CN116471307B

CN116471307B - Internet of things heterogeneous data cascade transmission method, device, equipment and medium

Info

Publication number: CN116471307B
Application number: CN202310731540.6A
Authority: CN
Inventors: 陈浩; 张渊; 刘明扬
Original assignee: Beijing Zhongkelangyi Technology Co ltd
Current assignee: Beijing Zhongkelangyi Technology Co ltd
Priority date: 2023-06-20
Filing date: 2023-06-20
Publication date: 2023-08-22
Anticipated expiration: 2043-06-20
Also published as: CN116471307A

Abstract

The invention provides a method, a device, equipment and a medium for cascade transmission of heterogeneous data of the Internet of things, which relate to the technical field of operation and maintenance management and comprise the following steps: acquiring first information and second information; clustering the data condition of each type of internet of things equipment in the second information to obtain third information; performing data segmentation processing on the first information according to the time window information in the third information to obtain fourth information; performing time sequence analysis processing according to the fourth information and the second information to obtain fifth information; performing feature extraction on the fifth information according to a preset deep learning mathematical model to obtain sixth information; and transmitting the sixth information and performing layer-by-layer decoding processing to obtain seventh information. The invention can extract the important characteristics of the data by applying the technologies of deep learning, time sequence analysis and the like, and optimally codes the data, so that the volume of the data in the transmission process is obviously reduced, and the transmission efficiency of the data is improved.

Description

Internet of things heterogeneous data cascade transmission method, device, equipment and medium

Technical Field

The invention relates to the technical field of operation and maintenance management, in particular to a method, a device, equipment and a medium for cascade transmission of heterogeneous data of the Internet of things.

Background

In recent years, internet of things devices play an increasingly important role in real-time data collection and processing. The data generated by the devices has high heterogeneity, including various types of device states, environment parameters, device running logs, device fault logs and the like, and the accurate transmission and processing of the data are critical to the stable operation of the internet of things system. However, the existing data transmission method of the internet of things faces serious challenges due to the differences in data types and formats between devices, and the large scale and complexity of data. Currently, conventional data transmission of the internet of things mainly relies on some simple encoding and transmission strategies, for example, encoding data into a uniform format and then transmitting the data through a network. However, this method cannot solve the problem of cascade transmission of heterogeneous data, cannot effectively integrate and optimize various types of data, and cannot guarantee the robustness and integrity of the data in the transmission process.

Based on the shortcomings of the prior art, a method for cascade transmission of heterogeneous data of the Internet of things is needed.

Disclosure of Invention

The invention aims to provide a method, a device, equipment and a medium for cascade transmission of heterogeneous data of the Internet of things, so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:

In a first aspect, the present application provides a method for cascade transmission of heterogeneous data in the internet of things, including:

acquiring first information and second information, wherein the first information comprises current heterogeneous data generated by Internet of things equipment, and the second information comprises historical heterogeneous data generated by the Internet of things equipment;

clustering the data condition of each type of internet of things equipment in the second information to obtain third information, wherein the third information comprises a time window for optimal data analysis of each type of internet of things equipment;

performing data segmentation processing on the corresponding type of the Internet of things equipment data in the first information according to the time window information in the third information to obtain fourth information;

performing time sequence analysis processing according to the fourth information and the second information to obtain fifth information, wherein the fifth information comprises a data trend prediction result;

extracting features of the fifth information according to a preset deep learning mathematical model, and carrying out layered coding processing on the extracted data features to obtain sixth information;

and transmitting the sixth information, and performing layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, wherein the seventh information is original heterogeneous data restored after decoding.

In a second aspect, the present application further provides an internet of things heterogeneous data cascade transmission device, including:

the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring first information and second information, the first information comprises current heterogeneous data generated by Internet of things equipment, and the second information comprises historical heterogeneous data generated by the Internet of things equipment;

the clustering module is used for carrying out clustering processing on the data condition of each type of internet of things equipment in the second information to obtain third information, and the third information comprises a time window for optimal data analysis of each type of internet of things equipment;

the segmentation module is used for carrying out data segmentation processing on the corresponding type of the internet of things equipment data in the first information according to the time window information in the third information to obtain fourth information;

the analysis module is used for carrying out time sequence analysis processing according to the fourth information and the second information to obtain fifth information, wherein the fifth information comprises a data trend prediction result;

the coding module is used for extracting the characteristics of the fifth information according to a preset deep learning mathematical model, and carrying out layered coding processing on the extracted data characteristics to obtain sixth information;

And the decoding module is used for transmitting the sixth information, and carrying out layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, wherein the seventh information is original heterogeneous data restored after decoding.

In a third aspect, the present application further provides an internet of things heterogeneous data cascade transmission device, including:

a memory for storing a computer program;

and the processor is used for realizing the step of the internet of things heterogeneous data cascade transmission method when executing the computer program.

In a fourth aspect, the present application further provides a medium, where a computer program is stored, where the computer program when executed by a processor implements the steps of the method for cascade transmission of heterogeneous data in the internet of things.

The beneficial effects of the application are as follows:

the application can extract important characteristics of data by applying technologies such as deep learning, time sequence analysis and the like, and optimally codes the important characteristics, so that the volume of the data in the transmission process is obviously reduced, and the transmission efficiency of the data is improved; by means of clustering and data segmentation of the data, the method and the device can effectively integrate heterogeneous data from different devices and different formats, so that consistency of the data in the transmission and processing processes is ensured.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a cascade transmission method of heterogeneous data of the internet of things according to an embodiment of the invention;

fig. 2 is a schematic structural diagram of an internet of things heterogeneous data cascade transmission device according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an internet of things heterogeneous data cascade transmission device according to an embodiment of the present invention.

