An industrial internet of things data transmission optimization method under a complex environment

Abstract

The application relates to the technical field of data transmission, and discloses an industrial Internet of Things data transmission optimization method in a complex environment, which comprises collecting data transmission parameters and environment parameters of industrial Internet of Things equipment in a complex environment, generating original data transmission data sets and environment parameter data sets; preprocessing the original data transmission data sets and the environment parameter data sets, including data cleaning, normalization and feature extraction, generating preprocessed data transmission feature data and environment features; collecting data transmission parameters and environment parameters of the industrial Internet of Things equipment, and performing data cleaning, normalization and feature extraction on the data transmission parameters and the environment parameters, so as to generate preprocessed feature data, guarantee the accuracy and reliability of data transmission analysis, simultaneously perform timestamp alignment and space correlation processing on the environment parameters and the transmission parameters, perceive the influence of the complex environment, and guarantee the accuracy and integrity of the data transmission optimization basic data.

Description

Technical Field

[0001] This invention relates to the field of data transmission technology, specifically to an optimization method for industrial Internet of Things (IoT) data transmission in complex environments. Background Technology

[0002] The Industrial Internet of Things (IIoT) integrates various data acquisition and control sensors or controllers with sensing and monitoring capabilities, as well as technologies such as mobile communication and intelligent analysis, into all aspects of the industrial production process. This significantly improves manufacturing efficiency, enhances product quality, reduces product costs and resource consumption, and ultimately elevates traditional industries to a new stage of intelligent manufacturing.

[0003] Currently, in the process of industrial IoT data transmission in complex environments, due to the presence of various dynamic interference factors, the traditional transmission strategies used for real-time data transmission optimization are based on fixed rules or static models. These strategies cannot perceive and adapt in real time to the impact of changes in the transmission environment, such as lighting, electromagnetic noise, and moving obstacles, on signal quality. This may lead to increased data transmission delay and packet loss rate, making it difficult to guarantee the real-time performance and reliability of transmission, thereby affecting the overall operational efficiency of the industrial IoT.

[0004] Therefore, a method for optimizing data transmission in industrial IoT under complex environments is proposed to address the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an optimization method for industrial IoT data transmission in complex environments. This method solves the problems mentioned in the background, such as increased data transmission latency, higher packet loss rate, difficulty in ensuring real-time performance and reliability, and thus affecting the overall operational efficiency of the industrial IoT.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing data transmission in industrial IoT under complex environments, the method comprising the following steps:

[0007] S1. Collect data transmission parameters and environmental parameters of industrial IoT devices in complex environments, and generate raw data transmission datasets and environmental parameter datasets;

[0008] S2. Preprocess the original data transmission dataset and environmental parameter dataset, including data cleaning, normalization and feature extraction, to generate preprocessed data transmission feature data and environmental feature data;

[0009] S3. Based on the preprocessed data transmission characteristic data and environmental characteristic data, perform a quantitative analysis of the impact of complex environment on data transmission, and generate environmental interference assessment data, wherein environmental interference includes signal attenuation, multipath interference and electromagnetic noise.

[0010] S4. Based on the environmental interference assessment data, an adaptive optimization algorithm is used to select the optimal data transmission strategy, which includes route selection, modulation method adjustment, and transmission power control.

[0011] S5. Based on the optimal data transmission strategy, perform real-time transmission of industrial IoT data, and dynamically monitor transmission performance indicators, including throughput, latency and packet loss rate, during the transmission process.

[0012] S6. Based on dynamic monitoring results, use a feedback control mechanism to adjust the parameters of the data transmission strategy in real time to cope with environmental changes;

[0013] S7. Perform integrity verification and error recovery processing on the transmitted data, and generate an optimized data transmission report;

[0014] S8 stores and transmits data and analysis results, which are suitable for subsequent machine learning model training and strategy optimization.

[0015] Preferably, the step S1, which involves collecting data transmission parameters and environmental parameters of industrial IoT devices in complex environments, includes the following steps:

[0016] S11. Collect real-time data transmission parameters of connected devices through industrial IoT gateway devices, including data packet size, transmission frequency, protocol type and device identifier, and generate raw data transmission dataset;

[0017] S12. Collect complex environmental parameters using an environmental sensor network, including temperature, humidity, signal strength, noise level, and distribution of moving obstacles, and generate an environmental parameter dataset.

