Data transmission method and apparatus, electronic device, storage medium, and program product

Abstract

The application provides a data transmission method and device, electronic equipment, storage medium and program product, and belongs to the technical field of data transmission. The feature data segment is used to represent the characteristics of type data, the real-time transmission requirement is obtained with the minimum data, the data amount of type data processing is reduced, and the calculation rate of the real-time transmission requirement is improved. The real-time transmission requirement is automatically obtained through the transmission requirement model, and the efficiency of determining the real-time transmission requirement is improved. The application comprehensively considers the type information, the previous channel state information and the matching transmission channel of the real-time transmission requirement matching type data, improves the accuracy of determining the matching transmission channel, and is conducive to improving the transmission success rate and transmission efficiency of multi-source data. The matching transmission channel is automatically determined through the channel matching model, and the efficiency of determining the matching transmission channel is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data transmission, and in particular to a data transmission method and device, electronic equipment, storage medium and program product. BACKGROUND

[0002] For the collection and transmission of multi-source data, a plurality of sensors for collecting different types of data are usually placed at the data source, and then a plurality of types of data are collected through these sensors, and the data is centrally packaged and sent to the data receiving end in the form of a data packet queue through a wireless network, so that the data receiving end can remotely and real-timely obtain the multi-type data information at the data source.

[0003] In the above multi-source data transmission process, different types of data have different transmission requirements, resulting in low data transmission efficiency. SUMMARY

[0004] The present application provides a data transmission method, device, electronic equipment, storage medium and program product to solve the defect of low multi-source data transmission efficiency in the prior art and improve the transmission efficiency of multi-source data.

[0005] In a first aspect, the present application provides a data transmission method, comprising: obtaining a characteristic data segment of at least one type of data at a data source, inputting the characteristic data segment into a transmission requirement model, and obtaining a real-time transmission requirement of the type of data corresponding to the characteristic data segment output by the transmission requirement model; wherein the transmission requirement model is obtained by training a first preset model based on sample characteristic data segments and sample real-time transmission requirement labels; inputting type information of the type of data, previous channel state information of a plurality of transmission channels in a previous data transmission, and real-time transmission requirement of the type of data into a channel matching model, and obtaining a matching transmission channel of the type of data output by the channel matching model; wherein the channel matching model is obtained by training a second preset model based on a sample information set of sample type data and sample matching transmission channel labels, and the sample information set includes sample type information, sample previous channel state information and sample real-time transmission requirement; in the current data transmission, the corresponding type of data is transmitted based on the matching transmission channel.

[0006] In one embodiment, after obtaining the matching transmission channel of the type of data output by the channel matching model, the data transmission method further comprises: if the current channel state information of the matching transmission channel does not match the real-time transmission requirement in the current data transmission, inputting the current channel state information of a plurality of idle transmission channels, the real-time transmission requirement of the type of data and the type information of the type of data into the channel matching model, obtaining an updated transmission channel of the type of data output by the channel matching model; and in the current data transmission, the corresponding type of data is transmitted based on the updated transmission channel.

[0007] In one embodiment, the characteristic data segment of at least one type data of the data source is obtained by segmenting the type data based on a collection device or a collection time of the type data to obtain a plurality of data segments, and taking a data segment where the most important data is located as the characteristic data segment, or taking a data segment where the most similar and most data is located as the characteristic data segment.

[0008] In one embodiment, the corresponding type data is transmitted based on the matching transmission channel, including: performing spectrum segmentation on the matching transmission channel to obtain a plurality of matching sub-channels; sorting the collection data in the type data according to the importance; sorting the matching sub-channels according to the good and bad of the sub-channel state, and the maximum transmission capacity and the maximum transmission rate of the good sub-channel state are high; performing data splitting on the sorted type data based on the product value of the maximum transmission capacity of the sorted matching sub-channels and the preset safe transmission parameter to obtain a plurality of data packets, the data packet corresponds to the matching sub-channel one by one, the important collection data corresponds to the data packet of the matching sub-channel with the good matching sub-channel state, and the data packet includes a head data packet, a tail data packet and an intermediate data packet; identifying the data packet based on the data packet number, the total number of data packets, the data packet type and the completion information to obtain the identified data packet; and transmitting the identified data packet based on the matching sub-channel.

[0009] In one embodiment, the data packet is determined by: dividing the sorted type data into a plurality of unit data; and allocating the unit data to the sorted matching sub-channels according to the product value of the maximum transmission capacity and the preset safe transmission parameter, wherein the unit data corresponding to the important collection data is allocated to the matching sub-channel with the good sub-channel state; and taking the unit data allocated to the same matching sub-channel as a data packet.

