integrate device performance, touch content, content type and data size into comprehensive data and input it into a bandwidth prediction model to obtain a bandwidth estimation value;

[0025] Obtaining USB interface data; the USB interface data includes an input channel ID and a bandwidth margin;

[0026] Determine in turn whether the bandwidth estimation value corresponding to the maximum touch priority value is less than the maximum bandwidth margin corresponding to the input channel ID; if so, load the touch content corresponding to the maximum touch priority value into the input channel corresponding to the input channel ID; if not, divide the touch content into a number of touch sub-contents that satisfy the bandwidth margins corresponding to each input channel ID through a data segmentation algorithm; and load the touch sub-contents into the input channels corresponding to each input channel ID.

[0027] Furthermore, the bandwidth prediction model is constructed by an artificial intelligence model, including:

[0028] Acquire several historical touch contents and their corresponding content types and data sizes, historical device performance and corresponding channel bandwidth; and integrate device performance, touch contents, content types and data sizes into historical comprehensive data;

[0029] Divide the historical comprehensive data and its corresponding channel bandwidth into training data, verification data and test data;

[0030] Perform data preprocessing on training data, verification data and test data to obtain training set, verification set and test set;

[0031] Use the training set to train the artificial intelligence model, adjust the parameters on the validation set, and test the model on the test data to ultimately obtain a bandwidth estimation model with comprehensive data as input and bandwidth estimation value as output.

[0032] Furthermore, the data preprocessing includes data cleaning and data standardization.

[0033] Furthermore, the feedback prediction model is constructed by using an LSTM model, including:

[0034] Obtain some historical time data and its corresponding device data, touch content and feedback labels;

[0035] Performing data preprocessing on a number of time data and corresponding device data, touch content, and feedback labels to obtain a number of preprocessed time data and corresponding device data, touch content, and feedback labels;

[0036] Integrate the device data, touch content, feedback labels and time data with the same time data into a time feature vector;

[0037] Integrate each time feature vector into several prediction sequences according to the order of time data;

[0038] Divide several prediction sequences into training set, test set and validation set in chronological order;

[0039] Select the LSTM model as the base model;

[0040] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0041] By verifying the pre-trained model on the test set, we finally get the input time data and output a feedback prediction model with several feedback prediction labels.

[0042] Furthermore, dynamically adjusting the current prediction cache area according to the feedback prediction tag and the last prediction cache area includes:

[0043] Get the feedback prediction label and the last prediction cache;

[0044] Determine whether a number of feedback prediction tags are in the last prediction cache;

[0045] Yes, when the fixed coefficient corresponding to the remaining feedback tags in the prediction buffer is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction buffer; the prediction buffer is dynamically adjusted according to the feedback tags in the prediction buffer; otherwise, no operation is performed;

[0046] No, when the fixed coefficient corresponding to the remaining feedback tags in the prediction cache area is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction cache area; the space size of the prediction cache area is dynamically adjusted according to the feedback tags in the prediction cache area, and the feedback data corresponding to the prediction feedback tags are transferred to the prediction cache area; otherwise, no operation is performed.

[0047] Furthermore, the fixed coefficient is obtained by feedback times, including:

[0048] Get the number of feedbacks FCj corresponding to the feedback label in the prediction buffer within the time range T;

[0049] The fixed coefficient is calculated by the formula GXj=g×arctan(FCj); wherein j represents the number of the prediction feedback label in the prediction cache, j=1, 2, ..., N; N represents the total number of prediction feedback labels in the prediction cache; g is the proportional coefficient, g∈(0, π / 2).

[0050] Furthermore, dynamically adjusting the space size of the prediction buffer area according to the feedback tag in the prediction buffer area includes:

[0051] Obtaining a number of feedback labels in the prediction buffer and their corresponding data sizes, as well as unmatched prediction feedback labels in the prediction buffer and their corresponding data sizes;

[0052] The data size of all feedback tags is summed to obtain the theoretical size HLD of the buffer area;

[0053] The buffer area adjustment size is calculated by the formula HTD=(1+β)×HLD; wherein β is an adjustment factor, β∈(0,1).

