Lightweight network data encryption transmission method and system facing edge device
By using data preprocessing and lightweight encryption methods for edge devices, the problems of low compression rate and poor security in data transmission of edge devices are solved, achieving efficient and secure data transmission that is adapted to the computing and power consumption characteristics of edge devices.
Patent Information
- Application Number
- CN202511802801.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-03
- Publication Date
- 2026-02-13
AI Technical Summary
Edge devices suffer from low compression rates, large data recovery errors, and a lack of side-channel attack protection in their encryption schemes during data transmission, leading to bandwidth waste and security risks. They are also difficult to adapt to the characteristics of weak computing power, limited storage, and power sensitivity.
By collecting raw data, preprocessing and noise filtering are performed to obtain geometric center parameters and numerical distribution direction. Reference data range is defined, verification samples are selected to calculate data correction values, lossless compression and lightweight encryption are performed, and finally the data is sent through a wireless communication protocol.
It improves data compression rate, reduces bandwidth waste, ensures data recovery accuracy, enhances transmission security, adapts to edge device resource constraints, and improves overall reliability and security.
Smart Images
Figure CN121531373A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of data processing, in particular to a lightweight network data encryption transmission method and system for edge devices. BACKGROUND
[0002] With the deep integration of Internet of Things and edge computing, the number of edge devices deployed in industrial Internet of Things, intelligent property and other scenarios has increased significantly. The devices generally have the characteristics of weak computing ability, limited storage resources and sensitivity to power consumption. During the transmission of the collected physical environment monitoring data, the requirements of security, real-time performance and low resource consumption need to be met simultaneously.
[0003] For example, a plurality of edge vibration sensors based on low-cost chips are deployed on a plurality of production lines to collect vibration acceleration data of motor operation at a fixed frequency. A scheme using general lossless compression, conventional lightweight encryption and common Internet of Things protocol transmission is used to send data to a workshop edge gateway for realizing motor fault warning function. However, the scheme has technical defects. The compression process does not combine the distribution characteristics of the data itself, but only relies on general algorithms to process the original data. For vibration data with strong time series and relatively concentrated values, the compression rate is low, which still causes certain bandwidth waste. A dynamic data correction mechanism is not established before encryption. When the data distribution is shifted due to changes in motor load, errors will occur after the compressed data is decrypted and restored, which exceeds the accuracy requirements of the device warning. Moreover, the encryption scheme does not integrate side channel attack protection capability. In a strong electromagnetic interference environment in the workshop, the key is easily leaked by analyzing the power consumption change when the device is running, which has the risk of data tampering or theft. SUMMARY
[0004] The technical problem to be solved by the present application is to provide a lightweight network data encryption transmission method and system for edge devices to realize the dynamic balance of security, efficiency and resource consumption.
[0005] To solve the above technical problems, the technical solutions of the present application are as follows: In a first aspect, a lightweight network data encryption transmission method for edge devices, the method comprising: Collecting original monitoring data of a physical environment to obtain an original data set arranged in time sequence; Defining the data range according to the original data set to obtain a geometric center parameter and a numerical distribution direction of the data space; Determining the geometric center parameter as a reference data point; based on the numerical distribution direction of the data space, dividing the data space into at least two non-overlapping numerical intervals to determine each numerical interval as a reference data range; Based on the reference data range, verification data samples located in different numerical intervals are selected; a data correction value is calculated based on the numerical relationship between each verification data sample and the benchmark data point. The original data set is combined with the data correction value to obtain the combined data; the combined data is then subjected to lossless compression to obtain a compressed data packet with a reduced size. The compressed data packet is encrypted to obtain an encrypted data stream; the encrypted data stream is then sent to the edge gateway via a wireless communication protocol.
[0006] Furthermore, raw monitoring data of the physical environment are collected to obtain a raw dataset arranged in time series, including: The raw monitoring data stream containing invalid and noisy data is collected; the raw monitoring data stream is preprocessed to remove invalid data and filter noisy data, thereby obtaining a set of purified valid monitoring data. The purified and valid monitoring data is used as input, and sorting and caching operations are performed by timestamp to finally obtain the original data set arranged in time series.
[0007] Furthermore, based on the original dataset, the data range is defined to obtain the geometric center parameter and the numerical distribution direction of the data space, including: The original dataset arranged in time series is used as input data; based on the input data, the geometric center parameter is obtained by processing it through a central statistical algorithm to characterize the central position of the entire dataset in the numerical space. Using the geometric center parameter as the reference point for distribution analysis, statistical analysis is performed on the distribution characteristics of the input data around the reference point to determine the main numerical distribution direction of the obtained data space.