The marks in the figure: 1. an acquisition module; 2. a clustering module; 21. a first processing unit; 22. a first extraction unit; 23. a first clustering unit; 24. a first calculation unit; 3. a segmentation module; 31. a first distribution unit; 32. a first sorting unit; 33. a first dividing unit; 34. a first integration unit; 4. an analysis module; 41. a first detection unit; 42. a second processing unit; 43. a first fitting unit; 44. a third processing unit; 5. a coding module; 51. a second extraction unit; 52. a first mapping unit; 53. a first encoding unit; 54. a first conversion unit; 6. a decoding module; 61. a first enhancement unit; 62. a first scheduling unit; 63. a second enhancement unit; 64. a first decoding unit; 800. the heterogeneous data cascade transmission equipment of the Internet of things; 801. a processor; 802. a memory; 803. a multimedia component; 804. an I/O interface; 805. a communication component.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.

Example 1:

the embodiment provides a cascade transmission method for heterogeneous data of the Internet of things.

Referring to fig. 1, the method is shown to include steps S100, S200, S300, S400, S500, and S600.

Step S100, acquiring first information and second information, wherein the first information comprises current heterogeneous data generated by the Internet of things equipment, and the second information comprises historical heterogeneous data generated by the Internet of things equipment.

It can be understood that in this step, the first information and the second information read data from the device through the communication interface of the internet of things device, or acquire data from the data storage center of the device. The current heterogeneous data refer to data generated by the Internet of things equipment at the current time point or in the near-term time, the data comprise the latest state and the running condition of the equipment, and the current heterogeneous data play an important role in understanding the current working condition and the instant condition of the equipment. Historical heterogeneous data refers to data generated by equipment in a past period of time, records historical operation conditions and state changes of the equipment, and has important reference value in the aspects of analyzing operation rules and trends of the equipment, predicting future states of the equipment and the like.

Step 200, clustering the data condition of each type of internet of things equipment in the second information to obtain third information, wherein the third information comprises a time window for optimal data analysis of each type of internet of things equipment.

It can be understood that in this step, the clustering of the historical heterogeneous data, that is, the aggregation of the device data with similar characteristics, is not only helpful to discover the intrinsic rules of the device from the data, but also to optimize the subsequent data processing and analysis flow. For example, we can cluster together the same type of device data, and unified adjustments and optimizations can be made to the maintenance policies of this type of device. The optimal data analysis time window is the optimal time length for data analysis for each type of internet of things device. By selecting a proper time window, the integrity of data and the accuracy of analysis can be ensured, and meanwhile, the efficiency of data processing can be improved. It should be noted that step S200 includes step S210, step S220, step S230, and step S240.

And step S210, carrying out data preprocessing on the historical heterogeneous data of each type of Internet of things equipment according to the second information to obtain preprocessed data, wherein the preprocessing comprises data cleaning, data standardization and data missing value processing.

It will be appreciated that in this step, the data cleansing is to remove extraneous information in the data, such as noise data, anomaly data, etc., so as to ensure that the data is analyzed without being disturbed by such extraneous information. Data normalization is to eliminate the dimensional effects between different devices, different data types, so that the data can be compared on the same scale. For example, when comparing temperature data and humidity data, the dimensions of the two data are different, direct comparison can lead to distortion of the results, and by data normalization, they can be converted to the same scale for comparison. The data missing value processing is used for processing missing situations in data and ensuring the integrity of the data. In the operation process of the internet of things equipment, data loss conditions, such as network transmission faults, equipment faults and the like, may be generated due to various reasons. At this point, proper processing of these missing values is required to avoid affecting subsequent data analysis. The data preprocessing is used for improving the quality of data, so that useful information can be extracted from accurate and reliable data, and support is provided for subsequent equipment operation and maintenance management.

Step S220, carrying out feature extraction and natural language processing on the preprocessed data according to a preset self-encoder mathematical model to obtain a feature vector set, wherein the feature vector set comprises equipment state features, environment parameter features, time features and equipment log features, the equipment state features comprise equipment running states, equipment using frequencies and equipment fault frequencies, the environment parameter features comprise average values, maximum values, minimum values, fluctuation ranges and the like of parameters of the environment where the equipment is located, the time features comprise equipment data generation time, equipment using time and equipment using period, and the equipment log features comprise keywords, topics and emotion features.

It can be understood that in this step, the pre-processed device data is subjected to feature extraction and natural language processing using a preset self-encoder mathematical model. This step converts the operating information of the device, including status, frequency of use, failure frequency, etc., environmental parameters, time characteristics, and device log, etc., into feature vectors that can be analyzed. This process provides a comprehensive view of our understanding of various aspects of device operation. The equipment state characteristics can reflect the running health condition of the equipment, so that operation and maintenance personnel can predict and timely find possible problems; the environmental parameter characteristics can enable operation and maintenance personnel to know whether the operation environment of the equipment is ideal or not, because environmental factors can influence the performance and service life of the equipment; the time features can help operation staff to know the use mode of the equipment, including the use peak period, the most frequent use period and the like, so that equipment scheduling and maintenance planning can be effectively performed; the log feature of the device can reflect detailed information in the running process of the device, such as the specific situation of fault occurrence, which is also important for diagnosing and preventing the fault. In the embodiment, through the deep analysis of the equipment data, operation and maintenance personnel can understand the operation condition of the equipment more deeply and predict the future state of the equipment more accurately, so that operation and maintenance strategies and measures are formulated more effectively, the operation efficiency of the equipment is improved, faults are reduced, and the operation and maintenance cost is reduced.