[0010] In one embodiment, the transmission requirement model is used to obtain the real-time transmission requirement of the type data: performing feature extraction on the characteristic data segment to obtain extracted features; and analyzing the real-time transmission requirement of the characteristic data segment based on the extracted features, wherein the real-time transmission requirement at least includes transmission delay, anti-packet loss performance, bandwidth resource, transmittable signal frequency range and security mechanism.

[0011] In a second aspect, the present application also provides a data transmission apparatus, comprising: a real-time transmission requirement determination module, configured to acquire characteristic data segments of at least one type of data of a data source, input the characteristic data segments into a transmission requirement model, and acquire real-time transmission requirements of the type of data corresponding to the characteristic data segments output by the transmission requirement model; wherein the transmission requirement model is obtained by training a first preset model based on sample characteristic data segments and labels of sample real-time transmission requirements; a matching transmission channel determination module, configured to input type information of the type of data, previous channel state information of a plurality of transmission channels in a previous data transmission, and the real-time transmission requirements of the type of data into a channel matching model, and acquire a matching transmission channel of the type of data output by the channel matching model; wherein the channel matching model is obtained by training a second preset model based on a sample information set of sample type data and labels of sample matching transmission channels, and the sample information set comprises sample type information, sample previous channel state information, and sample real-time transmission requirements; and a transmission module, configured to transmit the corresponding type of data based on the matching transmission channel in a current data transmission.

[0012] In a third aspect, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements any of the above data transmission methods when executing the computer program.

[0013] In a fourth aspect, the present application also provides a non-transitory computer readable storage medium, having a computer program stored thereon, and the computer program is executable on a processor to implement any of the above data transmission methods.

[0014] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, and the computer program is executable on a processor to implement any of the above data transmission methods.

[0015] The data transmission method, apparatus, electronic device, storage medium, and program product provided by the present application represent the characteristics of the type of data by the characteristic data segments, realize the acquisition of real-time transmission requirements with the least amount of data, reduce the amount of data processed for the type of data, and improve the calculation rate of the real-time transmission requirements. The real-time transmission requirements are automatically acquired by the transmission requirement model, and the efficiency of determining the real-time transmission requirements is improved. The matching transmission channel of the type of data is matched by comprehensively considering the type information, the previous channel state information, and the real-time transmission requirements, and the accuracy of determining the matching transmission channel is improved, which is conducive to improving the transmission success rate and transmission efficiency of the multi-source data. The matching transmission channel is automatically determined by the channel matching model, and the efficiency of determining the matching transmission channel is improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0017] Figure 1 is one of the flowcharts of the data transmission method provided by the present application.

[0018] Figure 2 is another flowchart of the data transmission method provided by the present application.

[0019] Figure 3 is a structural schematic diagram of the data transmission system provided by the present application.

[0020] Figure 4 is a structural schematic diagram of the data transmission device provided by the present application.

[0021] Figure 5 is a structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely in the following with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the present application.

[0023] The data transmission method, device and electronic device of the present application will be described below with reference to the drawings. Figures 1-5

[0024] Figure 1 is one of the flowcharts of the data transmission method provided by the present application, as shown in Figure 1 the data transmission method comprises steps S100 to S300, and each step is as follows.

[0025] S100: Obtain the feature data segment of at least one type of data in the data source, input the feature data segment into the transmission requirement model, and obtain the real-time transmission requirement of the type of data corresponding to the feature data segment output by the transmission requirement model.

[0026] Among them, the transmission requirement model is obtained on the basis of the first preset model based on the sample feature data segment and the label of the sample real-time transmission requirement.

[0027] ​Multi-source data is acquired by installing multiple sensors at the data source location to collect different data types. For example, if the data source is within an automated factory, sensors are typically installed to collect various data types, monitoring the operational status of different equipment within the factory. These sensors include position sensors, image sensors, and operational data sensors. Side-side devices are also located at the data source location to process the acquired data.

[0028] Multi-source data is broken down into multiple data types. Edge devices group data collected from sensors of the same type or function into a single data type. Each data type has its primary transmission requirements; for example, location data requires high reliability and low latency during transmission, while environmental sensing data requires higher bandwidth and can have lower real-time requirements. However, these requirements are broad and not precise enough, and the same type of data collected at different times varies. Therefore, the transmission requirements for the same type of data fluctuate across different times, making it necessary to determine the real-time transmission requirements for each data type.

[0029] Obtain feature data segments of at least one type of data from the data source. Specifically, segment the type of data based on the acquisition device or acquisition time to obtain multiple data segments; take the data segment containing the most important data as the feature data segment, or take the data segment containing the data with the highest and most similarity as the feature data segment.

[0030] like Figure 2 As shown, the type of data is segmented according to the acquisition device or acquisition time, resulting in multiple data segments. Within the same type of sensor, the data acquired by one sensor is considered as one data segment. Alternatively, within the same type of sensor, data acquired within the same time period is considered as one data segment. For example, type 2 data is segmented into data segment 2-1, data segment 2-2, ..., data segment 2-M.