[0054] The present application calculates the fixed coefficient of the feedback tag by calculating the situation in which the feedback tag in the prediction cache area is called within a time range. The smaller the fixed coefficient is, the feedback data corresponding to the feedback tag has not been used for a long time and needs to be eliminated. The space size of the prediction cache area is dynamically adjusted to reduce resource waste.

[0055] Furthermore, searching for corresponding feedback data in the prediction buffer area and the database according to the touch content includes:

[0056] Get the touch content;

[0057] Input the touch content into the feedback label query table to find the corresponding feedback label;

[0058] Determine whether there is a corresponding feedback tag in the prediction buffer; if yes, obtain feedback data corresponding to the feedback tag;

[0059] If no, determine whether there is a corresponding feedback tag in the database; if yes, obtain feedback data corresponding to the feedback tag; if no, generate a no record alarm signal.

[0060] Furthermore, the feedback tag query table is constructed by the following steps, including:

[0061] Acquire touch content, feedback labels and feedback data; the touch content and the feedback labels correspond one to one; the feedback labels and the feedback data correspond one to one;

[0062] A feedback label query table is created according to the touch content, feedback label, and the corresponding feedback data.

[0063] Further, selecting the feedback interface according to the feedback data and the USB interface data includes:

[0064] Obtain the output channel ID and bandwidth margin in the feedback data and USB interface data;

[0065] Determine whether the feedback tag corresponding to the feedback data has a feedback priority; the feedback priority is consistent with the touch priority;

[0066] Yes, extract the feedback data with the highest feedback priority and its data size in turn; obtain the feedback bandwidth estimation value through the bandwidth prediction model;

[0067] It is determined in sequence whether the feedback bandwidth estimation value corresponding to the maximum feedback priority value is less than the maximum bandwidth margin corresponding to the feedback channel ID. If yes, the feedback data corresponding to the maximum feedback priority value is loaded into the input channel corresponding to the input channel ID; if no, the feedback data is split into a number of feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID by using a data splitting algorithm; and the feedback sub-data is loaded into the feedback channel corresponding to each feedback channel ID;

[0068] If no, obtain the feedback bandwidth estimation value through the bandwidth prediction model; determine whether the feedback bandwidth estimation value is less than the maximum bandwidth margin corresponding to the feedback channel ID; if yes, load the feedback data into the input channel corresponding to the input channel ID; if no, divide the feedback data into several feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID through the data division algorithm; load the feedback sub-data into the feedback channel corresponding to each feedback channel ID.

[0069] A second aspect of the present invention provides a touch data transmission optimization method based on a USB interface, comprising:

[0070] S1: Get touch data and USB interface data;

[0071] S2: obtaining touch content according to the touch data; selecting a transmission interface according to the touch content and USB interface data;

[0072] S3: Obtain feedback prediction labels through the feedback prediction model; dynamically adjust the current prediction cache area according to the feedback prediction labels and the previous prediction cache area;

[0073] S4: searching for corresponding feedback data in the prediction buffer area and the database according to the touch content; selecting a feedback interface according to the feedback data and the USB interface data;

[0074] S5: Present the feedback data in a visual way; issue prompts based on alarm signals and contact management personnel.

[0075] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon.

[0076] Compared with the prior art, the beneficial effects of this application are:

[0077] 1. The present application selects a transmission interface according to touch content and USB interface data; obtains a feedback prediction label through a feedback prediction model; dynamically adjusts the current prediction cache area according to the feedback prediction label and the previous prediction cache area; searches for corresponding feedback data in the prediction cache area and the database according to the touch content; selects a feedback interface according to the feedback data and USB interface data, prioritizes the touch data of simultaneous touches, transmits and feedbacks according to the priority, and sets a dynamic prediction cache area, thereby avoiding frequent data transmission with the database, improving the response speed of feedback data, and thereby improving the efficiency of touch feedback.