Step S230, carrying out cluster analysis on each type of internet of things equipment in the feature vector set according to a preset spectral clustering mathematical model to obtain a cluster result, wherein the cluster result comprises cluster labels, characteristic descriptions and statistical information of equipment states of the internet of things equipment.

It will be appreciated that cluster analysis in this step may help understand the behavior patterns and interrelationships of the various devices. For example, by means of the cluster labels of the devices, it is possible to identify which devices have similar usage patterns and which devices have similar failure rates, and even to find out environmental factors, such as temperature, humidity, etc., which may affect the performance of the devices. This information allows us to better understand the behavior of the device and formulate a more efficient operation and maintenance strategy accordingly. Spectral clustering is a graph theory-based clustering method, and a data set is divided into different subsets or classes by searching for optimal segmentation of the data. The method has the advantages that the complex structure and mode of the data can be found, and the method is not limited by the shape and the size of the data, so that the method is suitable for processing high-dimensional, complex and heterogeneous data generated by the Internet of things equipment. Furthermore, the spectral clustering method converts original high-dimensional data into low-dimensional data by introducing a similarity matrix and a Laplacian matrix, so that the complexity of data processing is reduced, and the clustering result is more explanatory. This is helpful for understanding the behavior pattern of the internet of things device, finding out the statistical information of the device state, and determining the optimal data analysis time window.

And step S240, calculating an average time window of the data of the Internet of things equipment in each cluster according to the clustering result.

It can be appreciated that in operation and maintenance management of devices of the internet of things, it is extremely important to monitor, analyze and adapt to the state of the devices in time. For heterogeneous device data processing, if each device or each class of devices uses the same time window for analysis, the variability between devices and within devices may be ignored, thereby affecting the accuracy and efficiency of data analysis. Therefore, the step realizes personalized processing of different devices and different types of devices by calculating the average time window of the data of the devices of the Internet of things in each cluster. The average time window is not fixed, but dynamically adjusted according to the clustering result, so that the change of the equipment state is better adapted, and the accuracy and the efficiency of data analysis are improved. The calculation formula is as follows:

；

wherein, the liquid crystal display device comprises a liquid crystal display device,an average time window representing a kth cluster; />Representing the number of the internet of things devices in the kth cluster; />And->Representing the real and imaginary parts of the eigenvectors of the device i in the cluster k on the complex plane, respectively, which determine the frequency and phase of the complex exponential function; t represents the moment of data acquisition of the equipment of the Internet of things; / >The average time of data acquisition of the internet of things device i in the cluster k. />And representing the data value of the internet of things device i in the cluster k at the time t.

And step S300, carrying out data segmentation processing on the corresponding type of the Internet of things equipment data in the first information according to the time window information in the third information to obtain fourth information.

It can be understood that in this step, the original heterogeneous data is divided according to the time windows, so that the data in each time window has similar characteristics and statistical properties, and the subsequent time sequence analysis and data processing operations are convenient. Such data division processing is useful for extracting time series information of data and exploring a trend of change of the data. It should be noted that step S300 includes step S310, step S320, step S330, and step S340.

And step S310, identifying each type of the Internet of things equipment and corresponding time window information thereof according to the first information and the third information, and performing data distribution processing to obtain an equipment data set, wherein the equipment data set comprises all equipment data with the same equipment type label.

It will be appreciated that by dividing the device data set by device type and time window in this step, the operator can more finely monitor and manage the status and performance changes of different types of devices, helping to discover differences between devices, compare and analyze device performance, and identify problems and optimization directions. Meanwhile, the division of the time window provides a basis for time sequence analysis, and the periodic change, trend and abnormal behavior of the equipment are revealed, so that the equipment fault prediction and maintenance planning are facilitated. In addition, different types of internet of things device data have different characteristics and statistical properties. By partitioning the data set by device type and time window, the heterogeneous data can be better organized and managed. By the aid of the method, more accurate and reliable input can be provided for subsequent processing methods such as feature extraction, time sequence analysis and data mining, and accordingly data processing efficiency and accuracy are improved.

Step S320, sorting the device data sets according to the time window information in the third information to obtain ordered device data.

It can be appreciated that in this step, by sorting the time window of the device data set, the device data can be organized according to the time sequence, so that the operation and maintenance personnel can better understand the state evolution, the performance change and the fault condition of the device. Further, the ordered device data can reveal the relevance and interaction between devices, and help build a relevance model between the devices, so that intelligent decision making and operation and maintenance strategy optimization are supported.

And step S330, cutting the equipment data according to the time window information to obtain data blocks, wherein each data block comprises the equipment data in a period of continuous time.

It can be understood that in this step, the original data is cut according to the time window, so that the data can be organized into data blocks within a period of continuous time, so that the time sequence of the data is clearer and manageable. The method is favorable for operation and maintenance personnel to carry out finer observation and analysis on the equipment data and grasp the change and trend of the working state of the equipment.

And step S340, carrying out data integration processing according to the data block and the equipment type to obtain fourth information.

It can be understood that the data of the same equipment type is integrated in the step, so that the common characteristics and the statistical characteristics of the equipment of the type can be better mined. By extracting and analyzing the characteristics of the integrated data, the specific behavior mode, performance index and abnormal condition of the equipment can be revealed. The method is beneficial to operation and maintenance personnel to quickly know the difference and commonality between the types of the equipment, and provides basis for formulating corresponding maintenance and management strategies.

