Internet of Things front-end data acquisition method

Abstract

The invention discloses an Internet of Things front-end data acquisition method, and particularly relates to the technical field of data acquisition, which comprises the following steps: sending data acquired by Internet of Things front-end equipment to a gateway, analyzing the acquired data by the gateway, determining trend information and change information of the Internet of Things front-end data in a monitoring interval, and sending the trend information and the change information to the Internet of Things front-end equipment; through comprehensive analysis of trend information and change information of the Internet of Things front-end data, the transmission frequency of the Internet of Things front-end data uploaded to the cloud platform from the gateway is adjusted, and historical data of different types of Internet of Things front-end data and security problems existing in practical application of the different types of Internet of Things front-end data are analyzed. According to the method and the system, the uncertainty information and the risk information of different types of Internet of Things front-end data are acquired, the priority of the different types of Internet of Things front-end data sent by the gateway is determined, and the method and the system are beneficial to intelligently adjusting a data transmission strategy, so that the security and the stability of the system are improved, and the computing pressure of a cloud platform is effectively relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data acquisition, and more specifically, to a method for acquiring data at the front end of an Internet of Things. Background Art

[0002] With the rapid development of IoT technology, various IoT front-end devices are widely used in many fields such as industry, agriculture, urban management, and health monitoring. These devices collect data such as environmental parameters and device status and upload them to the cloud platform to support intelligent decision-making and real-time monitoring.

[0003] In the Internet of Things, front-end devices may be widely distributed and large in number. If each device sends data directly to the cloud platform, the cloud platform will receive a large number of independent data streams, which may cause network bandwidth constraints and excessive cloud platform load. Through the gateway, data from multiple devices can be aggregated or merged into fewer data packets and then sent to the cloud platform, which significantly reduces the complexity and amount of data transmitted. However, most IoT systems fail to effectively evaluate and control the priority of data upload, resulting in some high-risk and high-importance data not being processed in a timely manner, while a large amount of low-priority data occupies transmission bandwidth and computing resources, reducing the overall performance of the system.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an Internet of Things front-end data collection method to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for collecting data from an Internet of Things front end comprises the following steps:

[0008] S1: Send the data collected by the IoT front-end device to the gateway, and the gateway analyzes the collected data to determine the trend information and change information of the IoT front-end data within the monitoring interval;

[0009] S2: Adjust the transmission frequency of IoT front-end data uploaded from the gateway to the cloud platform through comprehensive analysis of trend information and change information of IoT front-end data;

[0010] S3: Collect uncertainty and risk information of different types of IoT front-end data by analyzing historical data of different types of IoT front-end data and security issues existing in actual applications of different types of IoT front-end data;

[0011] S4: Comprehensively evaluate the uncertainty information and risk information of different types of IoT front-end data to determine the priority of sending different types of IoT front-end data at the gateway.

[0012] In a preferred embodiment, determining trend information of IoT front-end data within a monitoring interval includes:

[0013] The change information of IoT front-end data is expressed by high-frequency filtering amplitude coefficient and data regularization similarity coefficient;

[0014] The acquisition logic of the high-frequency filtering amplitude coefficient is: obtain the IoT front-end data within the monitoring interval, combine and analyze the IoT front-end data within the monitoring interval in the time domain and frequency domain through wavelet transform, and transform the IoT front-end data within the monitoring interval through wavelet transform expression: Among them, WL is the data of the IoT front-end data in the monitoring interval through wavelet transformation, a is the scale of the wavelet function, b is the translation parameter, is the wavelet function;

[0015] By determining the peak frequency and center frequency of the IoT front-end data after wavelet transformation, the frequency difference coefficient is calculated. The calculation formula is: CY pl =|argmax(WL)-WL zx |; Among them, CY pl is the frequency difference coefficient, WL zx It is the center frequency of the IoT front-end data after wavelet transformation;

[0016] Set the wavelet coefficient threshold and calculate the high-frequency filter ratio coefficient. The calculation formula is: Among them, BZ gp is the high frequency filter ratio coefficient, t c ~t d is the time period greater than the wavelet coefficient threshold;

[0017] Calculate the high-frequency filtering amplitude coefficient, the calculation formula is: FZ gp =CY pl ×BZ gp Among them, FZ gp is the high frequency filtering amplitude coefficient.

