Internet-of-things data transmission protocol adaptation method and system based on big data

Abstract

The invention belongs to the technical field of data transmission, and provides an internet-of-things data transmission protocol adaptation method and system based on big data, and the method comprises the steps: obtaining data transmission protocol adaptation features according to data collection features, data essential feature coefficients and protocol parameters of internet-of-things equipment; and according to the data transmission protocol adaptation characteristics and the corresponding transmission protocol adaptation method, constructing an Internet of Things data transmission adaptation model, using the model to carry out adaptive transmission protocol adaptation on subsequent collected data and output a real-time transmission protocol adaptation method, and according to the real-time transmission protocol adaptation method, controlling a protocol converter device. And thus, the data transmission of the internet of things is subjected to adaptive management. The conversion efficiency of the data adaptation transmission protocol is enhanced through multi-dimensional feature processing of the data of the internet of things and the improved activation function big data model, the transmission speed and quality of the data of the internet of things are improved, and the manual adaptation loss of the data transmission protocol is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of data transmission, and particularly relates to an Internet of Things data transmission protocol adaptation method and system based on big data. Background Art

[0002] In industrial intelligent manufacturing, there are various data acquisition source devices with different functional attributes and hardware compositions that need to be connected to the network, such as numerically controlled machine tools, industrial robots, automatic guided vehicles, cleaning equipment, inspection equipment, automated production lines, and automatic storage warehouses, etc., and also include industrial accessory devices such as controllers of multiple sensors / actuators. For diverse industrial production data, it is necessary to access the Internet for unified storage and analysis in order to monitor, control, and adjust the production process.

[0003] However, currently for multi-source heterogeneous device data, the collected data will be directly sent to the node server for storage and analysis, and coordinated and managed through the main service, and load balancing is achieved by combining technologies such as data caching and message queues. These methods can better handle the massive data acquisition tasks, but for the extended access of various heterogeneous acquisition devices and acquisition servers, the processes of server and acquisition device configuration, deployment, etc. are very cumbersome.

[0004] Through the above situation description, the problems that need to be solved in the prior art are as follows: How to fully consider the processing of targeted application scenarios for protocol adaptation affecting Internet of Things data transmission under the condition that the adaptation protocol library remains unchanged, how to perform specific processing on the data itself, the protocol itself, and even the vector features affecting data transmission, and use the protocol, data, and corresponding acquisition and transmission processes to construct a scenario-based neural network model, so as to use the model to perform adaptive transmission protocol adaptation on the collected data to output an instant transmission protocol adaptation method, and control the protocol converter device according to the instant transmission protocol adaptation method, so as to perform adaptation management on Internet of Things data transmission. For this reason, an Internet of Things data transmission protocol adaptation method and system based on big data are proposed. Summary of the Invention

[0005] To solve the above technical problems, the present invention proposes an Internet of Things data transmission protocol adaptation method and system based on big data.

[0006] In the first aspect of the present invention, an Internet of Things data transmission protocol adaptation method based on big data is provided: Upload and obtain the data acquisition characteristics, data essence characteristic coefficients, and protocol parameters of the transmitted Internet of Things data that affect the selection of the Internet of Things data adaptation protocol, and obtain the corresponding Internet of Things data transmission protocol adaptation method; Process the data acquisition characteristics, the data essential feature coefficients, and the protocol parameters of the transmitted IoT data to obtain data transmission protocol adaptation characteristics; Construct an IoT data transmission adaptation model according to the transmission protocol adaptation characteristics and the corresponding IoT data transmission protocol adaptation method; Use the IoT data transmission adaptation model to process the IoT data to be transmitted to obtain an IoT data transmission protocol adaptation method for the data to be transmitted; Based on the IoT data transmission protocol adaptation method for the data to be transmitted, perform adaptation management on the IoT data transmission.

[0007] Further, the data acquisition characteristics are obtained by processing the average acquisition transmission time, the data cache time, and the temperature difference coefficient between the data source and the data storage location.

[0008] Further, the data essential feature coefficients are obtained by processing the characteristics of the data itself, specifically by processing the data dimension, the data span size, and the data type.