[0018] S13. Perform timestamp alignment and spatial association processing on the original data transmission dataset and the environmental parameter dataset.

[0019] Preferably, the preprocessing in S2 includes the following steps:

[0020] S21. Perform data cleaning on the original data transmission dataset to remove outliers and duplicate data, and fill in missing values;

[0021] S22. The cleaned data is normalized using the min-max normalization method to scale all parameters to the [0,1] interval.

[0022] S23. Extract key features from normalized data, including data flow patterns, environmental periodic change features, and interference correlation features, to generate preprocessed data transmission feature data and environmental feature data.

[0023] Preferably, the quantitative analysis of the impact of complex environment on data transmission in S3 includes the following steps:

[0024] S31. Based on the preprocessed environmental feature data, calculate the path loss index and interference coefficient in the signal propagation model to generate preliminary environmental interference indicators.

[0025] S32. Combining data transmission characteristic data, use grey relational analysis to quantify the degree of influence of environmental parameters on data transmission quality and generate a correlation matrix.

[0026] S33. Based on the correlation matrix, generate environmental interference assessment data through weighted fusion, wherein the assessment data includes interference level, expected bit error rate and available bandwidth.

[0027] Preferably, selecting the optimal data transmission strategy in step S4 includes the following steps:

[0028] S41. Establish a multi-objective optimization function with the goal of maximizing throughput and minimizing latency and packet loss rate, and the constraints include equipment energy limits and environmental interference thresholds.

[0029] S42. Use reinforcement learning algorithm as adaptive optimization algorithm to initialize Q-learning model, where the state space is the environmental disturbance evaluation data and the action space is the combination of data transmission strategies.

[0030] S43. By iteratively training the Q-learning model, select the data transmission strategy that maximizes the long-term reward as the optimal strategy, and output the strategy parameters.

[0031] Preferably, the real-time transmission of industrial IoT data in S5 includes the following steps:

[0032] S51. Configure the communication module of the industrial IoT device according to the selected optimal data transmission strategy, including setting the routing path, modulation and coding scheme and transmission power;

[0033] S52. Start data stream transmission and use the embedded monitoring agent to collect transmission performance indicators in real time, sampling data once every predetermined time window;

[0034] S53. Aggregate the sampled data to generate a transmission performance log, which is suitable for subsequent dynamic adjustments.

[0035] Preferably, the parameters for real-time adjustment of the data transmission strategy in S6 include the following steps:

[0036] S61. Design a proportional-integral-derivative controller as a feedback control mechanism. The input is the deviation between the real-time transmission performance index and the target value, and the output is the adjustment amount of the strategy parameters.

[0037] S62. When the deviation exceeds the preset threshold, the controller is triggered to calculate new routing weights and modulation parameters;

[0038] S63. Dynamically update the adjusted parameters to the running data transmission strategy to achieve closed-loop optimization.

[0039] Preferably, the integrity verification and error recovery process in S7 includes the following steps:

[0040] S71. At the data receiving end, use cyclic redundancy check and hash algorithm to verify data integrity and request retransmission of erroneous data packets;

[0041] S72. Forward error correction coding technology is used to decode and recover retransmitted data packets, reducing the number of repeated transmissions;

[0042] S73, calculates transmission success rate, average latency, and energy efficiency ratio, and generates an optimized data transmission report.

[0043] Preferably, the storage of transmission process data and analysis results in step S8 includes the following steps:

[0044] S81. Store transmission process data, environmental interference assessment data, and optimization reports in a distributed database on the cloud platform;

[0045] S82. Regularly use stored data to train deep neural network models to improve the prediction accuracy of adaptive optimization algorithms;

[0046] S83. Based on the model output, update the policy library for future transmission tasks to call.