[0078] 2. This application prioritizes the touch data that occurs simultaneously, calculates the content level corresponding to the historical touch count and the touch position, and preferentially transmits the touch data according to the touch priority, thereby improving the transmission efficiency.

[0079] 3. This application calculates the fixed coefficient of the feedback tag by calculating the situation in which the feedback tag in the prediction cache is called within a time range. The smaller the fixed coefficient, the longer the feedback data corresponding to the feedback tag has not been used and needs to be eliminated. The space size of the prediction cache is dynamically adjusted to reduce resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0081] Figure 1This is a schematic diagram of the principle of the touch data transmission optimization system based on the USB interface of the present application;

[0082] Figure 2 This is a flow chart of the touch data transmission optimization method based on the USB interface of the present application. DETAILED DESCRIPTION

[0083] The technical solution of the present application will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0084] See also Figure 1 , the first aspect of the present application provides a touch data transmission optimization system based on a USB interface, including: a data acquisition module, a data analysis module, an early warning module, a display module and a database;

[0085] Data acquisition module: acquire touch data and USB interface data through data acquisition equipment; data acquisition equipment includes various sensors, etc.;

[0086] Data analysis module: obtain touch content according to touch data; select a transmission interface according to touch content and USB interface data, the transmission interface refers to a channel interface for transmitting touch data to a processing unit; obtain a feedback prediction label through a feedback prediction model; dynamically adjust the current prediction buffer area according to the feedback prediction label and the previous prediction buffer area; search for corresponding feedback data in the prediction buffer area and the database according to the touch content; select a feedback interface according to the feedback data and USB interface data, the feedback interface refers to a channel interface for transmitting feedback data to a display module;

[0087] Early warning module: issues prompts based on alarm signals and contacts management personnel; alarm signals include no recorded alarm signals, etc.;

[0088] Display module: used for user operation and displaying feedback data.

[0089] This embodiment selects a transmission interface according to touch content and USB interface data; obtains a feedback prediction label through a feedback prediction model; dynamically adjusts the current prediction buffer area according to the feedback prediction label and the previous prediction buffer area; searches for corresponding feedback data in the prediction buffer area and the database according to the touch content; selects a feedback interface according to the feedback data and USB interface data, prioritizes the touch data of simultaneous touches, transmits and feedbacks according to the priority, and sets a dynamic prediction buffer area, thereby avoiding frequent data transmission with the database, improving the response speed of feedback data, and further improving the efficiency of touch feedback.

[0090] The touch content is obtained according to the touch data, including:

[0091] Acquire contact data and the display content of the current contact display screen; the touch data includes the touch position, touch type and touch time, etc.;

[0092] Search the touch content corresponding to the contact position in the current display screen; the touch content may not be the same even if the touch position is the same, and it is also related to the information displayed on the touch screen at that time.

[0093] In this embodiment, selecting a transmission interface according to touch content and USB interface data includes:

[0094] Get USB interface data, touch content and its corresponding touch time;

[0095] Determine whether there is touch content with consistent touch time, that is, generate touch priority according to the touch content and its corresponding historical touch times; select transmission interface according to touch priority and data size corresponding to touch content; touch priority refers to the priority corresponding to touch content, and touch content with high priority is given priority for data transmission and data feedback; consistent touch time refers to clicking the screen at the same time;

[0096] No, select the transmission interface based on the data size corresponding to the touch content.

[0097] In this embodiment, the touch priority is generated according to the touch content and its corresponding historical touch times, including:

[0098] Obtaining a number of touch contents and their corresponding historical touch counts LCCi; searching for their corresponding content levels NRDi through a touch content level table; and evaluating the content levels by relevant experts;

[0099] The touch priority corresponding to the touch content is calculated by the formula CYi=1-exp(-(α1×LCCi / DC+α2×NRDi / DD)); wherein i represents the number corresponding to the simultaneously touched content, i=1, 2, ..., M; M represents the total number of simultaneously touched content; α1 and α2 are weight coefficients, α1 and α2∈(0, 1), and the specific values ​​are set according to experience; the higher the content level, the higher the touch priority corresponding to the touch content; the more historical touches, the higher the corresponding touch priority, in the case of simultaneous touch content; the more historical touches, the higher the content level, and the more inclined to respond to it; therefore, the touch priority increases accordingly.