[0018] In a preferred embodiment, the data regularization similarity coefficient includes:

[0019] The logic for obtaining the data regularization similarity coefficient is as follows: obtain the IoT front-end data in the current monitoring interval, and mark the IoT front-end data in the current monitoring interval as: SJ n , n=1, 2, 3, ..., N, N is a positive integer, n is the number of IoT front-end data received by the gateway at different times within the current monitoring interval;

[0020] Get the IoT front-end data in the previous monitoring interval and mark the IoT front-end data in the previous monitoring interval as: SJ i , i = 1, 2, 3, ..., I, I is a positive integer, i is the number of IoT front-end data received by the gateway at different times in the previous monitoring interval;

[0021] Construct a cost matrix, use the absolute difference between the IoT front-end data at each moment in the current monitoring interval and the previous monitoring interval as the distance metric, determine each element in the cost matrix, and mark the cost matrix as: C(n,i), C(n,i)=|SL n -SL i |;

[0022] Construct a cumulative cost matrix, which records the minimum total cost from the starting point to each point. The cumulative cost matrix is ​​marked as: D(n,i), D(n,i) = C(n,i) + min(D(n-1,i), D(n,i-1), D(n-1,i-1));

[0023] Through recursive cumulative calculation, the DTW distance is determined, the last element of the cumulative cost matrix is ​​obtained, and the data regularization similarity coefficient is calculated. The expression is: GZ xs =D(N,I); where GZ xs is the data normalization similarity coefficient.

[0024] In a preferred embodiment, determining the change information of the IoT front-end data within the monitoring interval includes:

[0025] The change information of IoT front-end data is expressed by time series prediction coefficients;

[0026] The acquisition logic of the time series prediction coefficient is as follows: collect historical IoT front-end data, use IoT front-end data in multiple monitoring intervals in the historical IoT front-end data as training data, divide the training data into input training data and prediction training data, train the autoregressive moving average model according to the input training data and the prediction training data, and estimate the regression coefficient and the moving average coefficient of the autoregressive moving average model by the least squares method;

[0027] Based on the trained autoregressive moving average model, determine the IoT front-end data input into the autoregressive moving average model within the monitoring interval, obtain the predicted value of the IoT front-end data, and mark the predicted value of the IoT front-end data as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the predicted IoT front-end data;

[0028] Determine the actual value of the IoT front-end data in the current monitoring interval, and mark the actual value of the IoT front-end data in the current monitoring interval as: SJ m ;

[0029] Calculate the time series prediction coefficient, the calculation formula is: Among them, XL yc is the time series prediction coefficient.

[0030] In a preferred embodiment, adjusting the transmission frequency of the IoT front-end data uploaded from the gateway to the cloud platform includes:

[0031] Through comprehensive analysis of the change information and trend information of the IoT front-end data, the high-frequency filtering amplitude coefficient, data regularization similarity coefficient and time series prediction coefficient are weighted and calculated to build a transmission evaluation model and generate a transmission evaluation coefficient. The calculation formula is: Among them, pg cs is the transmission evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the high-frequency filtering amplitude coefficient, the data regularization similarity coefficient, and the time series prediction coefficient, respectively, and α1, α2, and α3 are all greater than 0;

[0032] Set a transmission evaluation coefficient threshold and compare the transmission evaluation coefficient threshold with the transmission evaluation coefficient. If the transmission evaluation coefficient threshold is greater than the transmission evaluation coefficient, generate a collection signal, collect IoT front-end data through the pre-set mode of collecting IoT front-end data, and transmit it to the cloud platform. If the transmission evaluation coefficient threshold is less than the transmission evaluation coefficient, do not generate a collection signal and maintain the current rate of IoT front-end data transmission to the cloud platform.

[0033] In a preferred embodiment, uncertainty information of different types of IoT front-end data is collected, including:

[0034] The uncertainty information of IoT front-end data is represented by the entropy value instability coefficient;

[0035] The logic for obtaining the entropy instability coefficient is: according to the historical data of the IoT front-end data, determine the difference between adjacent data points of the IoT front-end data, and mark the difference between adjacent data points of the IoT front-end data as: CZ j , where j = 1, 2, 3, ..., J, J is a positive integer, and j is the number of adjacent data points in the historical data of the IoT front-end data;

[0036] According to the difference between adjacent data points of IoT front-end data, kernel density estimation is used to determine the probability distribution of the difference between adjacent data points. The calculation formula is: Among them, P(CZ) is the probability distribution of the difference between adjacent data points, K is the kernel function, and h is the smoothing parameter;

[0037] Calculate the entropy instability coefficient, the calculation formula is: Among them, SZ wd is the entropy instability coefficient.