[0047] Preferably, the method further includes a security enhancement step:

[0048] S91. During data transmission, the lightweight encryption algorithm AES-128 is integrated to perform end-to-end encryption on sensitive data;

[0049] S92. Use blockchain technology to record transmission logs to make the data immutable and traceable;

[0050] S93. Implement identity authentication and access control mechanisms to prevent unauthorized devices from accessing the network.

[0051] Compared with existing technologies, this invention provides a method for optimizing data transmission in industrial IoT under complex environments, which has the following beneficial effects:

[0052] 1. In this invention, data transmission parameters and environmental parameters of industrial IoT devices are collected, and the data is cleaned, normalized, and feature extracted to generate preprocessed feature data, ensuring the accuracy and reliability of data transmission analysis. At the same time, environmental parameters and transmission parameters are time-stamp aligned and spatially correlated to perceive the impact of complex environments and ensure the accuracy and completeness of the basic data for data transmission optimization.

[0053] 2. In this invention, by quantitatively analyzing the impact of complex environments on data transmission, environmental interference assessment data is generated, and an adaptive optimization algorithm is used to select the optimal data transmission strategy, dynamically adapting to environmental changes. At the same time, routing selection, modulation method adjustment, and transmission power control are optimized to ensure high efficiency and low latency of data transmission and reduce packet loss rate.

[0054] 3. In this invention, transmission performance indicators are monitored in real time and policy parameters are dynamically adjusted using a feedback control mechanism to cope with environmental changes. At the same time, integrity verification and error recovery are performed to correct transmission deviations in real time, ensuring the integrity and reliability of data transmission. Furthermore, by storing transmission process data and analysis results, subsequent machine learning model training and policy optimization are supported, thereby improving the system's continuous optimization capabilities. Attached Figure Description

[0055] Figure 1 This is a flowchart of an industrial IoT data transmission optimization method under complex environments according to the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] For specific implementation examples, please refer to: Figure 1 A method for optimizing data transmission in industrial IoT under complex environments, comprising the following steps:

[0058] S1. Collect data transmission parameters and environmental parameters of industrial IoT devices in complex environments, and generate raw data transmission datasets and environmental parameter datasets;

[0059] S2. Preprocess the original data transmission dataset and environmental parameter dataset, including data cleaning, normalization and feature extraction, to generate preprocessed data transmission feature data and environmental feature data;

[0060] S3. Based on the preprocessed data transmission characteristic data and environmental characteristic data, perform a quantitative analysis of the impact of complex environment on data transmission, and generate environmental interference assessment data, which includes signal attenuation, multipath interference and electromagnetic noise.

[0061] S4. Based on the environmental interference assessment data, an adaptive optimization algorithm is used to select the optimal data transmission strategy. The data transmission strategy includes route selection, modulation method adjustment and transmission power control.

[0062] S5. Based on the optimal data transmission strategy, perform real-time transmission of industrial IoT data and dynamically monitor transmission performance indicators, including throughput, latency and packet loss rate, during the transmission process.

[0063] S6. Based on dynamic monitoring results, use a feedback control mechanism to adjust the parameters of the data transmission strategy in real time to cope with environmental changes;

[0064] S7. Perform integrity verification and error recovery processing on the transmitted data, and generate an optimized data transmission report;

[0065] S8 stores and transmits data and analysis results, which are suitable for subsequent machine learning model training and strategy optimization.

[0066] S1 collects data transmission parameters and environmental parameters of industrial IoT devices in complex environments, including the following steps:

[0067] S11. Collect real-time data transmission parameters of connected devices through industrial IoT gateway devices, including data packet size, transmission frequency, protocol type and device identifier, and generate raw data transmission dataset;

[0068] S12. Collect complex environmental parameters using an environmental sensor network, including temperature, humidity, signal strength, noise level, and distribution of moving obstacles, and generate an environmental parameter dataset.

[0069] S13. Perform timestamp alignment and spatial correlation processing on the original data transmission dataset and the environmental parameter dataset;

[0070] Specific implementation steps for timestamp alignment and spatial association processing:

[0071] First, the data acquired from industrial IoT gateways and device sensors will contain timestamps and spatial coordinates. Then, a time series alignment algorithm is used to interpolate and aggregate data from different sources according to time windows, so that the timestamps of all data points are aligned to a unified time axis. For spatial correlation, spatial indexing technology is used to match the device location with the location of environmental sensors to ensure that data in the same area are correlated.