[0100] This embodiment prioritizes the touch data occurring simultaneously, calculates the content level corresponding to the historical touch times and the touch position, and preferentially transmits the touch data according to the touch priority, thereby improving the transmission efficiency.

[0101] In this embodiment, the transmission interface is selected according to the touch priority and the data size corresponding to the touch content, including:

[0102] Obtain the device performance, touch content corresponding to the maximum value of the touch priority, content type, and data size in turn;

[0103] Integrate device performance, touch content, content type and data size into comprehensive data and input it into a bandwidth prediction model to obtain a bandwidth estimation value;

[0104] Get USB interface data; USB interface data includes input channel ID and bandwidth margin, etc.

[0105] Determine in turn whether the bandwidth estimation value corresponding to the maximum touch priority value is less than the maximum bandwidth margin corresponding to the input channel ID; if so, load the touch content corresponding to the maximum touch priority value into the input channel corresponding to the input channel ID; if not, divide the touch content into a number of touch sub-contents satisfying the bandwidth margins corresponding to each input channel ID through a data segmentation algorithm; load the touch sub-contents into the input channels corresponding to each input channel ID; the segmentation algorithm includes a hash segmentation algorithm, etc.

[0106] The bandwidth prediction model in this embodiment is constructed by an artificial intelligence model, including:

[0107] Acquire several historical touch contents and their corresponding content types and data sizes, historical device performance and corresponding channel bandwidth; and integrate the device performance, touch contents, content types and data sizes into historical comprehensive data; the historical comprehensive data and channel bandwidth correspond one to one;

[0108] Divide the historical comprehensive data and its corresponding channel bandwidth into training data, verification data and test data;

[0109] The training data, validation data and test data are preprocessed to obtain the training set, validation set and test set; the ratio between the training set, test set and validation set is 7:2:1;

[0110] The artificial intelligence model is trained using the training set, the parameters are adjusted on the validation set, and the model is tested on the test data, ultimately obtaining a bandwidth estimation model whose input is comprehensive data and whose output is a bandwidth estimation value; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0111] The data preprocessing in this embodiment includes data cleaning and data standardization;

[0112] In another embodiment, data cleaning: directly removing missing values ​​from the acquired historical comprehensive data to obtain normal data;

[0113] Data standardization: Normal data is standardized using the maximum and minimum normalization method to obtain preprocessed comprehensive data.

[0114] The feedback prediction model in this embodiment is constructed by using an LSTM model, including:

[0115] Obtain some historical time data and its corresponding device data, touch content and feedback labels;

[0116] Performing data preprocessing on a number of time data and corresponding device data, touch content, and feedback labels to obtain a number of preprocessed time data and corresponding device data, touch content, and feedback labels;

[0117] Integrate the device data, touch content, feedback labels and time data with the same time data into a time feature vector;

[0118] Integrate each time feature vector into several prediction sequences according to the order of time data;

[0119] Divide several prediction sequences into training set, test set and validation set in chronological order; the ratio between training set, test set and validation set is 7:2:1;

[0120] Select the LSTM model as the base model;

[0121] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0122] By verifying the pre-trained model on the test set, we finally get the input time data and output a feedback prediction model with several feedback prediction labels.