[0038] In a preferred embodiment, risk information of different types of IoT front-end data is collected, including:

[0039] The risk information of IoT front-end data is expressed through risk concealment assessment coefficient;

[0040] The logic for obtaining the risk concealment assessment coefficient is as follows: according to the historical records of different types of IoT front-end data, determine the risk events that may be caused after each type of IoT front-end data is uploaded to the cloud platform and the cloud platform determines that the IoT front-end data is abnormal data, obtain the risk event mapping corresponding to each type of IoT front-end data, and determine all risk events caused by different types of IoT front-end data. Based on different risk events, score different risk events according to the expert equal weight principle to determine the risk score of each type of IoT front-end data;

[0041] A logistic regression model is constructed based on the risk scores of risk events caused by IoT front-end data and the proportion of risk events caused by IoT front-end data in all risk events to obtain the risk concealment assessment coefficient. The calculation formula of the risk concealment assessment coefficient is:

[0042]

[0043] Among them, FX yn is the risk concealment assessment coefficient, q = 1, 2, 3, ..., Q, Q is a positive integer, q is the number of the risk event caused by the IoT front-end data, PF1, PF2, PF3, ..., PF Q Score the risk of risk events caused by IoT front-end data, BZ1, BZ2, BZ3, ..., BZ Q It is the proportion of risk events caused by IoT front-end data in all risk events, and e is a natural number.

[0044] In a preferred embodiment, determining the priority of sending different types of IoT front-end data at the gateway includes:

[0045] Through the uncertainty information and risk information of different types of IoT front-end data, the entropy instability coefficient and risk concealment assessment coefficient are weighted and calculated to build a priority assessment model and generate a priority assessment coefficient. The calculation formula is: PG yx =β1SZ wd +β2FX yn Among them, PG yxis the priority assessment coefficient, β1 and β2 are the proportional coefficients of the entropy value instability coefficient and the risk concealment assessment coefficient, respectively, and β1 and β2 are both greater than 0;

[0046] Obtain priority evaluation coefficients of different types of IoT front-end data, sort the priority evaluation coefficients of different types of IoT front-end data from large to small, and upload IoT front-end data with a higher ranking first according to the IoT front-end data received by the gateway.

[0047] Technical effects and advantages of the present invention:

[0048] The present invention analyzes the IoT front-end data at the gateway, extracts trend information and change information of the IoT front-end data within the monitoring interval, and based on the comprehensive analysis of the trend information and change information, the gateway dynamically adjusts the transmission frequency of uploading the IoT front-end data to the cloud platform, and assigns priorities to different types of IoT front-end data by analyzing their historical data and security issues arising in actual applications. The present invention helps to intelligently adjust the data transmission strategy, thereby improving the security and stability of the system and effectively reducing the computing pressure of the cloud platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0050] Figure 1 The present invention is a flowchart of a method for collecting data from an Internet of Things front end. DETAILED DESCRIPTION

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

[0052] Example 1

[0053] Figure 1 The present invention is a flow chart of a method for collecting data from an Internet of Things front end, which specifically includes the following steps:

[0054] S1: Send the data collected by the IoT front-end device to the gateway, and the gateway analyzes the collected data to determine the trend information and change information of the IoT front-end data within the monitoring interval;

[0055] S2: Adjust the transmission frequency of IoT front-end data uploaded from the gateway to the cloud platform through comprehensive analysis of trend information and change information of IoT front-end data;

[0056] S3: Collect uncertainty and risk information of different types of IoT front-end data by analyzing historical data of different types of IoT front-end data and security issues existing in actual applications of different types of IoT front-end data;

[0057] S4: Comprehensively evaluate the uncertainty information and risk information of different types of IoT front-end data to determine the priority of sending different types of IoT front-end data at the gateway.

[0058] The architecture of the Internet of Things system usually includes multiple levels of devices, sensors, gateways and cloud platforms. The interaction and data flow between each level are the core of Internet of Things data transmission. The collection of Internet of Things front-end data is usually carried out through sensors, devices or terminals. These devices are usually installed in the physical environment to monitor specific physical quantities. After the Internet of Things front-end data is collected, it needs to be transmitted to the downstream cloud platform through an appropriate transmission method.

[0059] IoT front-end data collection: IoT front-end data is raw data generated by various sensors, smart devices, machines and other devices. The types of front-end data can be temperature, humidity, location, light, acceleration, etc. Sensors continuously collect data, generate digital signals, and transmit them to the gateway through certain communication protocols;

[0060] Gateway layer: The gateway is mainly responsible for collecting, processing, filtering, compressing and forwarding the data collected by distributed front-end devices. The gateway also acts as a bridge between IoT devices and cloud platforms or other downstream systems;

[0061] Cloud platform layer: The cloud platform is the core of IoT data storage and processing. IoT front-end devices upload data to the cloud platform through the gateway. The cloud platform receives, stores and further processes and analyzes the data. The cloud platform usually has powerful data processing, storage and analysis capabilities, and can handle massive data generated by thousands of devices.