And step 400, performing time sequence analysis processing according to the fourth information and the second information to obtain fifth information, wherein the fifth information comprises a data trend prediction result.

It can be understood that in this step, by analyzing and predicting the trend of the equipment data, the operation and maintenance personnel can find potential problems and abnormal situations in time and take corresponding measures to intervene. In addition, the data trend prediction result can also help an operation and maintenance team to make a reasonable maintenance plan, optimize the allocation and utilization of resources and improve the reliability and performance of equipment. It should be noted that step S400 includes step S410, step S420, step S430, and step S440.

And step S410, detecting abnormal values of the second information according to a preset isolated forest mathematical model to obtain normal data.

It will be appreciated that an isolated forest is an unsupervised learning algorithm for detecting outliers. In this step, the preset isolated forest mathematical model evaluates and analyzes the device data based on the features and statistical properties in the second information, and identifies samples having similar features to the normal data, and regards the samples as the normal data. And carrying out outlier detection on the second information through the isolated forest model to obtain normal equipment data, wherein the normal equipment data have characteristics and statistical characteristics which are obviously different from other data samples, so that the normal equipment data are considered to be normal equipment behaviors. The screening of normal data is very important for subsequent data processing and analysis, and the accuracy of data analysis and the credibility of results can be improved.

And S420, carrying out trending processing according to the normal data, and eliminating the seasonality and the trend of the data through difference and moving average calculation to obtain trending data.

It will be appreciated that in time series data analysis, trend refers to the overall trend or regular change in data over time. Trend analysis is an important method for understanding the long-term development and trend change of data, and can reveal long-term change trend, periodic fluctuation and seasonal change of equipment data. However, in the application scenario of the present invention, the presence of trends may interfere with the analysis and prediction of data. Factors such as long-term trends and seasonal variations may mask short-term variations and abnormal behavior of the device, making data analysis and prediction difficult. Differencing is a common trending method that eliminates trends and seasonal variations in data by calculating differences between data points. By performing a first order differential or a high order differential operation on normal data, the trend influence of the data can be reduced, so that the data is more stable. In addition, the moving average calculation is a method of smoothing time-series data, and noise and random fluctuation of the data are reduced by calculating a sliding window average value of the data, so that the data are more stable. The normal data is thus trended by a combination of differential and moving average calculations to eliminate the seasonal and trending of the data.

And step S430, performing time sequence model fitting processing on the trending data according to a preset state space mathematical model to obtain a model fitting result for describing the change rule of the equipment data.

It can be understood that in this step, by fitting a time series model, a mathematical description of the change of the device data can be established, and parameters and status information of the model can be obtained. The method has the advantages that the characteristics, the trend and the periodic change of the equipment data can be better understood, and the method provides basis for subsequent data analysis, prediction and decision. And by using the fitting result, the future state change of the equipment can be predicted and simulated, so that the operation and maintenance management and the resource allocation are further optimized, and the stability and the performance of the equipment are improved.

And step S440, carrying out data prediction processing on the fourth information according to the model fitting result to obtain a data trend prediction result, wherein the data trend prediction result comprises the state change condition of the equipment in the future preset time.

It will be appreciated that by using the established time series model in this step, the trend of the change in the future operating state, performance index or other key index of the device can be inferred from past data trends and model parameters. The data prediction result is critical to operation and maintenance management, and can help operation and maintenance personnel to identify possible problems, faults or changes of equipment in advance, take corresponding measures and make maintenance plans and resource scheduling in advance. Through data trend prediction, the operation and maintenance team can manage the equipment more efficiently, reduce downtime, improve the reliability and performance of the equipment, optimize maintenance cost and improve operation and maintenance efficiency.

And S500, carrying out feature extraction on the fifth information according to a preset deep learning mathematical model, and carrying out layered coding processing on the extracted data features to obtain sixth information.

It can be understood that in this step, by extracting the features of the deep learning model, high-level abstract features in the data can be captured, which is helpful for better understanding and analyzing the internal modes and association relations of the data of the internet of things device. In addition, the transmission efficiency and the safety of the data can be further improved by carrying out hierarchical coding processing on the extracted data characteristics. Layered coding can compress and transform data features, making the data more compact and easier to transmit. Meanwhile, hierarchical coding can also provide a certain degree of data protection and privacy confidentiality, so that the safety of data transmission is enhanced. It should be noted that step S500 includes step S510, step S520, step S530, and step S540.

And step S510, extracting principal components from trend data of the state change of the Internet of things equipment in the fifth information to obtain principal component vectors.

It will be appreciated that principal component analysis is a commonly used dimension reduction technique in which raw data is projected by linear transformation into a new coordinate system such that the projected variables are uncorrelated with each other. According to the method, the dimension reduction is carried out on the state change trend data of the Internet of things equipment in the fifth information by using a principal component analysis method, and the principal component vector which can most represent the original data change mode is extracted. Through principal component extraction, the main mode and the characteristics of the state change of the equipment of the Internet of things can be captured without being interfered by secondary factors. This helps to reduce the dimensionality of the data, simplify the expression and transmission of the data, and improve the efficiency of data processing and analysis.

And step S520, mapping the principal component vector to a high-dimensional space through a preset kernel function mathematical model to obtain a mapped equipment state feature vector.

It is appreciated that the kernel function is capable of mapping low-dimensional data into high-dimensional space and capturing complex relationships in the data through nonlinear transformation. In this embodiment, the mapping is performed on the principal component vector through the preset kernel function mathematical model, and the principal component vector is converted into a vector in the high-dimensional feature space, so that the expression capability and the discrimination of the feature are improved, the state change of the internet of things equipment can be better understood and analyzed, and more accurate and comprehensive equipment state information is provided for subsequent data processing and analysis.