[0009] Further, the protocol parameters include the adaptation scenario characteristics and the adaptation characteristic parameters of the protocol.

[0010] Further, the adaptation scenario characteristics obtain adaptation values according to the adaptation scenario. Specifically, according to the broadband and latency conditions in the network environment, the adaptation scenario characteristics are a two-dimensional matrix vector, and the adaptation characteristic parameters are obtained by processing the distance, power consumption, and data volume of the protocol adaptation.

[0011] Further, the process of processing the data acquisition characteristics, the data essential feature coefficients, and the protocol parameters of the transmitted IoT data to obtain data transmission protocol adaptation characteristics uses a vector dimensionality increase method. The vector dimensionality increase method represents the data acquisition characteristics 、the data essential feature coefficients and the protocol parameters in matrix form.

[0012] Further, the IoT data transmission adaptation model uses a neural network model with an improved activation function based on the adaptation scenario characteristics and the adaptation characteristic parameters.

[0013] Further, the improved activation function used by the IoT data transmission adaptation model based on the adaptation scenario characteristics and the adaptation characteristic parameters is; ; In the formula, S(X) is the activation function value, is the rank of the matrix of the adaptation scenario characteristics, is the adaptation characteristic parameter, is obtained by calculating the weighted sum of the transmission protocol adaptation characteristics through neurons plus the bias.

[0014] There is also provided an Internet of Things data transmission protocol adaptation system based on big data, including an Internet of Things device data feature acquisition module, an Internet of Things data feature module, a protocol adaptation feature module, an Internet of Things data transmission adaptation terminal module, a transmission protocol adaptation feature processing module, and an Internet of Things data transmission adaptation model construction module, characterized in that: The Internet of Things device data feature acquisition module: acquires the first data acquisition feature in data acquisition, and is also used to acquire the second data acquisition feature in data acquisition; The Internet of Things data feature module: acquires the first data essential feature coefficient, and is also used to acquire the second data essential feature coefficient; The protocol adaptation feature module: uploads the acquired first protocol parameter; The transmission protocol adaptation feature processing module: performs parameter feature processing on the first data acquisition feature, the first data essential feature coefficient, and the first protocol parameter to obtain the first data transmission protocol adaptation feature, and also performs parameter feature processing on the second data acquisition feature, the second data essential feature coefficient, and the first protocol parameter to obtain the second data transmission protocol adaptation feature; The Internet of Things data transmission adaptation model construction module: constructs an Internet of Things data transmission adaptation model according to the first data transmission protocol adaptation feature and the first Internet of Things data transmission protocol adaptation method; The Internet of Things data transmission adaptation terminal module: acquires the uploaded first Internet of Things data transmission protocol adaptation method, is also connected to the transmission protocol adaptation feature processing module and the Internet of Things data transmission adaptation model construction module, acquires the second data transmission protocol adaptation feature and processes to obtain the second Internet of Things data transmission protocol adaptation method, and controls the protocol converter device to perform an adaptive selection of the Internet of Things data transmission protocol.

[0015] Furthermore, the Internet of Things data transmission adaptation model uses a neural network model with an improved activation function based on adaptation scenario features and adaptation characteristic parameters; The improved activation function is; ; In the formula, S(X) is the activation function value, is the rank of the matrix of adaptation scenario features, is the adaptation characteristic parameter, is obtained by weighted addition of the bias of the transmission protocol adaptation feature through neurons.

[0016] The present invention fully considers that, under the condition that the adaptation protocol library remains unchanged, for the protocol adaptation that affects the transmission of Internet of Things data, a feature processing of the fusion nature of the data itself, the protocol itself, and even the vector features that affect data transmission is constructed. The activation function of the neural network model is constructed in a scenario-based manner by using the protocol, data, and the corresponding acquisition and transmission process, so as to use the model to perform adaptive transmission protocol adaptation on the acquired data and output an immediate transmission protocol adaptation method. The protocol converter device is controlled according to the immediate transmission protocol adaptation method, so as to perform adaptation management on the transmission of Internet of Things data. Through multi-dimensional feature processing of Internet of Things data and by using an improved activation function big data model to enhance the conversion efficiency of the data adaptation transmission protocol, the transmission speed and quality of Internet of Things data are improved, and the manual adaptation loss of the data transmission protocol is reduced.