[0031] After obtaining multiple data segments, a feature data segment that can be used to characterize this type of data is selected from among them. For example, the data segment containing the most important data in the type of data can be used as the feature data segment. Alternatively, among multiple data segments of the type of data, the data segment containing the most similar and abundant collected data (public data) can be used as the feature data segment for this type of data.

[0032] This application characterizes the data features of type data through feature data segments, realizes the real-time transmission requirements of type data based on minimal data acquisition, and improves the efficiency of determining real-time transmission requirements.

[0033] The transmission requirement model is used to obtain the real-time transmission requirement of the type data: feature extraction is performed on the feature data segment to obtain extracted features; and the real-time transmission requirement of the feature data segment is analyzed based on the extracted features, and the real-time transmission requirement at least includes transmission delay, anti-packet loss performance, bandwidth resource, transmittable signal frequency range and security mechanism.

[0034] The feature data segment of each type of data is determined, and the feature data segment (for example, feature data segment 1, feature data segment 2,..., feature data segment N, etc.) of each type of data is input into the pre-constructed transmission requirement model one by one. The transmission requirement model is an analysis model pre-constructed by using machine learning or the like, which is used to perform feature extraction on different feature data segments to analyze the requirements of various channel parameters (for example, transmission delay, anti-packet loss performance, bandwidth resource, transmittable signal frequency range, security mechanism, etc.) in the transmission channel during data transmission, so as to obtain the real-time transmission requirement (for example, the quantified requirement score value of each channel parameter of each type of data) of each type of data.

[0035] The transmission requirement model is obtained based on the sample feature data segment and the label of the sample real-time transmission requirement on the basis of the first preset model. A plurality of sample feature data segments and corresponding sample real-time transmission requirements are obtained according to the analysis process of the historical feature data segment. The sample feature data segment is labeled according to the sample real-time transmission requirement to obtain sample transmission requirement data carrying labels, and then a sample transmission requirement data set is obtained. The sample transmission requirement data set is divided into a training set and a test set according to a preset ratio (for example, 6:4). The first preset model is constructed according to the transmission requirement of the type data, the first preset model is trained according to the training set, and the first preset model is tested according to the test set. When the error between the real-time transmission requirement output by the first preset model and the label of the sample real-time transmission requirement is less than a set threshold, it is determined that the training of the first preset model is completed, and the transmission requirement model is obtained.

[0036] According to the transmission requirement model, the real-time transmission requirement of the type data is automatically obtained, and the efficiency and accuracy of obtaining the real-time transmission requirement are improved.

[0037] S200: inputting the type information of the type data, the previous channel state information of the plurality of transmission channels in the previous data transmission, and the real-time transmission requirement of the type data into a channel matching model, and obtaining the matching transmission channel of the type data output by the channel matching model.

[0038] The channel matching model is obtained based on the sample information set and the label of the sample matching transmission channel of the sample type data on the basis of the second preset model, and the sample information set includes the sample type information, the sample previous channel state information and the sample real-time transmission requirement.

[0039] Obtain type information of type data, such as location type, environment perception type, etc. Obtain previous channel state information in a previous data transmission continuous with the current data transmission. Input type information of each type data (e.g., type information 1, type information 2, …, type information N), real-time transmission requirements of type data (e.g., real-time transmission requirement 1, real-time transmission requirement 2, …, real-time transmission requirement N), and previous channel state information (e.g., previous channel state information 1, previous channel state information 2, …, previous channel state information K) into a channel matching model. The channel matching model respectively extracts features from the type information, real-time transmission requirements, and previous channel state information, and comprehensively analyzes each feature extraction result to match a matching transmission channel for each type of data.

[0040] The channel matching model is an analysis model constructed in advance using machine learning technology. A second preset model is constructed according to the characteristics of the channel matching process. A plurality of sample information sets and corresponding sample matching transmission channels are obtained according to the transmission process of historical data. The sample information set includes sample type information, sample previous channel state information, and sample real-time transmission requirements. The sample information set is labeled according to the sample matching transmission channel to obtain sample channel matching data carrying labels, and then a sample channel matching data set is obtained. The sample channel matching data set is divided into a training set and a test set according to a preset ratio (e.g., 6:4). The second preset model is trained according to the training set and tested according to the test set until the error between the matching transmission channel output by the second preset model and the sample matching transmission channel is less than a set threshold, the training of the second preset model is determined to be complete, and the channel matching model is obtained.

[0041] S300: In the current data transmission, the corresponding type data is transmitted based on the matching transmission channel.