[0123] In this embodiment, dynamically adjusting the current prediction cache area according to the feedback prediction tag and the previous prediction cache area includes:

[0124] Obtain the feedback prediction label and the last prediction buffer; the prediction buffer includes the feedback label, feedback data and its corresponding fixed coefficient, etc.;

[0125] Determine whether a number of feedback prediction tags are in the last prediction cache;

[0126] Yes, when the fixed coefficient corresponding to the remaining feedback tags in the prediction buffer is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction buffer; the prediction buffer is dynamically adjusted according to the feedback tags in the prediction buffer; otherwise, no operation is performed; the fixed threshold is set based on experience;

[0127] No, when the fixed coefficient corresponding to the remaining feedback tags in the prediction cache is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction cache; the spatial size of the prediction cache is dynamically adjusted according to the feedback tags in the prediction cache, and the feedback data corresponding to the prediction feedback tags are transferred to the prediction cache; otherwise, no operation is performed.

[0128] In this embodiment, a feedback prediction tag is first obtained, and then a search is performed in the prediction cache to determine whether there is a corresponding feedback tag. If there is a consistent one, the relevant data in the prediction cache is retained. Otherwise, the remaining feedback tags in the prediction cache are retained or deleted according to a fixed coefficient, and the data of the prediction feedback tags not in the prediction cache are transferred to the prediction cache, thereby reducing the number of times the database is accessed and improving feedback efficiency.

[0129] The fixed coefficient is obtained by feedback times, including:

[0130] Get the number of feedbacks FCj corresponding to the feedback label in the prediction buffer within the time range T;

[0131] The fixed coefficient is calculated by the formula GXj=g×arctan(FCj); wherein j represents the number of the prediction feedback label in the prediction cache, j=1, 2, ..., N; N represents the total number of prediction feedback labels in the prediction cache; g is the proportional coefficient, g∈(0, π / 2), and the specific value is set according to experience; within the time range T, if the number of feedbacks corresponding to the feedback label is more, it means that the feedback data is frequently used and should not be removed from the prediction cache. Therefore, as the number of feedbacks increases, the fixed coefficient will also increase accordingly.

[0132] In this embodiment, dynamically adjusting the space size of the prediction buffer area according to the feedback tag in the prediction buffer area includes:

[0133] Obtaining a number of feedback labels in the prediction buffer and their corresponding data sizes, as well as unmatched prediction feedback labels in the prediction buffer and their corresponding data sizes;

[0134] The data size of all feedback tags is summed to obtain the theoretical size HLD of the buffer area;

[0135] The cache adjustment size is calculated by the formula HTD=(1+β)×HLD; wherein β is an adjustment factor, β∈(0,1), and the specific value is set based on experience, and is set to 0.1 in this embodiment; HTD increases with the increase of HLD; the spatial size of the initial prediction cache in this embodiment is calculated by the initial prediction feedback label, and the spatial size of each subsequent prediction cache is adjusted based on the previous prediction cache.

[0136] This embodiment calculates the fixed coefficient of the feedback tag by calculating the situation in which the feedback tag in the prediction cache area is called within a time range. The smaller the fixed coefficient is, the feedback data corresponding to the feedback tag has not been used for a long time and needs to be eliminated. The space size of the prediction cache area is dynamically adjusted to reduce resource waste.

[0137] In this embodiment, searching for corresponding feedback data in the prediction buffer area and the database according to the touch content includes:

[0138] Get the touch content;

[0139] Input the touch content into the feedback label query table to find the corresponding feedback label;

[0140] Determine whether there is a corresponding feedback tag in the prediction buffer; if yes, obtain feedback data corresponding to the feedback tag;

[0141] If no, determine whether there is a corresponding feedback tag in the database; if yes, obtain the feedback data corresponding to the feedback tag; if no, generate a no record alarm signal.

[0142] The feedback tag query table in this embodiment is constructed by the following steps, including:

[0143] Acquire touch content, feedback labels and feedback data; touch content and feedback labels correspond one to one; feedback labels and feedback data correspond one to one;

[0144] A feedback label query table is created according to the touch content, feedback label, and the corresponding feedback data.