[0062] Among them, the gateway is responsible for receiving data from the IoT front-end device and performing preliminary analysis and processing, controlling the transmission frequency of different types of IoT front-end data, analyzing the IoT front-end data within the monitoring interval, collecting change information and trend information of the IoT front-end data, and representing the change information of the IoT front-end data through high-frequency filtering amplitude coefficients and data regularization similarity coefficients, and representing the trend information of the IoT front-end data through time series prediction coefficients.

[0063] It should be noted that the monitoring interval is a specific time period set by professional staff. Usually, the frequency of sending IoT front-end data is high, so there are IoT front-end data at multiple time points within the monitoring interval.

[0064] The acquisition logic of the high-frequency filtering amplitude coefficient is: obtain the IoT front-end data within the monitoring interval, combine and analyze the IoT front-end data within the monitoring interval in the time domain and frequency domain through wavelet transform, and transform the IoT front-end data within the monitoring interval through wavelet transform expression: Among them, WL is the data of the IoT front-end data in the monitoring interval through wavelet transformation, a is the scale of the wavelet function, b is the translation parameter, is the wavelet function;

[0065] It should be noted that the translation parameter b of the wavelet function is determined according to the length of the monitoring interval, and the wavelet function and the scale a of the wavelet function are determined by professional staff based on the characteristics of different types of IoT front-end data. If the IoT front-end data is usually stable in historical data, a larger scale is selected; if the IoT front-end data is usually fluctuating in historical data, a smaller scale is selected.

[0066] By determining the peak frequency and center frequency of the IoT front-end data after wavelet transformation, the frequency difference coefficient is calculated. The calculation formula is: CY pl =|argmax(WL)-WL zx |; Among them, CY pl is the frequency difference coefficient, WL zx It is the center frequency of the IoT front-end data after wavelet transformation;

[0067] Set the wavelet coefficient threshold and calculate the high-frequency filter ratio coefficient. The calculation formula is: Among them, BZ gp is the high frequency filter ratio coefficient, t c ~t d is the time period greater than the wavelet coefficient threshold;

[0068] It should be noted that the wavelet coefficient threshold is set by professional field staff and adjusted based on different types of IoT front-end data.

[0069] Calculate the high-frequency filtering amplitude coefficient, the calculation formula is: FZ gp =CY pl ×BZ gp Among them, FZ gp is the high frequency filtering amplitude coefficient.

[0070] It can be seen from the formula that the larger the high-frequency filtering amplitude coefficient is, the greater the change in the IoT front-end data may be. Therefore, the transmission frequency of the IoT front-end data may be increased, which will help the cloud platform to analyze the transmitted IoT front-end data in a timely manner and make corresponding decisions.

[0071] The logic for obtaining the data regularization similarity coefficient is as follows: obtain the IoT front-end data in the current monitoring interval, and mark the IoT front-end data in the current monitoring interval as: SJ n , n=1, 2, 3, ..., N, N is a positive integer, n is the number of IoT front-end data received by the gateway at different times within the current monitoring interval;

[0072] Get the IoT front-end data in the previous monitoring interval and mark the IoT front-end data in the previous monitoring interval as: SJ i , i = 1, 2, 3, ..., I, I is a positive integer, i is the number of IoT front-end data received by the gateway at different times in the previous monitoring interval;

[0073] Construct a cost matrix, use the absolute difference between the IoT front-end data at each moment in the current monitoring interval and the previous monitoring interval as the distance metric, determine each element in the cost matrix, and mark the cost matrix as: C(n,i), C(n,i)=|SL n -SL i |;

[0074] Construct a cumulative cost matrix, which records the minimum total cost from the starting point to each point. The cumulative cost matrix is ​​marked as: D(n,i), D(n,i) = C(n,i) + min(D(n-1,i), D(n,i-1), D(n-1,i-1));

[0075] Through recursive cumulative calculation, the DTW distance is determined, the last element of the cumulative cost matrix is ​​obtained, and the data regularization similarity coefficient is calculated. The expression is: GZ xs =D(N,I); where GZ xs is the data normalization similarity coefficient.

[0076] It can be seen from the formula that the larger the data regularization similarity coefficient is, the greater the difference between the IoT front-end data in the current monitoring interval and the IoT front-end data in the previous monitoring interval. Therefore, the transmission frequency of the IoT front-end data may be increased, and the IoT front-end data in the current monitoring interval may be uploaded to the cloud platform at a higher frequency.