[0042] The corresponding type data is transmitted based on the matching transmission channel. Specifically, the matching transmission channel is spectrally segmented to obtain a plurality of matching sub-channels; the collected data in the type data is sorted according to importance; the matching sub-channels are sorted according to the goodness of the sub-channel state, and the maximum transmission capacity and the maximum transmission rate of the good sub-channel state are high; the sorted type data is data split based on the product value of the maximum transmission capacity of the sorted matching sub-channels and the preset safe transmission parameter, to obtain a plurality of data packets, the data packets correspond one-to-one to the matching sub-channels, the important collected data correspond to the data packets of the matching sub-channels with good matching sub-channel states, and the data packets include head data packets, tail data packets, and intermediate data packets; the data packets are identified based on the data packet number, the total number of data packets, the data packet type, and the completion information to obtain identified data packets; and the identified data packets are transmitted based on the matching sub-channels.

[0043] The matching sub-channels corresponding to the type data are spectrally segmented to obtain a plurality of matching sub-channels. For example, the bandwidth of each matching sub-channel is divided into a plurality of matching sub-channels.

[0044] To further improve the transmission success rate of important collected data in multi-source data transmission, the collected data in the type data is sorted according to the importance before transmission. For example, in the working data of the automation equipment, the collected data in the same type data is sorted according to the importance according to the importance of different automation equipment.

[0045] The matching sub-channels are sorted according to the good or bad of the sub-channel state. The maximum transmission capacity and the maximum transmission rate of the good sub-channel state are high. Based on the sorted matching sub-channels and the preset safe transmission parameter, the sorted type data is data split to obtain a plurality of data packets. According to the maximum transmission capacity of the matching sub-channel and the preset safe transmission parameter, the type data is split, and the collected data with high importance in the type data is preferentially allocated to the matching sub-channel with good sub-channel state. The preset safe transmission parameter is a value in the interval [0, 1]. The size of the preset safe transmission parameter can be set according to user demand, or can be set according to the historical fluctuation value of the matching sub-channel. In theory, in order to maximize the utilization rate of data transmission in the wireless network, the optimal case of data packet splitting is that the size of the split data packet is just corresponding to the maximum transmission capacity of the matching sub-channel. However, in the actual wireless network, the maximum transmission capacity of the matching sub-channel usually fluctuates, and if the optimal data packet splitting described above is used, there may be a data transmission moment when the maximum transmission capacity of the fluctuated matching sub-channel is less than the data size of the pre-divided data packet, thereby causing the transmission failure of the data packet. Therefore, in order to further improve the reliability of data transmission, the type data is data split according to the product of the maximum transmission capacity of the matching sub-channel at the division moment and the preset safe transmission parameter to obtain a data packet. A data packet corresponds to a matching sub-channel, and the data size of the data packet corresponds to the product of the maximum transmission capacity of the matching sub-channel and the preset safe transmission parameter. For example, the product of the maximum transmission capacity of the matching sub-channel 1 and the preset safe transmission parameter is 1G, the product of the matching sub-channel 2 is 2G, and the product of the matching sub-channel 3 is 3G. Then the type data is divided into 1G data packet, 2G data packet and 3G data packet.

[0046] Optionally, determining the data packet further includes: dividing the sorted type data into a plurality of unit data; allocating the unit data to the sorted matching sub-channels according to the product of the maximum transmission capacity and the preset safe transmission parameter; wherein the unit data corresponding to the important collected data is allocated to the matching sub-channel with good sub-channel state; and the unit data allocated to the same matching sub-channel is taken as a data packet.

[0047] The sorted type data is uniformly split to obtain a plurality of unit data, each unit data having the same data size. The plurality of unit data is allocated to the same matching sub-channel of the sorted type data according to the product value of the maximum transmission capacity and the preset safe transmission parameter. The unit data allocated to the same matching sub-channel is taken as a data packet. For example, the size of each unit data is 1G. The product value corresponding to the matching sub-channel 1 is 1G, the product value corresponding to the matching sub-channel 2 is 2G, and the product value corresponding to the matching sub-channel 3 is 3G. Then, one unit data is allocated to the matching sub-channel 1, two unit data are allocated to the matching sub-channel 2, and three unit data are allocated to the matching sub-channel 3. The unit data of the same matching sub-channel is taken as a data packet.

[0048] The application can not only realize uniform splitting of the type data by the unit data, but also meet the on-demand allocation of the unit data to the matching sub-channel, which is conducive to improving the efficiency of determining the data packet.

[0049] The data packet of each type data includes a head data packet, a plurality of intermediate data packets, and a tail data packet. Each data packet corresponds to a matching sub-channel. The head data packet, the tail data packet, and the intermediate data packet are identified in different formats. For example, each data packet carries data packet number and data packet type information, and the head data packet further includes the total number of data packets, and the tail data packet further includes completion information (for example, completion identification). Each matching sub-channel transmits the corresponding identified data packet.