[0072] Linear interpolation formula:

[0073] ;

[0074] in For parameters Target timestamp after alignment The estimated value at that location, , For parameters In the original data and The adjacent previous one and the next one Known observations at the timestamp , For time intervals.

[0075] Preprocessing in S2 includes the following steps:

[0076] S21. Clean the original data transmission dataset, remove outliers and duplicate data, and fill in missing values;

[0077] First, for outlier detection, the statistical method Z-score algorithm is used: the mean and standard deviation of each data parameter are calculated, and values ​​exceeding ±3 standard deviations are considered outliers and deleted. For duplicate data, the MD5 hash algorithm is used to generate a unique identifier for each data record, and completely duplicate entries are removed by comparing the identifiers. For missing value imputation, the method is selected according to the data characteristics: when the data is a time series, linear interpolation is used; when the data is a categorical parameter, mode imputation is used.

[0078] The formula for calculating the Z-score standardized value is:

[0079] ;

[0080] in The Z-score is the standardized value. For the first in the dataset Data points, The mean of the dataset. The standard deviation is denoted as .

[0081] S22. The cleaned data is normalized using the min-max normalization method to scale all parameters to the [0,1] interval.

[0082] The minimum-maximum normalization formula is:

[0083] ;

[0084] in Represents the original data value. and These represent the minimum and maximum values ​​of the parameter in the entire dataset, respectively. This represents the normalized value;

[0085] S23. Extract key features from normalized data, including data flow patterns, environmental periodic change features, and interference correlation features, to generate preprocessed data transmission feature data and environmental feature data.

[0086] The quantitative analysis of the impact of complex environments on data transmission in S3 includes the following steps:

[0087] S31. Based on the preprocessed environmental characteristic data, calculate the path loss index and interference coefficient in the signal propagation model to generate preliminary environmental interference indicators.

[0088] First, based on the preprocessed environmental feature data, namely signal strength and distance, the logarithmic distance path loss model is used for calculation. The model formula is as follows:

[0089] ;

[0090] in In distance Total path loss at the location, For reference distance Free space path loss at that location The path damage index, The actual distance of signal transmission. For reference distance, This represents the interference coefficient; in practice, it is obtained by fitting the measured signal strength data using linear regression. and ;

[0091] S32. Combining data transmission characteristic data, use grey relational analysis to quantify the degree of influence of environmental parameters on data transmission quality and generate a correlation matrix.

[0092] First, data transmission quality indicators are used as the reference sequence, and environmental parameters are used as the comparison sequence. Then, the grey relational analysis formula is applied:

[0093] ;

[0094] in For comparing sequences With reference sequence The gray relational degree, with values ​​between 0 and 1, For the reference sequence in the 1st The value of each data point For the first The value of the comparison sequence at the k-th data point The resolution coefficient, For standardization factors, and Let these represent the minimum and maximum differences of all sequences at all points, respectively. The total number of data points; during implementation, the sequence is first initialized to eliminate dimensions, then the correlation coefficient of each point is calculated, and finally the correlation degree is obtained by averaging; the correlation degrees of all environmental parameters form a correlation degree matrix, which is applicable to subsequent weighted fusion;

[0095] S33. Based on the correlation matrix, environmental interference assessment data is generated by weighted fusion, wherein the assessment data includes interference level, expected bit error rate and available bandwidth.

[0096] Weighted fusion formula:

[0097] ;

[0098] in This represents the comprehensive environmental disturbance assessment value. It is the first The weights of each parameter, This is the interference value of the parameter. The total number of environmental interference indicators participating in the fusion; during implementation, each interference indicator is first normalized to the [0,1] interval, and then the weighted sum is calculated; the output includes the interference level, expected bit error rate and available bandwidth.

[0099] In S4, the optimal data transmission strategy is selected, including the following steps:

[0100] S41. Establish a multi-objective optimization function with the goal of maximizing throughput and minimizing latency and packet loss rate, and the constraints include equipment energy limits and environmental interference thresholds.