[0077] The acquisition logic of the time series prediction coefficient is as follows: collect historical IoT front-end data, use IoT front-end data in multiple monitoring intervals in the historical IoT front-end data as training data, divide the training data into input training data and prediction training data, train the autoregressive moving average model according to the input training data and the prediction training data, and estimate the regression coefficient and the moving average coefficient of the autoregressive moving average model by the least squares method;

[0078] Based on the trained autoregressive moving average model, determine the IoT front-end data input into the autoregressive moving average model within the monitoring interval, obtain the predicted value of the IoT front-end data, and mark the predicted value of the IoT front-end data as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the predicted IoT front-end data;

[0079] Determine the actual value of the IoT front-end data in the current monitoring interval, and mark the actual value of the IoT front-end data in the current monitoring interval as: SJ m ;

[0080] Calculate the time series prediction coefficient, the calculation formula is: Among them, XL yc is the time series prediction coefficient.

[0081] It can be seen from the formula that the larger the time series prediction coefficient is, the greater the difference between the IoT front-end data and the historical data in the monitoring interval is, and the more unstable it is. Therefore, it is possible to increase the transmission frequency of the IoT front-end data and upload the IoT front-end data in the current monitoring interval to the cloud platform at a higher frequency.

[0082] Through comprehensive analysis of the change information and trend information of the IoT front-end data, the high-frequency filtering amplitude coefficient, data regularization similarity coefficient and time series prediction coefficient are weighted and calculated to build a transmission evaluation model and generate a transmission evaluation coefficient. The calculation formula is: Among them, pg cs is the transmission evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the high-frequency filtering amplitude coefficient, the data regularization similarity coefficient, and the time series prediction coefficient, respectively. α1, α2, and α3 are all greater than 0.

[0083] It can be seen from the formula that the larger the high-frequency filtering amplitude coefficient, the data regularization similarity coefficient and the time series prediction coefficient, the larger the transmission evaluation coefficient, which means that the IoT front-end data received at the gateway may have irregular and obvious changes, and may need to be transmitted to the cloud platform more frequently to update the data in the cloud platform in time. Conversely, the smaller the high-frequency filtering amplitude coefficient, the data regularization similarity coefficient and the time series prediction coefficient, the smaller the transmission evaluation coefficient, which means that the IoT front-end data received at the gateway may need to be compressed, selectively transmitted or the transmission frequency reduced, which helps to save bandwidth and storage resources, and reduce the computing pressure of the cloud platform, ensuring that only key data is transmitted and processed, and achieving the goal of efficient resource utilization.

[0084] Set a transmission evaluation coefficient threshold and compare the transmission evaluation coefficient threshold with the transmission evaluation coefficient. If the transmission evaluation coefficient threshold is greater than the transmission evaluation coefficient, generate a collection signal, collect IoT front-end data through the pre-set mode of collecting IoT front-end data, and transmit it to the cloud platform. If the transmission evaluation coefficient threshold is less than the transmission evaluation coefficient, do not generate a collection signal and maintain the current rate of IoT front-end data transmission to the cloud platform.

[0085] Example 2

[0086] The above-mentioned embodiment determines the transmission behavior of different types of IoT front-end data through comprehensive analysis of the change information and trend information of IoT front-end data, and reduces the computing pressure of the cloud platform by reducing the transmission of unnecessary data. In this embodiment, by considering the priority of different types of IoT front-end data, the gateway can decide the data that needs to be transmitted first and the data that needs to be uploaded with delay.

[0087] By analyzing the historical data of different types of IoT front-end data and the security issues existing in the actual applications of different types of IoT front-end data, the uncertainty information and risk information of different types of IoT front-end data are collected, and the uncertainty information of different types of IoT front-end data is represented by the entropy instability coefficient, and the risk information of different types of IoT front-end data is represented by the risk concealment assessment coefficient.

[0088] The advantages of the entropy instability coefficient are:

[0089] Because IoT front-end data may be affected by a variety of environmental factors and device changes, the entropy instability coefficient adapts to more complex and irregular data distribution;

[0090] Entropy can measure the randomness or uncertainty of data. In IoT data, high entropy means that data changes are complex, which may be caused by emergencies or abnormal behaviors, while low entropy means that the data is relatively stable and the system is running normally. The change of entropy value helps to optimize the data upload strategy as a whole and reduce unnecessary transmission and delay;

[0091] For IoT front-end data, the fluctuations and trend changes in time series data often have strong dependencies. The entropy instability coefficient can help capture and quantify the mutations or fluctuations in this time dependency, thus playing an important role in time series-based analysis.