[0008] Furthermore, the geometric center parameter is determined as the reference data point; based on the numerical distribution direction of the data space, the data space is divided into at least two non-overlapping numerical intervals, so that each numerical interval is determined as a reference data range, including: Receive the geometric center parameters and the numerical distribution direction of the data space; directly set the geometric center parameters as the reference data points for data processing; Based on the numerical distribution direction of the data space, numerical space partitioning parameters are obtained to define different data regions; By using numerical space partitioning parameters, the data space is divided into at least two non-overlapping numerical intervals, and each of the partitioned numerical intervals is ultimately determined as the reference data range.
[0009] Furthermore, based on the reference data range, verification data samples located in different numerical intervals are selected; a data correction value is calculated based on the numerical relationship between each verification data sample and the benchmark data point, including: Receive the output reference data ranges; based on the interval boundaries of each reference data range, select representative verification data samples from each reference data range; Each selected validation data sample is compared with the baseline data point, and the numerical difference between each validation data sample and the baseline data point is calculated. Based on the calculated differences between the various values, a unified data correction value is generated through a pre-defined statistical algorithm.
[0010] Furthermore, the original data set is combined with the data correction value to obtain combined data; the combined data is then subjected to lossless compression to obtain a compressed data packet with a reduced size, including: Based on the original data set and the data correction value, a data combination operation is performed, and the data correction value is used as metadata to be concatenated with the original data set to obtain a combined data block; Lossless compression is performed on the combined data blocks, and a preset lossless compression algorithm is applied to process them to obtain compressed data packets with reduced size.
[0011] Further, the compressed data packet is encrypted to obtain an encrypted data stream; the encrypted data stream is then sent to the edge gateway via a wireless communication protocol, including: Encryption is performed on the compressed data packet using a preset lightweight encryption algorithm to obtain an encrypted data stream; Perform wireless transmission operations on the encrypted data stream, and send the encrypted data stream to the edge gateway through a wireless communication protocol.
[0012] Secondly, a lightweight network data encryption transmission system for edge devices includes: The acquisition module is used to collect raw monitoring data of the physical environment and obtain a raw data set arranged in time series. The analysis module is used to define the data range based on the original data set, and obtain the geometric center parameters and the numerical distribution direction of the data space; The calculation module is used to determine the geometric center parameter as the reference data point; divide the data space into at least two non-overlapping numerical intervals based on the numerical distribution direction of the data space, so as to determine each numerical interval as the reference data range; select verification data samples located in different numerical intervals based on the reference data range; and calculate a data correction value according to the numerical relationship between each verification data sample and the reference data point. A compression module is used to combine the original data set with the data correction value to obtain the combined data; and to perform lossless compression processing on the combined data to obtain a compressed data packet with a reduced size. The processing module is used to encrypt the compressed data packet to obtain an encrypted data stream; and to send the encrypted data stream to the edge gateway via a wireless communication protocol.
[0013] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.
[0014] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.
[0015] The above-described solution of the present invention has at least the following beneficial effects: This approach overcomes the technical challenges of traditional solutions. It involves preprocessing and sorting the raw monitoring data, obtaining geometric center parameters and numerical distribution directions through central statistical algorithms and distribution characteristic analysis, defining a reference data range, selecting verification samples to calculate and fuse data correction values, combining the correction values with the original data for lossless compression, and finally processing the data using a lightweight encryption algorithm and transmitting it to the edge gateway via a wireless communication protocol. These challenges include low compression rates due to the lack of data distribution characteristics in compression, excessive data recovery errors due to the absence of dynamic correction mechanisms, lack of side-channel attack protection for encryption leading to key leakage, and difficulty in adapting to the limited computing power, storage capacity, and power sensitivity of edge devices. The solution achieves a balance between low resource consumption requirements of edge devices, improved data compression rates to reduce bandwidth waste, dynamic correction to ensure data recovery accuracy meets monitoring and early warning requirements, enhanced data transmission security to avoid key leakage and data tampering / theft risks, and a comprehensive improvement in the overall reliability and adaptability of edge device data transmission. Attached Figure Description
[0016] Figure 1 This is a flowchart illustrating a lightweight network data encryption transmission method for edge devices provided in an embodiment of the present invention.
[0017] Figure 2 This is a schematic diagram of a lightweight network data encryption transmission system for edge devices provided by an embodiment of the present invention. Detailed Implementation
[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0019] like Figure 1 As shown, embodiments of the present invention propose a lightweight network data encryption transmission method for edge devices, the method comprising the following steps: Step 1: Collect raw monitoring data of the physical environment to obtain a raw data set arranged in time series; Step 2: Based on the original dataset, define the data range to obtain the geometric center parameter and the numerical distribution direction of the data space; Step 3: Determine the geometric center parameter as the reference data point; based on the numerical distribution direction of the data space, divide the data space into at least two non-overlapping numerical intervals, so as to determine each numerical interval as the reference data range; Step 4: Based on the reference data range, select verification data samples located in different numerical intervals; calculate a data correction value based on the numerical relationship between each verification data sample and the benchmark data point; Step 5: Combine the original data set with the data correction value to obtain the combined data; perform lossless compression on the combined data to obtain a compressed data packet with a reduced size. Step 6: Encrypt the compressed data packet to obtain an encrypted data stream; send the encrypted data stream to the edge gateway via a wireless communication protocol.