[0017] More embodiments and improvement effects of the present invention will be further introduced in combination with the drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a diagram for training the model of a method for adapting the Internet of Things data transmission protocol based on big data according to the present invention; Figure 2 is a schematic diagram of a system for adapting the Internet of Things data transmission protocol based on big data according to the present invention; Figure 3 is a schematic diagram of the MQTT message transmission protocol in the present invention; Figure 4 is a schematic diagram of the improved activation function principle in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, with reference to the drawings and specific embodiments, the invention will be further described.

[0020] In the first aspect of the present invention, a method and system for adapting the Internet of Things data transmission protocol based on big data are proposed.

[0021] In the first aspect of the present invention, a method for adapting the Internet of Things data transmission protocol based on big data is provided: Upload and obtain the data acquisition features, data essence feature coefficients, and protocol parameters of the transmitted Internet of Things data that affect the selection of the adaptation protocol for Internet of Things data, and obtain the corresponding method for adapting the Internet of Things data transmission protocol; Process the data acquisition features, the data essence feature coefficients, and the protocol parameters of the transmitted Internet of Things data to obtain data transmission protocol adaptation features; Construct an Internet of Things data transmission adaptation model according to the transmission protocol adaptation features and the corresponding method for adapting the Internet of Things data transmission protocol; A method for adapting the transmission protocol of Internet of Things (IoT) data to be transmitted is obtained by processing the IoT data to be transmitted using an IoT data transmission adaptation model; Based on the method for adapting the transmission protocol of IoT data to be transmitted, the adaptation management of IoT data transmission is carried out.

[0022] There are various adapted transmission protocols, such as lightweight protocols (MQTT / CoAP).

[0023] MQTT: Based on the publish / subscribe model, supports QoS levels (0 - 2), suitable for low-bandwidth and unstable network environments (such as remote monitoring, sensor data transmission). Its retained message and will message mechanisms can improve the efficiency of device status management.

[0024] CoAP: Based on the RESTful architecture, designed specifically for resource-constrained devices, supports UDP transmission and the observer pattern, suitable for smart home and industrial control scenarios. The block transfer mechanism can optimize the transmission of large data fragments.

[0025] Comparison: MQTT pays more attention to reliability and message management, while CoAP emphasizes low power consumption and simple interaction.

[0026] Real-time and high-performance protocols (DDS / HTTP2) DDS: Oriented towards real-time systems, supports low-latency and high-throughput communication, suitable for fields with high real-time requirements such as autonomous driving and aerospace.

[0027] HTTP / 2: Improves transmission efficiency through multiplexing and header compression technologies, suitable for scenarios that require high-frequency interaction such as intelligent transportation systems.

[0028] Low-power wide-area network protocols (LoRaWAN / NB-IoT) LoRaWAN: Supports long-distance communication and ADR (Adaptive Data Rate), suitable for smart cities and agricultural monitoring.

[0029] NB-IoT: Based on cellular networks, has wide coverage and low cost, suitable for large-scale deployment scenarios such as smart metering and asset tracking.

[0030] In the present invention, according to actual needs, it is divided into two required categories according to the speed of data transmission, and further matching of the IoT data adaptation method is carried out based on the model.

[0031] Furthermore, the data acquisition characteristics are obtained by processing the average acquisition and transmission time, data caching time, and the temperature difference coefficient between the data source location and the data storage location, and its calculation formula is: ; In the formula, Regarding the data acquisition characteristics, in this application, through multiple experiments, it has been found that the longer the average acquisition and transmission time of the IoT data, the stronger the enhancement effect on the model value corresponding to the adaptation method obtained by training the model. And the longer the data caching time, the more the data transmission rate will be affected accordingly, so it shows an inverse ratio. is the average acquisition and transmission time of the IoT data. is the data caching time. is the temperature at the data source location. is the temperature at the data storage location, and the temperature difference coefficient is calculated and processed using the two.