And step S530, carrying out feature vector layering coding by utilizing a preset Huffman coding mathematical model according to the equipment state feature vector to obtain a coded feature vector of the equipment state.

It will be appreciated that in this step, the device state feature vectors may be encoded in terms of their importance and frequency by huffman encoding, such that common feature vectors may be represented by shorter encodings while unusual feature vectors may be represented by longer encodings. The coding mode can effectively reduce the storage space and transmission bandwidth of the feature vector, and meanwhile, the integrity and the reducibility of data are maintained. By utilizing a preset Huffman coding mathematical model to code the equipment state feature vector, the data volume can be reduced in the data transmission and storage process of the Internet of things, and the data transmission efficiency and response speed can be improved. Meanwhile, the transmission and storage of the coding feature vector saves more resources, and meets the requirements on bandwidth and storage capacity in the environment of the Internet of things.

Step S540, performing format conversion processing on the coding feature vector to obtain a coded data sequence of the equipment state.

It can be understood that in this step, the coded feature vector of the device state is converted from a vector form to a sequence form, so as to facilitate effective transmission and storage of data in the transmission process of the internet of things. Preferably, the data sequence is converted into a binary format. The execution of the step enables the equipment state data to be transmitted, stored and processed more effectively, and provides a basis for realizing cascade transmission of heterogeneous data of the Internet of things.

And step S600, transmitting sixth information, and carrying out layer-by-layer decoding processing on the transmitted data sequence according to a deep learning mathematical model to obtain seventh information, wherein the seventh information is original heterogeneous data restored after decoding.

It will be appreciated that this step involves communication means such as network transmission, packet transmission, etc. to ensure that data can be transmitted from the sender to the receiver. And carrying out layer-by-layer decoding processing on the transmitted data sequence by using a preset deep learning mathematical model. This decoding process is the inverse of the previous encoding operation, aimed at restoring the original heterogeneous data. The original characteristics and structure of the data can be gradually restored through layer-by-layer decoding to obtain seventh information, namely the restored original heterogeneous data after decoding. The data comprise heterogeneous data generated by original Internet of things equipment, wherein the heterogeneous data comprise equipment state, environment parameters, equipment logs and the like. It should be noted that step S600 includes step S610, step S620, step S630, and step S640.

Step S610, signal enhancement is performed on the sixth information according to a preset WaveNet mathematical model, and a first data sequence is obtained.

It can be appreciated that WaveNet is a deep learning model that is mainly used in the fields of speech synthesis and audio processing. It is based on the idea of generating a model, which is capable of modeling a high quality audio signal. In the step, the characteristic of the WaveNet model is used as a reference, and the WaveNet model is applied to the signal enhancement process of heterogeneous data of the Internet of things. By means of the WaveNet model, we can model the data sequence in the sixth information and generate a high quality data sequence using the model. This process of signal enhancement helps to improve the accuracy, integrity and reliability of the data. It can eliminate noise, distortion or other interference possibly introduced in the transmission process, so that the data sequence is clearer and more reliable.

Step S620, according to the first data sequence, dynamic scheduling is carried out according to the characteristics of heterogeneous data such as equipment states, environment parameters and the like and a preset scheduling strategy to obtain a second data sequence, wherein the second data sequence comprises data transmission parameters, and the data transmission parameters comprise data priority and transmission rate.

It is understood that in an internet of things environment, the nature and importance of heterogeneous data may vary. Some data may have a higher priority and need to be handled and transmitted preferentially during transmission. Some data may have higher transmission rate requirements, and timeliness and real-time performance of the data need to be guaranteed. Further, in this step, according to the real-time change of the device state, the change of the environmental parameter, and other factors, the transmission sequence and the transmission rate of the data are dynamically adjusted, so as to meet the priority and the transmission requirement of the data. Through dynamic scheduling, the transmission sequence and the transmission rate of the data can be reasonably arranged according to the characteristics and the importance of different data, so that the transmission efficiency and the instantaneity of the data are improved. Therefore, timely transmission of important data can be ensured, data transmission delay is reduced, and the performance and response capability of the whole system are improved.

And step 630, performing redundancy information enhancement according to the second data sequence, and performing redundancy transmission on the data with the enhanced redundancy information to obtain a third data sequence.

It is understood that redundancy information enhancement refers to adding redundancy information during data transmission to improve the reliability and fault tolerance of the data. In the internet of things, because of the unstable network and the possible loss or damage of data transmission, in order to ensure the integrity and reliability of data, a redundant information technology is required. In this step, redundancy encoding is performed on the second data sequence according to a preset redundancy information generating algorithm, and redundancy information is introduced into the data by adding redundancy bits or using redundancy check codes. By the aid of the method, the redundancy of the data can be improved, the reliability of the data can be improved, and the data has higher fault tolerance. And then carrying out redundant transmission on the data enhanced by the redundant information. Redundant transmission refers to transmitting redundant information together with original data in a data transmission process. Therefore, even if partial data is lost or damaged in the transmission process, the original data can be recovered and corrected through the redundant information, so that the receiving end can be ensured to correctly reconstruct. By the redundancy information enhancement and redundancy transmission, the reliability and fault tolerance of the data can be improved, and errors and losses in the data transmission can be reduced. This is very important for scenarios with high requirements for data integrity and reliability in the internet of things, such as remote monitoring, device control, etc.