[0020] In this embodiment of the invention, a complete technical approach is adopted. This approach involves collecting raw monitoring data in a time-series manner, obtaining geometric center parameters and numerical distribution directions by defining the data range, dividing the reference data range accordingly, selecting verification samples to calculate data correction values, combining the raw data with the correction values for lossless compression, and finally encrypting the data before transmitting it to the edge gateway via a wireless communication protocol. This overcomes the technical problems of traditional edge device data transmission schemes, such as low compression rates due to compression not incorporating data characteristics, excessive data recovery errors due to a lack of dynamic correction mechanisms, insufficient encryption protection leading to security risks, and difficulty in adapting to the weak computing power, limited storage, and power-sensitive characteristics of edge devices. Therefore, this approach achieves the technical effect of improving data compression efficiency to reduce bandwidth waste while meeting the low resource consumption requirements of edge devices, ensuring data recovery accuracy meets monitoring requirements, enhancing data transmission security to avoid tampering or theft risks, and comprehensively balancing data transmission security, real-time performance, and low resource consumption, thereby improving the overall reliability of edge device data transmission.
[0021] In a preferred embodiment of the present invention, step 1 above may include: Step 1.1: Collect the raw monitoring data stream containing invalid and noisy data. Preprocess the raw monitoring data stream to remove invalid data and filter noisy data, resulting in a set of purified valid monitoring data. Specifically, this includes: edge vibration sensors based on low-cost chips deployed on multiple production lines in the smart factory continuously collecting vibration acceleration data of the motors at a preset fixed frequency, forming the raw monitoring data stream. Due to electromagnetic interference, temporary sensor contact problems, or brief malfunctions in the workshop environment, the raw monitoring data stream will naturally contain invalid and noisy data. Invalid data includes meaningless values generated when sensors malfunction, while noise data comes from electromagnetic radiation interference from other equipment in the workshop.
[0022] For this raw monitoring data stream, the invalid data removal operation is first performed. Based on the reasonable value range of vibration acceleration during normal operation of the corresponding model motor, combined with the pre-set factory motor process parameters, each data point in the raw monitoring data stream is judged one by one, and meaningless data that exceeds the reasonable value range is removed from the data stream. Next, the noise data filtering operation is performed. Considering the characteristics of weak computing power and limited storage resources of edge devices, a lightweight sliding window mean filtering method is adopted. A fixed-size sliding window is set, and the average value of the data in each window is used to replace the original data in the center of the window. This smooths out data fluctuations, filters out noise caused by electromagnetic interference, and finally obtains a set of purified monitoring data with invalid information and noise interference removed.
[0023] Step 1.2 involves using the purified valid monitoring data as input to perform timestamp sorting and caching operations, ultimately obtaining a raw data set arranged in time sequence. Specifically, this includes: using the purified valid monitoring data as the processing object, firstly extracting the timestamp information corresponding to the acquisition time from each valid monitoring data. The timestamp is automatically recorded by the edge vibration sensor while acquiring data, accurately reflecting the acquisition time of the data; then, sorting all valid monitoring data sequentially according to the acquisition time recorded by the timestamp, ensuring that the sorted valid monitoring data completely presents the law of change of motor vibration state over time; after sorting, storing the valid monitoring data in the edge device to avoid data loss or disorder, ultimately forming a raw data set arranged in time sequence.
[0024] In this embodiment of the invention, by first acquiring the original monitoring data stream containing invalid and noisy data, then performing preprocessing operations such as invalid data removal and noise data filtering, and finally sorting and caching the purified monitoring data by timestamp, the technical problems of interference from invalid and noisy data in the original monitoring data and low analysis accuracy caused by disordered data are overcome. This achieves the technical effect of providing high-quality, time-series input data for data distribution feature analysis, data correction value calculation, and other links, ensuring the reliability of the entire encrypted transmission method.
[0025] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1: Using the original dataset arranged in time series as input data; based on the input data, a central statistical algorithm is used to obtain a geometric center parameter that characterizes the central position of the entire dataset in the numerical space. Specifically, this includes: receiving the original monitoring data set arranged in time sequence, which consists of physical environment data such as vibration acceleration collected by edge devices; for the time series dataset, a central statistical algorithm suitable for low resource consumption of edge devices is used to extract the monitoring values at each time point in sequence, and by calculating the mean or median of all values, a geometric center parameter that can represent the central position of the entire dataset in the numerical space is obtained.