[0145] In this embodiment, the feedback interface is selected according to the feedback data and the USB interface data, including:

[0146] Obtain the output channel ID and bandwidth margin in the feedback data and USB interface data;

[0147] Determine whether the feedback tag corresponding to the feedback data has feedback priority; the feedback priority is consistent with the touch priority;

[0148] Yes, extract the feedback data with the highest feedback priority and its data size in turn; obtain the feedback bandwidth estimation value through the bandwidth prediction model;

[0149] It is determined in sequence whether the feedback bandwidth estimation value corresponding to the maximum feedback priority value is less than the maximum bandwidth margin corresponding to the feedback channel ID. If yes, the feedback data corresponding to the maximum feedback priority value is loaded into the input channel corresponding to the input channel ID; if no, the feedback data is split into a number of feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID by using a data splitting algorithm; and the feedback sub-data is loaded into the feedback channel corresponding to each feedback channel ID;

[0150] If no, obtain the feedback bandwidth estimation value through the bandwidth prediction model; determine whether the feedback bandwidth estimation value is less than the maximum bandwidth margin corresponding to the feedback channel ID; if yes, load the feedback data into the input channel corresponding to the input channel ID; if no, divide the feedback data into several feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID through the data division algorithm; load the feedback sub-data into the feedback channel corresponding to each feedback channel ID.

[0151] See also Figure 2 The second aspect of the present application provides a touch data transmission optimization method based on a USB interface, including:

[0152] S1: Get touch data and USB interface data;

[0153] S2: obtaining touch content according to the touch data; selecting a transmission interface according to the touch content and USB interface data;

[0154] S3: Obtain feedback prediction labels through the feedback prediction model; dynamically adjust the current prediction cache area according to the feedback prediction labels and the previous prediction cache area;

[0155] S4: searching for corresponding feedback data in the prediction buffer area and the database according to the touch content; selecting a feedback interface according to the feedback data and the USB interface data;

[0156] S5: Present the feedback data in a visual way; issue prompts based on alarm signals and contact management personnel.

[0157] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon.

[0158] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0159] The working principle of the present application is as follows: by acquiring touch data and USB interface data; obtaining touch content according to the touch data; selecting a transmission interface according to the touch content and USB interface data; obtaining a feedback prediction label through a feedback prediction model; dynamically adjusting the current prediction cache area according to the feedback prediction label and the previous prediction cache area; searching for corresponding feedback data in the prediction cache area and the database according to the touch content; selecting a feedback interface according to the feedback data and USB interface data; issuing a prompt according to the alarm signal, contacting the management personnel, and visualizing the feedback data, prioritizing the touch data of simultaneous touches, transmitting and feedbacking according to the priority, and setting a dynamic prediction cache area, thereby avoiding frequent data transmission with the database, improving the response speed of feedback data, and thereby increasing the efficiency of touch feedback, and avoiding the problem of low touch feedback efficiency caused by delays and errors when the OPS box transmits touch data through the USB interface.

[0160] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A touch data transmission optimization system based on USB interface, characterized in that: include: Data acquisition module, data analysis module, early warning module, display module and database; The data acquisition module is used to acquire touch data and USB interface data through a data acquisition device; The data analysis module: obtains touch content according to touch data; selects a transmission interface according to touch content and USB interface data; obtains a feedback prediction label through a feedback prediction model; dynamically adjusts the current prediction buffer area according to the feedback prediction label and the previous prediction buffer area; searches for corresponding feedback data in the prediction buffer area and the database according to the touch content; and selects a feedback interface according to the feedback data and USB interface data; The early warning module: issues a prompt according to the alarm signal and contacts the management personnel; The display module is used for user operation and displaying feedback data.

2. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The step of obtaining touch content according to touch data includes: Acquire contact data and display content in the current contact display screen; the touch data includes touch position, touch type and touch time; Find the touch content corresponding to the contact position in the current display.

3. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The selecting of the transmission interface according to the touch content and the USB interface data comprises: Get USB interface data, touch content and its corresponding touch time; Determine whether there is touch content with consistent touch time, if yes, generate touch priority according to the touch content and its corresponding historical touch times; select a transmission interface according to the touch priority and the data size corresponding to the touch content; No, select the transmission interface based on the data size corresponding to the touch content.