[0050] The application splits the sorted type data by the sorted matching sub-channel, ensures that important data is matched to a good matching sub-channel, and is conducive to improving the transmission success rate of important data in the type data. According to the transmission of the corresponding data packet by the matching sub-channel, the accurate speed of each data packet is realized, which is conducive to improving the transmission efficiency of the type data. The transmission capacity of the plurality of sub-channels of each matching transmission channel is utilized, and data transmission is performed in parallel, thereby improving the transmission rate of the multi-source data, and avoiding data congestion and blockage.

[0051] Each matching sub-channel transmits the corresponding identified data packet. When receiving the data packets of various types of data, the data receiving end can store the real-time received data packets in a buffer queue, and recombine and verify the data packets according to the data packet number, the total number of data packets, the data packet type, and the completion information.

[0052] Further, the parallel transmission method of multi-source data of the application further provides a request retransmission mechanism. The data receiving end confirms whether the correct data packet is received through the data packet number, and the missing data packet can be triggered for retransmission through the data packet number. The data receiving end combines all received data packets into various types of data according to the combination method, and then combines all types of data into multi-source data sent by the sending end. Then, the data receiving end performs corresponding data processing and analysis operations according to the obtained multi-source data. For example, backup of video stream data, verification of preliminary processing results, generation of control instructions according to device working data, etc.

[0053] The data transmission method provided by the embodiment of the application represents the characteristics of the type data by the feature data segment, realizes the real-time transmission requirement with the minimum data, reduces the data amount of the type data processing, and improves the calculation rate of the real-time transmission requirement. The real-time transmission requirement is automatically obtained through the transmission requirement model, which improves the efficiency of determining the real-time transmission requirement. The matching transmission channel of the type data is matched by comprehensively considering the type information, the previous channel state information and the real-time transmission requirement, which improves the accuracy of determining the matching transmission channel, and further improves the transmission success rate and the transmission efficiency of the multi-source data. The matching transmission channel is automatically determined through the channel matching model, which improves the efficiency of determining the matching transmission channel.

[0054] Based on the above embodiment, after obtaining the matching transmission channel of the type data output by the channel matching model, the data transmission method further includes steps S400 to S500, which are specifically as follows.

[0055] S400: If the current channel state information of the matching transmission channel and the real-time transmission requirement do not match in the current data transmission, input the current channel state information of the plurality of idle transmission channels, the real-time transmission requirement of the type data and the type information of the type data into the channel matching model, and obtain the updated transmission channel of the type data output by the channel matching model.

[0056] S500: In the current data transmission, the corresponding type data is transmitted based on the updated transmission channel.

[0057] The channel state information in the wireless communication environment is generally time-varying fluctuation, so the previous channel state information in the previous data transmission and the current channel state information in the current data transmission are deviated in matching the transmission channel. Therefore, in order to improve the transmission reliability of multi-source data, the current channel state information is obtained in real time during the current data transmission after the transmission channel is matched for each type of data and the data is transmitted. When it is determined that at least one current channel state information of the matched transmission channel does not meet the real-time transmission requirement of the type data of the corresponding transmission channel, the transmission channel matching for the type data is re-performed. The real-time transmission requirement of the type data, type information and current channel state information of a plurality of idle transmission channels are input into the above-mentioned channel matching model for data analysis to obtain an updated transmission channel of the type data. In the subsequent process of the current data transmission, the type data is transmitted through the updated transmission channel.

[0058] The embodiment of the present application realizes the timely updating of the transmission channel by matching the current channel state information and the real-time transmission requirement, which is beneficial to improve the transmission reliability of multi-source data. By inputting the current channel state information of a plurality of idle transmission channels, the real-time transmission requirement and the type information into the channel matching model, the updated transmission channel is obtained, that is, the factors affecting the transmission channel are considered, and the updated transmission channel is automatically obtained, which is beneficial to improve the accuracy and efficiency of the updated transmission channel.

[0059] In order to further analyze and describe the data transmission method provided by the embodiment of the present application, the following embodiments and Figure 3 are used for description.

[0060] As Figure 3 shown, a data transmission system includes a multi-source data parallel transmission agent, a data transmission channel and a data aggregation server. The multi-source data parallel transmission agent includes a data splitting module, a data channel management module and a data sending module. The data transmission channel includes a plurality of transmission channels, such as channel 1, channel 2, channel 3, etc. The data aggregation server includes a data receiving module and a data aggregation window calculation module. The multi-source data parallel transmission agent is a fast-deployable edge computing agent service, which can receive type data reported from different types of devices, and process the type data into standard multi-source data processing packages (data packages) through analysis and splitting. The data aggregation server can support configuring data packages of specified transmission channels, and hand over the data packages to the data aggregation window calculation module for processing. The data aggregation window calculation module uses a time window function to associate and combine the data packages arriving within the window time range.