[0032] Furthermore, the data essential feature coefficient is obtained by processing the characteristics of the data itself, specifically processed by the data dimension, data span size, and data type, and its calculation formula is: ; Since the essential features of the data other than knowing the data source can intuitively show the situation of the data, and the dimension, span size, and data type of the data have a significant impact on the adaptation of the IoT data transmission protocol. In this application, it has been found through experiments that the higher the data dimension and the larger the data span, the more difficult the data transmission is, and the corresponding value of the transmission protocol adaptation method is significantly enhanced. In the formula, is the data essential feature coefficient, is the data dimension, is the total span size of each dimension of the -dimensional data, is the set data type, and the corresponding tag value is set according to the needs of those skilled in the art for the data acquisition source. For example, agricultural data is set to 1, industrial data is set to 2, and service industry data is set to 3.

[0033] Furthermore, the protocol parameters include the adaptation scenario characteristics and adaptation feature parameters of the protocol.

[0034] Furthermore, the adaptation scenario characteristics obtain the adaptation value according to the adaptation scenario. Specifically, according to the broadband and latency conditions in the network environment, the adaptation scenario characteristics are a one- or two-dimensional matrix vector, specifically expressed as the matrix If the matrix is , it means high broadband and low latency. The adaptation feature parameters are calculated from the distance, power consumption, and data volume of the protocol adaptation:

[0035] Affected by the protocol itself, the adaptability of IoT data transmission will be different. Therefore, the model considers the situation of the protocol itself to correct the efficiency of the model output for adapting the protocol. In the formula, To adapt to the characteristic parameters, d is the distance adapted by the protocol. Since there is ultra-long-distance transmission in the distance adapted by the transmission protocol, in order to reduce the span of the input data of the training model and improve the robustness of the model, the logarithmic function is used to scale the distance value. And the power consumption of different protocols has different effects. To reflect the power consumption effects between different protocols, the average power consumption of P protocols is calculated. , is the maximum power consumption among the P protocols, and is the amount of transmitted data.

[0036] Furthermore, the processing of the data acquisition feature, the data essential feature coefficient, and the protocol parameter of the transmitted IoT data to obtain the data transmission protocol adaptation feature is processed by the vector dimension elevation method. The vector dimension elevation method represents the data acquisition feature , the data essential feature coefficient , and the protocol parameter in matrix form as , making the subsequent processing of the model more convenient and facilitating the processing of the neural network model.

[0037] Furthermore, the IoT data transmission adaptation model uses a neural network model with an improved activation function based on the adaptation scenario features and adaptation characteristic parameters.

[0038] Furthermore, the improved activation function used by the IoT data transmission adaptation model based on the adaptation scenario features and adaptation characteristic parameters is; ; In the formula, S(X) is the activation function value, is the rank of the matrix of the adaptation scenario features, which plays a role in correcting the activation function value, is the adaptation characteristic parameter. Since the adaptation protocol output by the model has a corresponding impact on the adaptation protocol itself, and few existing models consider the case of the IoT data adaptation protocol manually calibrated and selected in the database, therefore, by correcting the calculation of the activation function in the model based on the selection library of the adaptation protocol itself, it can be determined that the content sent to the next neuron of the neural network is more in line with the required IoT data transmission protocol. The transmission protocol adaptation feature is obtained by weighted addition of the neuron plus the bias.