4. The touch data transmission optimization system based on USB interface according to claim 3, characterized in that: The touch content and its corresponding historical touch times generate touch priorities, including: Obtain a number of touch contents and their corresponding historical touch counts LCCi; search for their corresponding content levels NRDi through a touch content level table; The touch priority corresponding to the touch content is calculated by the formula CYi=1-exp(-(α1×LCCi / DC+α2×NRDi / DD)); wherein i represents the number corresponding to the simultaneously touched content, i=1, 2, ..., M; M represents the total number of simultaneously touched contents; α1 and α2 are weight coefficients, α1 and α2∈(0, 1).

5. The touch data transmission optimization system based on USB interface according to claim 3 is characterized in that: The selecting of the transmission interface according to the touch priority and the data size corresponding to the touch content includes: Obtain the device performance, touch content corresponding to the maximum value of the touch priority, content type, and data size in turn; Integrate device performance, touch content, content type and data size into comprehensive data and input it into a bandwidth prediction model to obtain a bandwidth estimation value; Obtaining USB interface data; the USB interface data includes an input channel ID and a bandwidth margin; Determine in turn whether the bandwidth estimation value corresponding to the maximum touch priority value is less than the maximum bandwidth margin corresponding to the input channel ID; if so, load the touch content corresponding to the maximum touch priority value into the input channel corresponding to the input channel ID; if not, divide the touch content into a number of touch sub-contents that satisfy the bandwidth margins corresponding to each input channel ID through a data segmentation algorithm; and load the touch sub-contents into the input channels corresponding to each input channel ID.

6. The touch data transmission optimization system based on USB interface according to claim 5, characterized in that: The bandwidth prediction model is constructed through an artificial intelligence model, including: Acquire several historical touch contents and their corresponding content types and data sizes, historical device performance and corresponding channel bandwidth; and integrate device performance, touch contents, content types and data sizes into historical comprehensive data; Divide the historical comprehensive data and its corresponding channel bandwidth into training data, verification data and test data; Perform data preprocessing on training data, verification data and test data to obtain training set, verification set and test set; Use the training set to train the artificial intelligence model, adjust the parameters on the validation set, and test the model on the test data to ultimately obtain a bandwidth estimation model with comprehensive data as input and bandwidth estimation value as output.

7. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The feedback prediction model is constructed by using the LSTM model, including: Obtain some historical time data and its corresponding device data, touch content and feedback labels; Performing data preprocessing on a number of time data and corresponding device data, touch content, and feedback labels to obtain a number of preprocessed time data and corresponding device data, touch content, and feedback labels; Integrate the device data, touch content, feedback labels and time data with the same time data into a time feature vector; Integrate each time feature vector into several prediction sequences according to the order of time data; Divide several prediction sequences into training sets, test sets and validation sets in chronological order; Select the LSTM model as the base model; Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally get the input time data and output a feedback prediction model with several feedback prediction labels.

8. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The dynamically adjusting the current prediction cache area according to the feedback prediction tag and the last prediction cache area includes: Get the feedback prediction label and the last prediction cache; Determine whether a number of feedback prediction tags are in the last prediction cache; Yes, when the fixed coefficient corresponding to the remaining feedback tags in the prediction buffer is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction buffer; the prediction buffer is dynamically adjusted according to the feedback tags in the prediction buffer; otherwise, no operation is performed; No, when the fixed coefficient corresponding to the remaining feedback tags in the prediction cache area is less than the fixed threshold, the remaining feedback tags less than the fixed threshold and their corresponding feedback data are removed from the prediction cache area; the space size of the prediction cache area is dynamically adjusted according to the feedback tags in the prediction cache area, and the feedback data corresponding to the prediction feedback tags are transferred to the prediction cache area; otherwise, no operation is performed.