And step 640, according to the third data sequence, decoding by using a deep learning mathematical model and removing redundant information to recover the original heterogeneous data, thereby obtaining the recovered original heterogeneous data after decoding.

It can be understood that in this step, the third data sequence is first input into the deep learning model, the model learns the patterns and rules in the sequence, and the data is decoded through the calculation and mapping process of the network layer, so as to gradually recover the form and content of the original heterogeneous data. The relationship between the original data and the redundant data is then analyzed and processed using the model, and the redundant information can be removed from the decoded data by appropriate computation and processing to recover the original heterogeneous data.

Example 2:

as shown in fig. 2, this embodiment provides a device for cascade transmission of heterogeneous data of internet of things, where the device includes:

the acquisition module 1 is configured to acquire first information and second information, where the first information includes current heterogeneous data generated by the internet of things device, and the second information includes historical heterogeneous data generated by the internet of things device.

And the clustering module 2 is used for carrying out clustering processing on the data condition of each type of the internet of things equipment in the second information to obtain third information, wherein the third information comprises a time window for optimal data analysis of each type of the internet of things equipment.

And the segmentation module 3 is used for carrying out data segmentation processing on the corresponding type of the internet of things equipment data in the first information according to the time window information in the third information to obtain fourth information.

And the analysis module 4 is used for carrying out time sequence analysis processing according to the fourth information and the second information to obtain fifth information, wherein the fifth information comprises a data trend prediction result.

And the encoding module 5 is used for extracting the characteristics of the fifth information according to a preset deep learning mathematical model, and carrying out layered encoding processing on the extracted data characteristics to obtain sixth information.

And the decoding module 6 is used for transmitting the sixth information, and carrying out layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, wherein the seventh information is the original heterogeneous data restored after decoding.

In one embodiment of the present disclosure, the clustering module 2 includes:

the first processing unit 21 is configured to perform data preprocessing on historical heterogeneous data of each type of internet of things device according to the second information, so as to obtain preprocessed data, where the preprocessing includes data cleaning, data standardization and data missing value processing.

The first extracting unit 22 is configured to perform feature extraction and natural language processing on the preprocessed data according to a preset self-encoder mathematical model to obtain a feature vector set, where the feature vector set includes a device state feature, an environment parameter feature, a time feature and a device log feature, the device state feature includes an operation state of a device, a use frequency of the device and a fault frequency of the device, the environment parameter feature includes an average value, a maximum value, a minimum value, a fluctuation range and the like of parameters such as temperature, humidity, air pressure and the like of an environment where the device is located, the time feature includes a time of generating the device data, a use time of the device and a use period of the device, and the device log feature includes a keyword, a theme and an emotion feature.

The first clustering unit 23 is configured to perform cluster analysis on each type of internet of things device in the feature vector set according to a preset spectral clustering mathematical model to obtain a clustering result, where the clustering result includes a clustering label, a feature description and statistical information of a device state of the internet of things device.

The first calculating unit 24 is configured to calculate an average time window of the data of the devices of the internet of things in each cluster according to the clustering result.

In one embodiment of the present disclosure, the segmentation module 3 includes:

the first allocation unit 31 is configured to identify each type of internet of things device and its corresponding time window information according to the first information and the third information, and perform data allocation processing to obtain a device data set, where the device data set includes all device data with the same device type tag.

A first ordering unit 32, configured to order the device data sets according to the time window information in the third information, so as to obtain ordered device data.

The first dividing unit 33 is configured to perform a cutting process on the device data according to the time window information, so as to obtain data blocks, where each data block includes the device data in a continuous time.

The first integrating unit 34 is configured to perform data integration processing according to the device type according to the data block, so as to obtain fourth information.

In one embodiment of the present disclosure, the analysis module 4 includes:

the first detection unit 41 is configured to perform outlier detection on the second information according to a preset isolated forest mathematical model, so as to obtain normal data.

The second processing unit 42 is configured to perform trending processing according to the normal data, and remove seasonality and trend of the data by difference and moving average calculation to obtain trended data.

The first fitting unit 43 is configured to perform a time-series model fitting process on the trending data according to a preset state space mathematical model, so as to obtain a model fitting result for describing a change rule of the device data.

The third processing unit 44 is configured to perform data prediction processing on the fourth information according to the model fitting result to obtain a data trend prediction result, where the data trend prediction result includes a state change condition of the device in a future preset time.

In one embodiment of the present disclosure, the encoding module 5 includes:

the second extracting unit 51 is configured to extract a principal component from trend data of the state change of the internet of things device in the fifth information, so as to obtain a principal component vector.

The first mapping unit 52 is configured to map the principal component vector to a high-dimensional space through a preset kernel function mathematical model, to obtain a mapped device state feature vector.

The first encoding unit 53 is configured to perform feature vector hierarchical coding according to the device state feature vector by using a preset huffman coding mathematical model, so as to obtain a coded feature vector of the device state.

The first converting unit 54 is configured to perform format conversion processing on the encoded feature vector to obtain an encoded data sequence of the device state.

In one embodiment of the present disclosure, the decoding module 6 includes:

the first enhancing unit 61 is configured to perform signal enhancement on the sixth information according to a preset WaveNet mathematical model, so as to obtain a first data sequence.

The first scheduling unit 62 is configured to dynamically schedule according to the first data sequence, according to the characteristics of heterogeneous data such as a device state and an environmental parameter, and a preset scheduling policy, to obtain a second data sequence, where the second data sequence includes a data transmission parameter, and the data transmission parameter includes a priority and a transmission rate of data.