[0101] The multi-objective optimization function is in the form of:

[0102] ;

[0103] in:

[0104] For network throughput, the objective is to maximize it.

[0105] For data transmission latency, the objective is to minimize it.

[0106] Given packet loss rate, the objective is to minimize it.

[0107] , , The weight coefficients for the corresponding objectives are respectively, and ;

[0108] S42. Use reinforcement learning algorithm as adaptive optimization algorithm to initialize Q-learning model, where the state space is the environmental disturbance evaluation data and the action space is the combination of data transmission strategies.

[0109] The Q-learning update formula is:

[0110] ;

[0111] in In the state Take action below Q value, For learning rate and , For instant rewards, Discount factor and , In order to be in the new state Below, all possible actions The largest Q value;

[0112] S43. By iteratively training the Q-learning model, select the data transmission strategy that maximizes the long-term reward as the optimal strategy, and output the strategy parameters.

[0113] Set a maximum number of iterations T. In each iteration, the agent starts from the current state. Select Action A new state was observed after execution. and rewards Then, the Q-value is adjusted using the update formula; the training stops when the Q-value converges and the maximum number of iterations is reached; the output is the optimal policy parameters, suitable for real-time transmission.

[0114] Real-time transmission of industrial IoT data in S5 includes the following steps:

[0115] S51. Configure the communication module of the industrial IoT device according to the selected optimal data transmission strategy, including setting the routing path, modulation and coding scheme and transmission power;

[0116] S52. Start data stream transmission and use the embedded monitoring agent to collect transmission performance indicators in real time, sampling data once every predetermined time window;

[0117] S53. Aggregate the sampled data to generate a transmission performance log, which is suitable for subsequent dynamic adjustments.

[0118] In S6, the parameters of the data transmission strategy are adjusted in real time, including the following steps:

[0119] S61. Design a proportional-integral-derivative controller as a feedback control mechanism. The input is the deviation between the real-time transmission performance index and the target value, and the output is the adjustment amount of the strategy parameters.

[0120] This step uses a PID controller to achieve real-time parameter adjustment. The PID controller formula is:

[0121] ;

[0122] in To control the output, This is a deviation signal. This is the proportional gain coefficient. This is the integral gain coefficient. This is the differential gain coefficient; during implementation, the deviation is sampled at each time window to calculate the control quantity. The system dynamically adjusts strategy parameters; if the deviation exceeds a preset threshold, adjustments are triggered to ensure the system responds quickly to environmental changes.

[0123] S62. When the deviation exceeds the preset threshold, the controller is triggered to calculate new routing weights and modulation parameters;

[0124] S63. Dynamically update the adjusted parameters to the running data transmission strategy to achieve closed-loop optimization.

[0125] Integrity verification and error recovery in S7 include the following steps:

[0126] S71. At the data receiving end, use cyclic redundancy check and hash algorithm to verify data integrity and request retransmission of erroneous data packets;

[0127] First, CRC checksums are applied to the received data packets: the sending end calculates the CRC code of the data packet (CRC-32), and the receiving end recalculates and compares the CRC codes. If they do not match, a retransmission is requested. Simultaneously, a digest is generated for the data packet using the SHA-256 hash algorithm, and the hash values ​​of the sending and receiving ends are compared. In implementation, a lightweight library is integrated to embed a checksum field into the data transmission protocol. When checksum verification fails, an error log is recorded and a retransmission mechanism is triggered to reduce the risk of data corruption.

[0128] S72. Forward error correction coding technology is used to decode and recover retransmitted data packets, reducing the number of repeated transmissions;

[0129] S73, calculates transmission success rate, average latency, and energy efficiency ratio, and generates an optimized data transmission report.