[0092] The logic for obtaining the entropy instability coefficient is: according to the historical data of the IoT front-end data, determine the difference between adjacent data points of the IoT front-end data, and mark the difference between adjacent data points of the IoT front-end data as: CZ j , where j = 1, 2, 3, ..., J, J is a positive integer, and j is the number of adjacent data points in the historical data of the IoT front-end data;

[0093] According to the difference between adjacent data points of IoT front-end data, kernel density estimation is used to determine the probability distribution of the difference between adjacent data points. The calculation formula is: Among them, P(CZ) is the probability distribution of the difference between adjacent data points, K is the kernel function, and h is the smoothing parameter;

[0094] Calculate the entropy instability coefficient, the calculation formula is: Among them, SZ wd is the entropy instability coefficient.

[0095] It can be seen from the formula that the larger the entropy instability coefficient, the more dispersed the distribution of this type of IoT front-end data is, and the higher the information uncertainty is. It may be necessary to give priority to uploading this type of IoT front-end data.

[0096] The advantages of the risk concealment assessment coefficient are:

[0097] The risk concealment assessment coefficient converts the risk of data into a clear value by combining data change information, the severity and probability of risk events, and provides an objective reference indicator for decision makers to quantify and compare the risks of different types of data;

[0098] The risk hidden assessment coefficient can quantify and evaluate the severity of potential risks based on the dangerousness of risk events caused by different types of IoT front-end data. Through the analysis of historical data and combined with expert opinions, the assessment of different risk events can be adjusted and optimized.

[0099] The logic for obtaining the risk concealment assessment coefficient is as follows: according to the historical records of different types of IoT front-end data, determine the risk events that may be caused after each type of IoT front-end data is uploaded to the cloud platform and the cloud platform determines that the IoT front-end data is abnormal data, obtain the risk event mapping corresponding to each type of IoT front-end data, and determine all risk events caused by different types of IoT front-end data. Based on different risk events, score different risk events according to the expert equal weight principle to determine the risk score of each type of IoT front-end data;

[0100] A logistic regression model is constructed based on the risk scores of risk events caused by IoT front-end data and the proportion of risk events caused by IoT front-end data in all risk events to obtain the risk concealment assessment coefficient. The calculation formula of the risk concealment assessment coefficient is:

[0101]

[0102] Among them, FX yn is the risk concealment assessment coefficient, q = 1, 2, 3, ..., Q, Q is a positive integer, q is the number of the risk event caused by the IoT front-end data, PF1, PF2, PF3, ..., PF Q Score the risk of risk events caused by IoT front-end data, BZ1, BZ2, BZ3, ..., BZ Q It is the proportion of risk events caused by IoT front-end data in all risk events, and e is a natural number.

[0103] It can be seen from the formula that the larger the risk concealment assessment coefficient, the more dangerous the risk events caused by the IoT front-end data are. If they are not handled in a timely manner, the risks may be greater and there may be a greater probability of causing risks. Therefore, it may be necessary to give priority to uploading this type of IoT front-end data.

[0104] Through the uncertainty information and risk information of different types of IoT front-end data, the entropy instability coefficient and risk concealment assessment coefficient are weighted and calculated to build a priority assessment model and generate a priority assessment coefficient. The calculation formula is: PG yx =β1SZ wd +β2FX yn Among them, PG yx is the priority assessment coefficient, β1 and β2 are the proportional coefficients of the entropy instability coefficient and the risk concealment assessment coefficient, respectively, and β1 and β2 are both greater than 0.

[0105] It can be seen from the formula that the larger the entropy instability coefficient and the risk hidden assessment coefficient, the larger the priority assessment coefficient, which means that the priority of uploading this type of IoT front-end data to the cloud platform for analysis is higher, ensuring that the system can discover problems as early as possible and take corresponding measures. Conversely, the smaller the entropy instability coefficient and the risk hidden assessment coefficient, the smaller the priority assessment coefficient, which means that the priority of uploading this type of IoT front-end data to the cloud platform for analysis is lower.

[0106] Obtain priority evaluation coefficients of different types of IoT front-end data, sort the priority evaluation coefficients of different types of IoT front-end data from large to small, and upload IoT front-end data with a higher ranking first according to the IoT front-end data received by the gateway.

[0107] The present invention analyzes the IoT front-end data at the gateway, extracts trend information and change information of the IoT front-end data within the monitoring interval, and based on the comprehensive analysis of the trend information and change information, the gateway dynamically adjusts the transmission frequency of uploading the IoT front-end data to the cloud platform, and assigns priorities to different types of IoT front-end data by analyzing their historical data and security issues arising in actual applications. The present invention helps to intelligently adjust the data transmission strategy, thereby improving the security and stability of the system and effectively reducing the computing pressure of the cloud platform.