And the second enhancing unit 63 is configured to enhance the redundant information according to the second data sequence, and perform redundant transmission on the data enhanced by the redundant information to obtain a third data sequence.

The first decoding unit 64 is configured to decode and remove redundant information by using the deep learning mathematical model according to the third data sequence to recover the original heterogeneous data, and obtain the recovered original heterogeneous data after decoding.

Example 3:

fig. 3 is a block diagram illustrating an internet of things heterogeneous data cascade transmission device 800 according to an exemplary embodiment. As shown in fig. 3, the internet of things heterogeneous data cascade transmission device 800 may include: a processor 801, a memory 802. The internet of things heterogeneous data cascade transmission device 800 can also include one or more of a multimedia component 803, an I/O interface 804, and a communication component 805.

The processor 801 is configured to control overall operation of the internet of things heterogeneous data cascade transmission device 800, so as to complete all or part of the steps in the internet of things heterogeneous data cascade transmission method. The memory 802 is used to store various types of data to support the operation of the internet of things heterogeneous data cascade transmission device 800, which may include, for example, instructions for any application or method operating on the internet of things heterogeneous data cascade transmission device 800, as well as application related data, such as contact data, transceived messages, pictures, audio, video, and the like. The Memory 802 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM for short), electrically erasable programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), erasable programmable Read-Only Memory (Erasable Programmable Read-Only Memory, EPROM for short), programmable Read-Only Memory (Programmable Read-Only Memory, PROM for short), read-Only Memory (ROM for short), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen, the audio component being for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, which may be a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is configured to perform wired or wireless communication between the heterogeneous data cascade transmission device 800 of the internet of things and other devices. Wireless communication, such as Wi-Fi, bluetooth, near field communication (Near FieldCommunication, NFC for short), 2G, 3G or 4G, or a combination of one or more thereof, the respective communication component 805 may thus comprise: wi-Fi module, bluetooth module, NFC module.

In an exemplary embodiment, the internet of things heterogeneous data cascade transmission device 800 may be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASIC), digital signal processors (DigitalSignal Processor, abbreviated as DSP), digital signal processing devices (Digital Signal Processing Device, abbreviated as DSPD), programmable logic devices (Programmable Logic Device, abbreviated as PLD), field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGA), controllers, microcontrollers, microprocessors, or other electronic components for performing the internet of things heterogeneous data cascade transmission method described above.

In another exemplary embodiment, a computer readable storage medium is also provided, which includes program instructions that, when executed by a processor, implement the steps of the internet of things heterogeneous data cascade transmission method described above. For example, the computer readable storage medium may be the memory 802 including the program instructions described above, which are executable by the processor 801 of the internet of things heterogeneous data cascade transmission device 800 to perform the internet of things heterogeneous data cascade transmission method described above.

Example 4:

Corresponding to the above method embodiment, a readable storage medium is further provided in this embodiment, and a readable storage medium described below and an internet of things heterogeneous data cascade transmission method described above may be referred to correspondingly.

A readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method for cascade transmission of heterogeneous data of the internet of things according to the above method embodiment.

The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, and the like.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims

1. The heterogeneous data cascade transmission method of the Internet of things is characterized by comprising the following steps of:

transmitting the sixth information, and performing layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, wherein the seventh information is original heterogeneous data restored after decoding;

Transmitting the sixth information, and performing layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, where the method includes:

carrying out signal enhancement on the sixth information according to a preset WaveNet mathematical model to obtain a first data sequence;

according to the first data sequence, carrying out dynamic scheduling according to the equipment state, the characteristic of environment parameter heterogeneous data and a preset scheduling strategy to obtain a second data sequence, wherein the second data sequence comprises data transmission parameters, and the data transmission parameters comprise the priority and the transmission rate of the data;

performing redundancy information enhancement according to the second data sequence, and performing redundancy transmission on the data with the enhanced redundancy information to obtain a third data sequence;

and according to the third data sequence, decoding by using the deep learning mathematical model and removing redundant information to recover the original heterogeneous data, so as to obtain the original heterogeneous data recovered after decoding.

2. The internet of things heterogeneous data cascade transmission method of claim 1, wherein clustering the data condition of each type of internet of things equipment in the second information to obtain third information comprises:

According to the second information, carrying out data preprocessing on historical heterogeneous data of each type of internet of things equipment to obtain preprocessed data, wherein the preprocessing comprises data cleaning, data standardization and data missing value processing;

performing feature extraction and natural language processing on the preprocessed data according to a preset self-encoder mathematical model to obtain a feature vector set, wherein the feature vector set comprises equipment state features, environment parameter features, time features and equipment log features, the equipment state features comprise the running state of equipment, the use frequency of the equipment and the fault frequency of the equipment, the environment parameter features comprise the average value, the maximum value, the minimum value and the fluctuation range of temperature, humidity and air pressure parameters of the environment where the equipment is located, the time features comprise the time for generating the equipment data, the use time of the equipment and the use period of the equipment, and the equipment log features comprise keywords, topics and emotion features;

carrying out cluster analysis on each type of internet of things equipment in the feature vector set according to a preset spectral clustering mathematical model to obtain a clustering result, wherein the clustering result comprises clustering labels, characteristic description and statistical information of equipment states of the internet of things equipment;

And calculating an average time window of the data of the Internet of things equipment in each cluster according to the clustering result.