[0130] S8 stores transmission process data and analysis results, including the following steps:

[0131] S81. Store transmission process data, environmental interference assessment data, and optimization reports in a distributed database on the cloud platform;

[0132] S82. Regularly use stored data to train deep neural network models to improve the prediction accuracy of adaptive optimization algorithms;

[0133] A deep neural network model is trained based on stored historical data, namely transmission process and environmental data. Input features include environmental interference data and transmission strategy parameters, and the output is the predicted performance index. In practice, the model adopts a multilayer perceptron structure with three fully connected layers, with 128, 64 and 32 neurons in each layer, respectively, and ReLU activation function. During training, mean squared error is used as the loss function, Adam optimizer is used, learning rate is set to 0.001, batch size is 64, and training cycle is 100 rounds. Through this training, the model can learn the mapping relationship from environmental state to optimal algorithm parameters.

[0134] S83. Based on the model output, update the policy library for future transmission tasks to call.

[0135] The method also includes security enhancement steps:

[0136] S91. During data transmission, the lightweight encryption algorithm AES-128 is integrated to perform end-to-end encryption on sensitive data;

[0137] S92. Use blockchain technology to record transmission logs to make the data immutable and traceable;

[0138] S93. Implement identity authentication and access control mechanisms to prevent unauthorized devices from accessing the network.

[0139] The steps of this method are as follows:

[0140] First, data transmission parameters and environmental parameters of industrial IoT devices in complex environments are collected to generate raw data transmission datasets and environmental parameter datasets. Then, the raw datasets are preprocessed, including data cleaning, normalization, and feature extraction, to generate preprocessed data transmission feature data and environmental feature data.

[0141] Then, based on the preprocessed data, a quantitative analysis of the impact of complex environments on data transmission is performed to generate environmental interference assessment data, which includes signal attenuation, multipath interference, and electromagnetic noise. According to the environmental interference assessment data, an adaptive optimization algorithm is used to select the optimal data transmission strategy, which covers routing selection, modulation method adjustment, and transmission power control. Subsequently, industrial IoT data is transmitted in real time according to the optimal data transmission strategy, and performance indicators such as throughput, latency, and packet loss rate are dynamically monitored during the transmission process.

[0142] Based on dynamic monitoring results, a feedback control mechanism is used to adjust the parameters of the data transmission strategy in real time to cope with environmental changes; integrity verification and error recovery are performed on the transmitted data, and an optimized data transmission report is generated; finally, the transmission process data and analysis results are stored, which are suitable for subsequent machine learning model training and strategy optimization.

[0143] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0144] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing data transmission in industrial IoT under complex environments, characterized in that: The method includes the following steps: S1. Collect data transmission parameters and environmental parameters of industrial IoT devices in complex environments, and generate raw data transmission datasets and environmental parameter datasets; S2. Preprocess the original data transmission dataset and environmental parameter dataset, including data cleaning, normalization and feature extraction, to generate preprocessed data transmission feature data and environmental feature data; S3. Based on the preprocessed data transmission characteristic data and environmental characteristic data, perform a quantitative analysis of the impact of complex environment on data transmission, and generate environmental interference assessment data, wherein environmental interference includes signal attenuation, multipath interference and electromagnetic noise. The quantitative analysis of the impact of the complex environment on data transmission in S3 includes the following steps: S31. Based on the preprocessed environmental feature data, calculate the path loss index and interference coefficient in the signal propagation model to generate preliminary environmental interference indicators. S32. Combining data transmission characteristic data, use grey relational analysis to quantify the degree of influence of environmental parameters on data transmission quality and generate a correlation matrix. S33. Based on the correlation matrix, environmental interference assessment data is generated by weighted fusion, wherein the assessment data includes interference level, expected bit error rate and available bandwidth. S4. Based on the environmental interference assessment data, an adaptive optimization algorithm is used to select the optimal data transmission strategy, which includes route selection, modulation method adjustment, and transmission power control. Selecting the optimal data transmission strategy in step S4 includes the following steps: S41. Establish a multi-objective optimization function with the goal of maximizing throughput and minimizing latency and packet loss rate, and the constraints include equipment energy limits and environmental interference thresholds. S42. Use reinforcement learning algorithm as adaptive optimization algorithm to initialize Q-learning model, where the state space is the environmental disturbance evaluation data and the action space is the combination of data transmission strategies. S43. By iteratively training the Q-learning model, select the data transmission strategy that maximizes the long-term reward as the optimal data transmission strategy, and output the strategy parameters. S5. Based on the optimal data transmission strategy, perform real-time transmission of industrial IoT data, and dynamically monitor transmission performance indicators, including throughput, latency and packet loss rate, during the transmission process. S6. Based on dynamic monitoring results, use a feedback control mechanism to adjust the parameters of the data transmission strategy in real time to cope with environmental changes; S7. Perform integrity verification and error recovery processing on the transmitted data, and generate an optimized data transmission report; S8 stores and transmits data and analysis results, which are suitable for subsequent machine learning model training and strategy optimization.