[0039] A system for adapting IoT data transmission protocols based on big data is also provided, including an IoT device data feature acquisition module, an IoT data feature module, a protocol adaptation feature module, an IoT data transmission adaptation terminal module, a transmission protocol adaptation feature processing module, and an IoT data transmission adaptation model construction module: The IoT device data feature acquisition module: acquires the first data acquisition feature during data acquisition, and is also used to acquire the second data acquisition feature during data acquisition; The IoT data feature module: acquires the first data essential feature coefficient, and is also used to acquire the second data essential feature coefficient; The protocol adaptation feature module: uploads the acquired first protocol parameter; The transmission protocol adaptation feature processing module: performs parameter feature processing on the first data acquisition feature, the first data essential feature coefficient, and the first protocol parameter to obtain the first data transmission protocol adaptation feature, and also performs parameter feature processing on the second data acquisition feature, the second data essential feature coefficient, and the first protocol parameter to obtain the second data transmission protocol adaptation feature; The IoT data transmission adaptation model construction module: constructs an IoT data transmission adaptation model according to the first data transmission protocol adaptation feature and the first IoT data transmission protocol adaptation method; The IoT data transmission adaptation terminal module: acquires the uploaded first IoT data transmission protocol adaptation method, is also connected to the transmission protocol adaptation feature processing module and the IoT data transmission adaptation model construction module, acquires the second data transmission protocol adaptation feature and processes to obtain the second IoT data transmission protocol adaptation method, and controls the protocol converter device to perform adaptive selection of the IoT data transmission protocol.

[0040] Furthermore, the IoT data transmission adaptation model uses a neural network model with an improved activation function based on adaptation scenario features and adaptation characteristic parameters; The improved activation function is: ; In the formula, S(X) is the activation function value, is the rank of the matrix of adaptation scenario features, which plays a role in correcting the activation function value, is the adaptation characteristic parameter. Since the adaptation protocol output by the model has a corresponding impact on the adaptation protocol itself, and few existing models consider the situation of the IoT data adaptation protocol manually calibrated and selected in the database, therefore, by correcting the calculation of the activation function in the model based on the situation of the selection library of the adaptation protocol itself, it can be determined that the content sent to the next neuron of the neural network is more in line with the required IoT data transmission protocol, is obtained by weighted addition of the bias of the transmission protocol adaptation feature through the neuron.

[0041] The present invention fully considers that, under the condition that the adaptation protocol library remains unchanged, feature processing of the fusion nature of the data itself, the protocol itself, and even the vector features affecting data transmission is carried out for protocol adaptation affecting Internet of Things data transmission. The activation function of the neural network model is constructed in a scenario-based manner by using the protocol, data, and the corresponding acquisition and transmission process, so as to use the model to adaptively output an immediate transmission protocol adaptation method for the acquisition data, and control the protocol converter device according to the immediate transmission protocol adaptation method, thereby performing adaptation management on Internet of Things data transmission. Through multi-dimensional feature processing of Internet of Things data, and by using an improved activation function big data model to enhance the conversion efficiency of data adaptation transmission protocols, the transmission speed and quality of Internet of Things data are improved, and the manual adaptation loss of data transmission protocols is reduced.

[0042] Of course, it can be understood that each embodiment of the present invention can achieve one of the effects alone, and the combination of multiple embodiments of the present invention can achieve all of the above effects. However, it is not required that each embodiment of the present invention achieve all of the above advantages and effects, because each embodiment of the present invention can constitute an independent technical solution and make one or more contributions to the prior art.

[0043] For the part of the module structure not specifically defined in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the foregoing background art part and the specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters. The protection scope of the present invention shall be subject to the content actually recorded in the claims.

Claims

1. A method for adapting an Internet of Things data transmission protocol based on big data, characterized in that, The method includes: Uploading and obtaining the data acquisition characteristics, data essential feature coefficients, and protocol parameters of the transmitted IoT data that affect the selection of the IoT data adaptation protocol, and obtaining the corresponding IoT data transmission protocol adaptation method; Processing the data acquisition characteristics, the data essential feature coefficients, and the protocol parameters of the transmitted IoT data to obtain data transmission protocol adaptation characteristics; Constructing an IoT data transmission adaptation model according to the transmission protocol adaptation characteristics and the corresponding IoT data transmission protocol adaptation method; Using the IoT data transmission adaptation model to process the to-be-transmitted IoT data to obtain a to-be-transmitted IoT data transmission protocol adaptation method; Based on the to-be-transmitted IoT data transmission protocol adaptation method, performing adaptation management on the IoT data transmission.