[0061] The various sensors, such as smoke sensors, temperature and humidity sensors, special gas sensors, etc. generate type data at regular intervals and transmit the type data to the edge computing server. The edge computing service determines the matching transmission channel according to the sensor type and the type of generated data through a matching algorithm, such as selecting a temperature data channel for temperature data and a humidity channel for humidity data, and transmits the data through a dedicated channel. The edge computing server splits the type data according to the sensor type and the data type, such as temperature data packets, humidity data packets, and air pressure data packets, and encapsulates them into specific data models. The temperature data packets, humidity data packets, and sensor types are input into the channel rule engine to obtain the channel number of the matching transmission channel (including the main channel number and the alternative channel number). The data packets are sent through the main channel number and the sending result is monitored. If the sending fails, the data packets are sent through the alternative channel number.

[0062] The data packets are sent to the matching transmission channel for data transmission, and the data transmission status is monitored at any time. The data receiving end supports simultaneous configuration of multiple channel numbers for data reception, and the received data packets are aggregated and associated through a time window aggregation function. For example, temperature data packets, humidity data packets, and air pressure data packets from different channel numbers are aggregated according to the device unique identifier and data generation time to form the final complete data, which is sent to the backend data service for processing.

[0063] The present application utilizes the transmission requirements of different types of data to allocate transmission channels for each type of data, which can further improve the transmission success rate of each type of data. The present application utilizes the previous channel state information to estimate the channel estimation value for the next data transmission, and utilizes the channel estimation value and the real-time transmission requirement of each type of data and the type information of the type data to more accurately match the corresponding matching transmission channel for each type of data, thereby improving the transmission success rate of multi-source data. The present application utilizes the transmission capacity of multiple sub-channels of each transmission channel to perform parallel data transmission, thereby improving the transmission rate of multi-source data and avoiding data congestion and blocking.

[0064] The data transmission device provided by the present application is described below. The data transmission device described below can be referred to in conjunction with the data transmission method described above.

[0065] As shown in FIG. 4, Figure 4 A data transmission device includes a real-time transmission requirement determination module 401 for obtaining a feature data segment of at least one type of data from a data source, inputting the feature data segment into a transmission requirement model, and obtaining the real-time transmission requirement of the type of data corresponding to the feature data segment output by the transmission requirement model. The transmission requirement model is trained based on sample feature data segments and sample real-time transmission requirements based on a first preset model.

[0066] The matching transmission channel determination module 402 is configured to input type information of the type data, previous channel state information of the plurality of transmission channels in previous data transmission, and real-time transmission requirement of the type data into a channel matching model, and obtain a matching transmission channel of the type data output by the channel matching model; the channel matching model is obtained based on sample information set and sample matching transmission channel label of sample type data based on the second preset model, and the sample information set includes sample type information, sample previous channel state information, and sample real-time transmission requirement.

[0067] The transmission module 403 is configured to transmit corresponding type data based on the matching transmission channel in the current data transmission.

[0068] The data transmission device provided by the embodiment of the application realizes obtaining the real-time transmission requirement with the least data by representing the characteristics of the type data by the feature data segment, reduces the data amount for processing the type data, and improves the calculation rate of the real-time transmission requirement. The real-time transmission requirement is automatically obtained by the transmission requirement model, and the efficiency of determining the real-time transmission requirement is improved. The matching transmission channel of the type data is matched by comprehensively considering the type information, the previous channel state information, and the real-time transmission requirement, the accuracy of determining the matching transmission channel is improved, and then the transmission success rate and the transmission efficiency of the multi-source data are improved. The matching transmission channel is automatically determined by the channel matching model, and the efficiency of determining the matching transmission channel is improved.

[0069] In one embodiment, the matching transmission channel determination module 402 is further configured to: if the current channel state information of the matching transmission channel and the real-time transmission requirement do not match in the current data transmission, input the current channel state information of the plurality of idle transmission channels, the real-time transmission requirement of the type data, and the type information of the type data into the channel matching model, and obtain an updated transmission channel of the type data output by the channel matching model; and transmit corresponding type data based on the updated transmission channel in the current data transmission.

[0070] In one embodiment, the real-time transmission requirement determination module 401 is configured to: segment the type data based on a collection device or a collection time of the type data to obtain a plurality of data segments; and take a data segment where the most important data is located as a feature data segment, or take a data segment where the most similar and most data is located as a feature data segment.