2. The method for optimizing industrial IoT data transmission in complex environments according to claim 1, characterized in that: The step S1, which involves collecting data transmission parameters and environmental parameters of industrial IoT devices in complex environments, includes the following steps: S11. Collect real-time data transmission parameters of connected devices through industrial IoT gateway devices, including data packet size, transmission frequency, protocol type and device identifier, and generate raw data transmission dataset; S12. Collect complex environmental parameters using an environmental sensor network, including temperature, humidity, signal strength, noise level, and distribution of moving obstacles, and generate an environmental parameter dataset. S13. Perform timestamp alignment and spatial association processing on the original data transmission dataset and the environmental parameter dataset.

3. The method for optimizing industrial IoT data transmission in complex environments according to claim 2, characterized in that: The preprocessing in S2 includes the following steps: S21. Perform data cleaning on the original data transmission dataset to remove outliers and duplicate data, and fill in missing values; S22. The cleaned data is normalized using the min-max normalization method to scale all parameters to the [0,1] interval. S23. Extract key features from normalized data, including data flow patterns, environmental periodic change features, and interference correlation features, to generate preprocessed data transmission feature data and environmental feature data.

4. The method for optimizing industrial IoT data transmission in complex environments according to claim 3, characterized in that: The real-time transmission of industrial IoT data in S5 includes the following steps: S51. Configure the communication module of the industrial IoT device according to the selected optimal data transmission strategy, including setting the routing path, modulation and coding scheme and transmission power; S52. Start data stream transmission and use the embedded monitoring agent to collect transmission performance indicators in real time, sampling data once every predetermined time window; S53. Aggregate the sampled data to generate a transmission performance log, which is suitable for subsequent dynamic adjustments.

5. The method for optimizing industrial IoT data transmission in complex environments according to claim 4, characterized in that: The parameters for real-time adjustment of the data transmission strategy in S6 include the following steps: S61. Design a proportional-integral-derivative controller as a feedback control mechanism. The input is the deviation between the real-time transmission performance index and the target value, and the output is the adjustment amount of the strategy parameters. S62. When the deviation exceeds the preset threshold, the controller is triggered to calculate new routing weights and modulation parameters; S63. Dynamically update the adjusted parameters to the running data transmission strategy to achieve closed-loop optimization.

6. The method for optimizing industrial IoT data transmission in a complex environment according to claim 5, characterized in that: The integrity verification and error recovery process in S7 includes the following steps: S71. At the data receiving end, use cyclic redundancy check and hash algorithm to verify data integrity and request retransmission of erroneous data packets; S72. Forward error correction coding technology is used to decode and recover retransmitted data packets, reducing the number of repeated transmissions; S73, calculates transmission success rate, average latency, and energy efficiency ratio, and generates an optimized data transmission report.

7. The method for optimizing industrial IoT data transmission in complex environments according to claim 6, characterized in that: The S8 process stores transmission data and analysis results, including the following steps: S81. Store transmission process data, environmental interference assessment data, and optimization reports in a distributed database on the cloud platform; S82. Regularly use stored data to train deep neural network models to improve the prediction accuracy of adaptive optimization algorithms; S83. Based on the model output, update the policy library for future transmission tasks to call.

8. The method for optimizing industrial IoT data transmission in complex environments according to claim 1, characterized in that: The method also includes security enhancement steps: S91. During data transmission, the lightweight encryption algorithm AES-128 is integrated to perform end-to-end encryption on sensitive data; S92. Use blockchain technology to record transmission logs to make the data immutable and traceable; S93. Implement identity authentication and access control mechanisms to prevent unauthorized devices from accessing the network.

Images