2. The method for adapting an IoT data transmission protocol based on big data according to claim 1, wherein: The data acquisition characteristics are obtained by processing the average acquisition transmission time, data cache time, and temperature difference coefficient between the data source and the data storage location.

3. The method for adapting an IoT data transmission protocol based on big data according to claim 1, wherein: The data essential feature coefficients are obtained by processing the characteristics of the data itself, specifically by processing the data dimension, data span size, and data type.

4. The method for adapting an IoT data transmission protocol based on big data according to claim 2, wherein: The protocol parameters include the adaptation scenario characteristics and adaptation characteristic parameters of the protocol.

5. The method for adapting an IoT data transmission protocol based on big data according to claim 4, wherein: The adaptation scenario characteristics obtain adaptation values according to the adaptation scenario. Specifically, according to the broadband and latency conditions in the network environment, the adaptation scenario characteristics are a two-dimensional matrix vector, and the adaptation characteristic parameters are obtained by processing the distance, power consumption, and data volume of the protocol adaptation.

6. The method for adapting an IoT data transmission protocol based on big data according to claim 1, wherein: The processing of the data acquisition feature, the data essential feature coefficient, and the protocol parameter of the transmitted IoT data to obtain the data transmission protocol adaptation feature is processed by means of vector dimensionality elevation. The vector dimensionality elevation method is to , the data essential feature coefficient , and the protocol parameter represented in matrix form.

7. The method for adapting an IoT data transmission protocol based on big data according to claim 2 or 3 or 4 or 6, wherein: The IoT data transmission adaptation model uses a neural network model with an improved activation function based on the adaptation scenario characteristics and adaptation characteristic parameters.

8. The method for adapting an IoT data transmission protocol based on big data according to claim 7, wherein: The improved activation function used by the IoT data transmission adaptation model based on the adaptation scenario characteristics and adaptation characteristic parameters is; ； Wherein, is the activation function value, is the rank of the matrix adapting to the scenario features, is the parameter adapting to the characteristics, is obtained by weighted addition of the transmission protocol adaptation features through neurons plus the bias calculation.

9. An IoT data transmission protocol adaptation system based on big data, including an IoT device data feature acquisition module, an IoT data feature module, a protocol adaptation feature module, an IoT data transmission adaptation terminal module, a transmission protocol adaptation feature processing module, and an IoT data transmission adaptation model construction module, wherein: The IoT device data feature acquisition module: acquires the first data acquisition characteristics in data acquisition and is also used to acquire the second data acquisition characteristics in data acquisition; The IoT data feature module: acquires the first data essential feature coefficient and is also used to acquire the second data essential feature coefficient; The protocol adaptation feature module: uploads the obtained first protocol parameter; The transmission protocol adaptation feature processing module: performs parameter feature processing on the first data acquisition feature, the first data essence feature coefficient, and the first protocol parameter to obtain a first data transmission protocol adaptation feature, and also performs parameter feature processing on the second data acquisition feature, the second data essence feature coefficient, and the first protocol parameter to obtain a second data transmission protocol adaptation feature; The IoT data transmission adaptation model construction module: constructs an IoT data transmission adaptation model according to the first data transmission protocol adaptation feature and the first IoT data transmission protocol adaptation method; The IoT data transmission adaptation terminal module: obtains the uploaded first IoT data transmission protocol adaptation method, is also connected to the transmission protocol adaptation feature processing module and the IoT data transmission adaptation model construction module, obtains the second data transmission protocol adaptation feature and processes to obtain the second IoT data transmission protocol adaptation method, and controls the protocol converter device to perform an adaptive selection of the IoT data transmission protocol.

10. An IoT data transmission protocol adaptation system based on big data according to claim 9, wherein: The IoT data transmission adaptation model uses a neural network model with an improved activation function based on adaptation scenario features and adaptation characteristic parameters; The improved activation function is; ； In the formula, is the activation function value, is the rank of the matrix adapting to the scenario features, is the parameter adapting to the characteristics, is obtained by the weighted sum of the transmission protocol adaptation features through neurons plus the bias.

Images