[0071] In one embodiment, the transmission module 403 is configured to: perform spectrum segmentation on the matched transmission channel to obtain a plurality of matched sub-channels; sort the collection data in the type data according to importance; sort the matched sub-channels according to the sub-channel state, and the maximum transmission capacity and the maximum transmission rate of the good sub-channel state are high; perform data splitting on the sorted type data based on the product of the maximum transmission capacity of the sorted matched sub-channels and the preset safe transmission parameter, to obtain a plurality of data packets, the data packet corresponds to the matched sub-channel in one-to-one correspondence, the important collection data corresponds to the data packet of the matched sub-channel with good state, and the data packet includes a head data packet, a tail data packet and an intermediate data packet; identify the data packet based on the data packet number, the total number of data packets, the data packet type and the completion information, to obtain an identified data packet; and transmit the identified data packet based on the matched sub-channel.

[0072] In one embodiment, the transmission module 403 is further configured to: divide the sorted type data into a plurality of unit data; and allocate the unit data to the sorted matched sub-channels according to the product of the maximum transmission capacity and the preset safe transmission parameter; wherein the unit data corresponding to the important collection data is allocated to the matched sub-channel with good state; and take the unit data allocated to the same matched sub-channel as a data packet.

[0073] In one embodiment, the real-time transmission requirement determination module 401 is configured to: perform feature extraction on the feature data segment to obtain extracted features; and analyze the real-time transmission requirement of the feature data segment based on the extracted features, and the real-time transmission requirement at least includes transmission delay, anti-packet loss performance, bandwidth resource, transmittable signal frequency range and security mechanism.

[0074] Figure 5 An example of an entity structure diagram of an electronic device is shown in FIG. 1. Figure 5As shown, the electronic device can include a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 complete communication with each other through the communications bus 540. The processor 510 can invoke a logic instruction in the memory 530 to execute a data transmission method, the method comprising: obtaining a feature data segment of at least one type of data of a data source, inputting the feature data segment into a transmission requirement model, and obtaining a real-time transmission requirement of the type of data corresponding to the feature data segment output by the transmission requirement model; wherein the transmission requirement model is obtained by training based on sample feature data segments and labels of sample real-time transmission requirements on the basis of a first preset model; inputting type information of the type of data, previous channel state information of a plurality of transmission channels in a previous data transmission, and the real-time transmission requirement of the type of data into a channel matching model, and obtaining a matching transmission channel of the type of data output by the channel matching model; wherein the channel matching model is obtained by training based on a sample information set of sample type data and sample matching transmission channels on the basis of a second preset model, and the sample information set includes sample type information, sample previous channel state information, and sample real-time transmission requirement; and in the current data transmission, transmitting the corresponding type of data based on the matching transmission channel.

[0075] In addition, the logic instruction in the memory 530 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or in part or parts of the technical solutions that make contributions to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0076] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program being stored in a non-transitory computer-readable storage medium, and the computer program being executable by a processor to enable the computer to perform the data transmission method provided by any of the above methods, the method comprising: obtaining a feature data segment of at least one type of data from a data source, inputting the feature data segment into a transmission requirement model, and obtaining real-time transmission requirements of the type of data corresponding to the feature data segment output by the transmission requirement model; wherein the transmission requirement model is obtained by training a first preset model based on sample feature data segments and labels of sample real-time transmission requirements; inputting type information of the type of data, previous channel state information of a plurality of transmission channels in a previous data transmission, and the real-time transmission requirements of the type of data into a channel matching model, and obtaining a matching transmission channel of the type of data output by the channel matching model; wherein the channel matching model is obtained by training a second preset model based on a sample information set of sample type data and labels of sample matching transmission channels, and the sample information set comprises sample type information, sample previous channel state information, and sample real-time transmission requirements; and in a current data transmission, transmitting the corresponding type of data based on the matching transmission channel.

[0077] In another aspect, the present application also provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executable by a processor to implement the data transmission method provided by any of the above methods, the method comprising: obtaining a feature data segment of at least one type of data from a data source, inputting the feature data segment into a transmission requirement model, and obtaining real-time transmission requirements of the type of data corresponding to the feature data segment output by the transmission requirement model; wherein the transmission requirement model is obtained by training a first preset model based on sample feature data segments and labels of sample real-time transmission requirements; inputting type information of the type of data, previous channel state information of a plurality of transmission channels in a previous data transmission, and the real-time transmission requirements of the type of data into a channel matching model, and obtaining a matching transmission channel of the type of data output by the channel matching model; wherein the channel matching model is obtained by training a second preset model based on a sample information set of sample type data and labels of sample matching transmission channels, and the sample information set comprises sample type information, sample previous channel state information, and sample real-time transmission requirements; and in a current data transmission, transmitting the corresponding type of data based on the matching transmission channel.