[0108] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0109] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0110] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0111] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0114] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0115] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for collecting data from the front end of the Internet of Things, characterized in that: The specific steps include: S1: Send the data collected by the IoT front-end device to the gateway, and the gateway analyzes the collected data to determine the trend information and change information of the IoT front-end data within the monitoring interval; S2: Adjust the transmission frequency of IoT front-end data uploaded from the gateway to the cloud platform through comprehensive analysis of trend information and change information of IoT front-end data; S3: Collect uncertainty and risk information of different types of IoT front-end data by analyzing historical data of different types of IoT front-end data and security issues existing in actual applications of different types of IoT front-end data; S4: Comprehensively evaluate the uncertainty information and risk information of different types of IoT front-end data to determine the priority of sending different types of IoT front-end data at the gateway.

2. The method for collecting data from an Internet of Things front end according to claim 1, characterized in that: Determine the trend information of IoT front-end data within the monitoring interval, including; The change information of IoT front-end data is expressed by high-frequency filtering amplitude coefficient and data regularization similarity coefficient; The acquisition logic of the high-frequency filtering amplitude coefficient is: obtain the IoT front-end data within the monitoring interval, combine and analyze the IoT front-end data within the monitoring interval in the time domain and frequency domain through wavelet transform, and transform the IoT front-end data within the monitoring interval through wavelet transform expression: Among them, WL is the data of the IoT front-end data in the monitoring interval through wavelet transformation, a is the scale of the wavelet function, b is the translation parameter, is the wavelet function; By determining the peak frequency and center frequency of the IoT front-end data after wavelet transformation, the frequency difference coefficient is calculated. The calculation formula is: CY pl =|argmax(WL)-WL zx |; Among them, CY pl is the frequency difference coefficient, WL zx It is the center frequency of the IoT front-end data after wavelet transformation; Set the wavelet coefficient threshold and calculate the high-frequency filter ratio coefficient. The calculation formula is: Among them, BZ gp is the high frequency filter ratio coefficient, t c ~t d is the time period greater than the wavelet coefficient threshold; Calculate the high-frequency filtering amplitude coefficient, the calculation formula is: FZ gp =CY pl ×BZ gp Among them, FZ gp is the high frequency filtering amplitude coefficient.

3. The method for collecting data from an Internet of Things front end according to claim 2, characterized in that: Data normalization similarity coefficient, including: The logic for obtaining the data regularization similarity coefficient is: obtain the IoT front-end data in the current monitoring interval, and mark the IoT front-end data in the current monitoring interval as: SJ n , n=1, 2, 3, ..., N, N is a positive integer, n is the number of IoT front-end data received by the gateway at different times within the current monitoring interval; Get the IoT front-end data in the previous monitoring interval and mark the IoT front-end data in the previous monitoring interval as: SJ i , i = 1, 2, 3, ..., I, I is a positive integer, i is the number of IoT front-end data received by the gateway at different times in the previous monitoring interval; Construct a cost matrix, use the absolute difference between the IoT front-end data at each moment in the current monitoring interval and the previous monitoring interval as the distance metric, determine each element in the cost matrix, and mark the cost matrix as: C(n,i), C(n,i)=|SL n -SL i |; Construct a cumulative cost matrix, which records the minimum total cost from the starting point to each point. The cumulative cost matrix is ​​marked as: D(n,i), D(n,i) = C(n,i) + min(D(n-1,i), D(n,i-1), D(n-1,i-1)); Through recursive cumulative calculation, the DTW distance is determined, the last element of the cumulative cost matrix is ​​obtained, and the data regularization similarity coefficient is calculated. The expression is: GZ xs =D(N,I); where GZ xs is the data normalization similarity coefficient.

4. The method for collecting data from an Internet of Things front end according to claim 3, characterized in that: Determine the change information of IoT front-end data within the monitoring interval, including: The change information of IoT front-end data is expressed by time series prediction coefficients; The acquisition logic of the time series prediction coefficient is as follows: collect historical IoT front-end data, use IoT front-end data in multiple monitoring intervals in the historical IoT front-end data as training data, divide the training data into input training data and prediction training data, train the autoregressive moving average model according to the input training data and the prediction training data, and estimate the regression coefficient and the moving average coefficient of the autoregressive moving average model by the least squares method; Based on the trained autoregressive moving average model, determine the IoT front-end data input into the autoregressive moving average model within the monitoring interval, obtain the predicted value of the IoT front-end data, and mark the predicted value of the IoT front-end data as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the predicted IoT front-end data; Determine the actual value of the IoT front-end data in the current monitoring interval, and mark the actual value of the IoT front-end data in the current monitoring interval as: SJ m ; Calculate the time series prediction coefficient, the calculation formula is: Among them, XL yc is the time series prediction coefficient.