9. The touch data transmission optimization system based on USB interface according to claim 8, characterized in that: The fixed coefficient is obtained by feedback times, including: Get the number of feedbacks FCj corresponding to the feedback label in the prediction buffer within the time range T; The fixed coefficient is calculated by the formula GXj=g×arctan(FCj); wherein j represents the number of the prediction feedback label in the prediction cache, j=1, 2, ..., N; N represents the total number of prediction feedback labels in the prediction cache; g is the proportional coefficient, g∈(0, π / 2).

10. The touch data transmission optimization system based on USB interface according to claim 8, characterized in that: The dynamically adjusting the space size of the prediction buffer area according to the feedback tag in the prediction buffer area includes: Obtaining a number of feedback labels in the prediction buffer and their corresponding data sizes, as well as unmatched prediction feedback labels in the prediction buffer and their corresponding data sizes; The data size of all feedback tags is summed to obtain the theoretical size HLD of the buffer area; The buffer area adjustment size is calculated by the formula HTD=(1+β)×HLD; wherein β is an adjustment factor, β∈(0,1).

11. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The searching corresponding feedback data in the prediction buffer area and the database according to the touch content includes: Get the touch content; Input the touch content into the feedback label query table to find the corresponding feedback label; Determine whether there is a corresponding feedback tag in the prediction buffer; if yes, obtain feedback data corresponding to the feedback tag; If no, determine whether there is a corresponding feedback tag in the database; if yes, obtain feedback data corresponding to the feedback tag; if no, generate a no record alarm signal.

12. The touch data transmission optimization system based on USB interface according to claim 11, characterized in that: The feedback tag query table is constructed by the following steps, including: Acquire touch content, feedback labels and feedback data; the touch content and the feedback labels correspond one to one; the feedback labels and the feedback data correspond one to one; A feedback label query table is created according to the touch content, feedback label, and the corresponding feedback data.

13. The touch data transmission optimization system based on USB interface according to claim 1, characterized in that: The step of selecting the feedback interface according to the feedback data and the USB interface data includes: Obtain the output channel ID and bandwidth margin in the feedback data and USB interface data; Determine whether the feedback tag corresponding to the feedback data has a feedback priority; the feedback priority is consistent with the touch priority; Yes, extract the feedback data with the highest feedback priority and its data size in turn; obtain the feedback bandwidth estimation value through the bandwidth prediction model; It is determined in sequence whether the feedback bandwidth estimation value corresponding to the maximum feedback priority value is less than the maximum bandwidth margin corresponding to the feedback channel ID. If yes, the feedback data corresponding to the maximum feedback priority value is loaded into the input channel corresponding to the input channel ID; if no, the feedback data is split into a number of feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID by using a data splitting algorithm; and the feedback sub-data is loaded into the feedback channel corresponding to each feedback channel ID; If no, obtain the feedback bandwidth estimation value through the bandwidth prediction model; determine whether the feedback bandwidth estimation value is less than the maximum bandwidth margin corresponding to the feedback channel ID; if yes, load the feedback data into the input channel corresponding to the input channel ID; if no, divide the feedback data into several feedback sub-data satisfying the bandwidth margin corresponding to each feedback channel ID through the data division algorithm; load the feedback sub-data into the feedback channel corresponding to each feedback channel ID.

14. A touch data transmission optimization method based on a USB interface, applied to a touch data transmission optimization system based on a USB interface according to any one of claims 1 to 13, characterized in that: include: S1: Get touch data and USB interface data; S2: obtaining touch content according to the touch data; selecting a transmission interface according to the touch content and USB interface data; S3: Obtain feedback prediction labels through the feedback prediction model; dynamically adjust the current prediction cache area according to the feedback prediction labels and the previous prediction cache area; S4: searching for corresponding feedback data in the prediction buffer area and the database according to the touch content; selecting a feedback interface according to the feedback data and the USB interface data; S5: Present the feedback data in a visual way; issue prompts based on alarm signals and contact management personnel.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the touch data transmission optimization system based on the USB interface as described in any one of claims 1 to 13 is implemented.

Images