[0078] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0079] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary general hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0080] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data transmission method, characterized by, The method comprises the following steps: obtaining a feature data segment of at least one type data of a data source, inputting the feature data segment into a transmission requirement model, and obtaining real-time transmission requirements of the type data corresponding to the feature data segment output by the transmission requirement model; wherein the transmission requirement model is obtained by training a first preset model based on sample feature data segments and labels of sample real-time transmission requirements; inputting type information of the type data, previous channel state information of a plurality of transmission channels in previous data transmission, and real-time transmission requirements of the type data into a channel matching model, and obtaining a matching transmission channel of the type data output by the channel matching model; wherein the channel matching model is obtained by training a second preset model based on a sample information set of sample type data and labels of sample matching transmission channels, and the sample information set comprises sample type information, sample previous channel state information, and sample real-time transmission requirements; in the current data transmission, performing spectrum segmentation on the matching transmission channel to obtain a plurality of matching sub-channels; sorting the collected data in the type data according to importance; sorting the matching sub-channels according to the good and bad of the sub-channel state, and the maximum transmission capacity and the maximum transmission rate of the good sub-channel state are high; based on the product of the maximum transmission capacity of the sorted matching sub-channels and the preset safety transmission parameter, data splitting is performed on the sorted type data to obtain a plurality of data packets, and the data packet and the matching sub-channel correspond one by one, the important collected data correspond to the data packet matching sub-channel with good matching sub-channel state, and the data packet comprises a head data packet, a tail data packet, and an intermediate data packet; based on the data packet number, the total number of data packets, the data packet type, and the completion identifier, the data packet is identified to obtain an identified data packet; transmitting the identified data packet based on the matching sub-channel.

2. The data transmission method of claim 1, wherein, After obtaining the matching transmission channel of the type data output by the channel matching model, the method further comprises: if the current channel state information of the matching transmission channel and the real-time transmission requirements do not match in the current data transmission, inputting the current channel state information of a plurality of idle transmission channels, the real-time transmission requirements of the type data, and the type information of the type data into the channel matching model, and obtaining an updated transmission channel of the type data output by the channel matching model; in the current data transmission, transmitting the corresponding type data based on the updated transmission channel.

3. The data transmission method of claim 1, wherein, The method of obtaining a feature data segment of at least one type data of a data source comprises: segmenting the type data based on a collection device or collection time of the type data to obtain a plurality of data segments; the data segment where the most important data is located is taken as the feature data segment, or the data segment where the data with the highest similarity and the most is located is taken as the feature data segment.

4. The data transmission method of claim 1, wherein, The data packet further comprises: dividing the sorted type data into a plurality of unit data; According to the product of the maximum transmission capacity and the preset safety transmission parameter, the unit data is allocated to the sorted matching sub-channel; wherein, the unit data corresponding to the important acquisition data is allocated to the matching sub-channel with good sub-channel state; The unit data allocated to the same matching sub-channel is taken as one data packet.

5. The data transmission method of claim 1, wherein, The transmission requirement model is used to obtain the real-time transmission requirement of the type data: Feature extraction is performed on the feature data segment to obtain extracted features; Based on the extracted features, the real-time transmission requirement of the feature data segment is analyzed, and the real-time transmission requirement at least includes transmission delay, anti-packet loss performance, bandwidth resource, transmittable signal frequency range and safety mechanism.

6. A data transmission apparatus characterized by comprising: Comprise: A real-time transmission requirement determination module is configured to obtain a feature data segment of at least one type data from a data source, input the feature data segment into a transmission requirement model, and obtain the real-time transmission requirement of the type data corresponding to the feature data segment output by the transmission requirement model; wherein, the transmission requirement model is obtained based on sample feature data segments and sample real-time transmission requirement labels on the basis of a first preset model; A matching transmission channel determination module is configured to input type information of the type data, previous channel state information of a plurality of transmission channels in previous data transmission, and the real-time transmission requirement of the type data into a channel matching model, and obtain the matching transmission channel of the type data output by the channel matching model; wherein, the channel matching model is obtained based on a sample information set of sample type data and sample matching transmission channel labels on the basis of a second preset model, and the sample information set includes sample type information, sample previous channel state information and sample real-time transmission requirement; A transmission module is configured to, in current data transmission, perform spectrum segmentation on the matching transmission channel to obtain a plurality of matching sub-channels, sort acquisition data in the type data according to importance, sort the matching sub-channels according to sub-channel state, and perform data splitting on the sorted type data based on the product of the maximum transmission capacity of the sorted matching sub-channel and a preset safety transmission parameter to obtain a plurality of data packets, wherein, a data packet corresponds to a matching sub-channel, important acquisition data corresponds to a data packet matching a matching sub-channel with good matching sub-channel state, and the data packet includes a head data packet, a tail data packet and an intermediate data packet; the data packet is identified based on packet number, total number of data packets, data packet type and completion identifier to obtain an identified data packet; and the identified data packet is transmitted based on the matching sub-channel.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the data transmission method of any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the data transmission method of any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the data transmission method of any one of claims 1 to 5.

Images