3. The method for cascade transmission of heterogeneous data of internet of things according to claim 1, wherein the data segmentation processing is performed on the corresponding type of the device data of internet of things in the first information according to the time window information in the third information to obtain fourth information, and the method comprises the following steps:

identifying each type of the Internet of things equipment and corresponding time window information thereof according to the first information and the third information, and performing data distribution processing to obtain an equipment data set, wherein the equipment data set comprises all equipment data with the same equipment type label;

sorting the equipment data sets according to the time window information in the third information to obtain ordered equipment data;

cutting the equipment data according to the time window information to obtain data blocks, wherein each data block comprises equipment data in a continuous time;

and carrying out data integration processing according to the data block and the equipment type to obtain fourth information.

4. The internet of things heterogeneous data cascade transmission method of claim 1, wherein performing time sequence analysis processing according to the fourth information and the second information to obtain fifth information comprises:

Performing outlier detection on the second information according to a preset isolated forest mathematical model to obtain normal data;

carrying out trending treatment according to the normal data, and eliminating seasonality and trend of the data through difference and moving average calculation to obtain trending data;

performing time sequence model fitting processing on the trending data according to a preset state space mathematical model to obtain a model fitting result for describing the change rule of the equipment data;

and carrying out data prediction processing on the fourth information according to the model fitting result to obtain a data trend prediction result, wherein the data trend prediction result comprises the state change condition of the equipment in the future preset time.

5. Heterogeneous data cascade transmission device of thing networking, characterized by, include:

the decoding module is used for transmitting the sixth information, and carrying out layer-by-layer decoding processing on the transmitted data sequence according to the deep learning mathematical model to obtain seventh information, wherein the seventh information is original heterogeneous data restored after decoding;

wherein the decoding module comprises:

the first enhancement unit is used for carrying out signal enhancement on the sixth information according to a preset WaveNet mathematical model to obtain a first data sequence;

the first scheduling unit is used for carrying out dynamic scheduling according to the first data sequence, the equipment state, the characteristic of environment parameter heterogeneous data and a preset scheduling strategy to obtain a second data sequence, wherein the second data sequence comprises data transmission parameters, and the data transmission parameters comprise the priority and the transmission rate of data;

The second enhancement unit is used for enhancing the redundant information according to the second data sequence and carrying out redundant transmission on the data enhanced by the redundant information to obtain a third data sequence;

and the first decoding unit is used for decoding and removing redundant information by utilizing the deep learning mathematical model according to the third data sequence to recover the original heterogeneous data, so as to obtain the recovered original heterogeneous data after decoding.

6. The internet of things heterogeneous data cascade transmission device of claim 5, wherein the clustering module comprises:

the first processing unit is used for carrying out data preprocessing on the historical heterogeneous data of each type of internet of things equipment according to the second information to obtain preprocessed data, wherein the preprocessing comprises data cleaning, data standardization and data missing value processing;

the first extraction unit is used for carrying out feature extraction and natural language processing on the preprocessed data according to a preset self-encoder mathematical model to obtain a feature vector set, wherein the feature vector set comprises equipment state features, environment parameter features, time features and equipment log features, the equipment state features comprise the running state of equipment, the use frequency of the equipment and the fault frequency of the equipment, the environment parameter features comprise the average value, the maximum value, the minimum value and the fluctuation range of the temperature, the humidity and the air pressure parameters of the environment where the equipment is located, the time features comprise the time for generating equipment data, the use time of the equipment and the use period of the equipment, and the equipment log features comprise keywords, topics and emotion features;

The first clustering unit is used for carrying out clustering analysis on each type of internet of things equipment in the feature vector set according to a preset spectral clustering mathematical model to obtain a clustering result, wherein the clustering result comprises clustering labels, characteristic description and statistical information of equipment states of the internet of things equipment;

and the first calculation unit is used for calculating the average time window of the data of the Internet of things equipment in each cluster according to the clustering result.

7. The internet of things heterogeneous data cascade transmission device of claim 5, wherein the segmentation module comprises:

the first distribution unit is used for identifying each type of the Internet of things equipment and corresponding time window information thereof according to the first information and the third information, and carrying out data distribution processing to obtain an equipment data set, wherein the equipment data set comprises all equipment data with the same equipment type label;

the first ordering unit is used for ordering the equipment data sets according to the time window information in the third information to obtain ordered equipment data;

the first segmentation unit is used for cutting the equipment data according to the time window information to obtain data blocks, wherein each data block comprises equipment data in a continuous time;

And the first integration unit is used for carrying out data integration processing according to the data block and the equipment type to obtain fourth information.

8. The internet of things heterogeneous data cascade transmission device of claim 5, wherein the analysis module comprises:

the first detection unit is used for detecting abnormal values of the second information according to a preset isolated forest mathematical model to obtain normal data;

the second processing unit is used for carrying out trending processing according to the normal data, and removing seasonality and trend of the data through difference and moving average calculation to obtain trending data;

the first fitting unit is used for performing time sequence model fitting processing on the trending data according to a preset state space mathematical model to obtain a model fitting result for describing the change rule of the equipment data;

and the third processing unit is used for carrying out data prediction processing on the fourth information according to the model fitting result to obtain a data trend prediction result, wherein the data trend prediction result comprises the state change condition of the equipment in the future preset time.

9. The heterogeneous data cascade transmission equipment of thing networking, characterized by, include:

A memory for storing a computer program;

a processor for implementing the steps of the internet of things heterogeneous data cascade transmission method according to any one of claims 1 to 4 when executing the computer program.

10. A medium, characterized by: the medium has stored thereon a computer program which, when executed by a processor, implements the steps of the internet of things heterogeneous data cascade transmission method according to any of claims 1 to 4.