5. The method for collecting data from the Internet of Things front end according to claim 4, characterized in that: Adjust the transmission frequency of IoT front-end data uploaded from the gateway to the cloud platform, including: Through comprehensive analysis of the change information and trend information of the IoT front-end data, the high-frequency filtering amplitude coefficient, data regularization similarity coefficient and time series prediction coefficient are weighted and calculated to build a transmission evaluation model and generate a transmission evaluation coefficient. The calculation formula is: Among them, pg cs is the transmission evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the high-frequency filtering amplitude coefficient, the data regularization similarity coefficient, and the time series prediction coefficient, respectively, and α1, α2, and α3 are all greater than 0; Set a transmission evaluation coefficient threshold and compare the transmission evaluation coefficient threshold with the transmission evaluation coefficient. If the transmission evaluation coefficient threshold is greater than the transmission evaluation coefficient, generate a collection signal, collect IoT front-end data through the pre-set mode of collecting IoT front-end data, and transmit it to the cloud platform. If the transmission evaluation coefficient threshold is less than the transmission evaluation coefficient, do not generate a collection signal and maintain the current rate of IoT front-end data transmission to the cloud platform.

6. The method for collecting data from the Internet of Things front end according to claim 5, characterized in that: Collect uncertainty information of different types of IoT front-end data, including: The uncertainty information of IoT front-end data is represented by the entropy value instability coefficient; The logic for obtaining the entropy instability coefficient is: according to the historical data of the IoT front-end data, determine the difference between adjacent data points of the IoT front-end data, and mark the difference between adjacent data points of the IoT front-end data as: CZ j , where j = 1, 2, 3, ..., J, J is a positive integer, and j is the number of adjacent data points in the historical data of the IoT front-end data; According to the difference between adjacent data points of IoT front-end data, kernel density estimation is used to determine the probability distribution of the difference between adjacent data points. The calculation formula is: Among them, P(CZ) is the probability distribution of the difference between adjacent data points, K is the kernel function, and h is the smoothing parameter; Calculate the entropy instability coefficient, the calculation formula is: Among them, SZ wd is the entropy instability coefficient.

7. The method for collecting data from the Internet of Things front end according to claim 6, characterized in that: Collect risk information of different types of IoT front-end data, including: The risk information of IoT front-end data is expressed through risk concealment assessment coefficient; The logic for obtaining the risk concealment assessment coefficient is as follows: according to the historical records of different types of IoT front-end data, determine the risk events that may be caused after each type of IoT front-end data is uploaded to the cloud platform and the cloud platform determines that the IoT front-end data is abnormal data, obtain the risk event mapping corresponding to each type of IoT front-end data, and determine all risk events caused by different types of IoT front-end data. Based on different risk events, score different risk events according to the expert equal weight principle to determine the risk score of each type of IoT front-end data; A logistic regression model is constructed based on the risk scores of risk events caused by IoT front-end data and the proportion of risk events caused by IoT front-end data in all risk events to obtain the risk concealment assessment coefficient. The calculation formula of the risk concealment assessment coefficient is: Among them, FX yn is the risk concealment assessment coefficient, q = 1, 2, 3, ..., Q, Q is a positive integer, q is the number of the risk event caused by the IoT front-end data, PF1, PF2, PF3, ..., PF Q Score the risk of risk events caused by IoT front-end data, BZ1, BZ2, BZ3, ..., BZ Q It is the proportion of risk events caused by IoT front-end data in all risk events, and e is a natural number.

8. The method for collecting data from the Internet of Things front end according to claim 7, characterized in that: Determine the priority of sending different types of IoT front-end data at the gateway, including: Through the uncertainty information and risk information of different types of IoT front-end data, the entropy instability coefficient and risk concealment assessment coefficient are weighted and calculated to build a priority assessment model and generate a priority assessment coefficient. The calculation formula is: PG yx =β1SZ wd +β2FX yn Among them, PG yx is the priority assessment coefficient, β1 and β2 are the proportional coefficients of the entropy value instability coefficient and the risk concealment assessment coefficient, respectively, and β1 and β2 are both greater than 0; Obtain priority evaluation coefficients of different types of IoT front-end data, sort the priority evaluation coefficients of different types of IoT front-end data from large to small, and upload IoT front-end data with a higher ranking first according to the IoT front-end data received by the gateway.

Images