[0026] Step 2.2: Using the geometric center parameter as the benchmark for distribution analysis, perform statistical analysis on the distribution characteristics of the input data around the benchmark to determine the main numerical distribution direction of the obtained data space. Specifically, this includes: using the obtained geometric center parameter as the analysis benchmark, calculating the numerical difference between each data point in the original data set and the benchmark; performing statistical analysis on the difference, observing the distribution of data points around the benchmark, such as counting the number of data points with positive and negative differences, or analyzing the distribution trend of the absolute value of the difference, thereby determining the main direction of the data's concentrated distribution, that is, the main trend direction of the overall numerical values in the data space tending to increase or decrease.
[0027] In this embodiment of the invention, because a technical means is adopted that takes the original data set arranged in time series as input, obtains the geometric center parameter through a central statistical algorithm, and uses the geometric center parameter as a reference point to perform statistical analysis on the distribution characteristics of the data around the reference point, it overcomes the technical problems in traditional data processing where the data spatial center position is ambiguous and the main numerical distribution direction is unclear, resulting in a lack of basis for subsequent data interval division and insufficient processing targeting. Thus, it achieves the accurate positioning of the core position and main distribution trend of the data set in the numerical space, providing reliable support for the reasonable division of subsequent reference data range, the scientific selection of verification data samples, and the accurate calculation of data correction values, thereby improving the targeting and effectiveness of the data preprocessing stage of edge devices.
[0028] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: Receive the geometric center parameters and the numerical distribution direction of the data space; directly set the geometric center parameters as the reference data point for data processing, specifically including: receiving the obtained geometric center parameters and the numerical distribution direction of the data space, the parameters being derived from the analysis of time-series data such as motor vibration acceleration collected from edge devices; directly determining the geometric center parameters as the reference data point for data processing.
[0029] Step 3.2: Based on the numerical distribution direction of the data space, obtain the numerical space partitioning parameters used to define different data regions. Specifically, based on the obtained numerical distribution direction of the data space, that is, the main distribution trend of data such as vibration acceleration around the geometric center parameter, such as the overall bias towards the direction of numerical increase or decrease, through light statistical analysis, combined with the distribution density and range of data in this direction, determine the numerical space partitioning parameters used to divide different data regions. The parameters can be specific numerical boundaries.
[0030] Step 3.3: Divide the data space into at least two non-overlapping numerical intervals using numerical space partitioning parameters, and ultimately determine each of the partitioned numerical intervals as the reference data range. Specifically, this includes: using the obtained numerical space partitioning parameters, determining specific interval boundaries in the data space, and dividing the entire data space into at least two non-overlapping numerical intervals. For example, for vibration acceleration data, the data space can be divided into intervals on both sides of the reference point according to the partitioning parameters, or into core intervals and peripheral intervals according to the distribution density, etc. After the partitioning is completed, these intervals are defined as the reference data range for subsequent selection of verification data samples.
[0031] In this embodiment of the invention, because a technical means is adopted to set the geometric center parameter as the reference data point, determine the numerical space partitioning parameter based on the numerical distribution direction of the data space, and then divide at least two non-overlapping numerical intervals as reference data ranges, the technical problems in traditional data processing caused by the lack of interval partitioning methods based on data centers and distribution characteristics, resulting in ambiguous data range definition, no clear reference for the selection of verification samples, and difficulty in adapting to data distribution characteristics, are overcome. Thus, the technical effect of making the reference data range accurately match the actual data distribution characteristics is achieved, improving the pertinence and effectiveness of data processing, and helping to optimize compression efficiency and reduce data recovery errors is achieved.
[0032] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1: Receive the output reference data ranges; based on the interval boundaries of each reference data range, select representative verification data samples from each reference data range. Specifically, this includes: receiving the divided reference data ranges, where each reference data range is a non-overlapping numerical interval that matches the distribution characteristics of time-series data such as vibration acceleration collected by the edge device; determining the data value span based on the interval boundaries of each reference data range; and, considering the temporal and numerical concentration characteristics of the data, uniformly selecting multiple data points from each interval as verification data samples to ensure that the samples can represent the numerical distribution of the corresponding interval. The selection process employs low-complexity operations.
[0033] Step 4.2 compares each selected verification data sample with the benchmark data point, and calculates the numerical difference between each verification data sample and the benchmark data point. Specifically, this includes: taking out all selected verification data samples, comparing each sample with the set benchmark data point one by one; obtaining the specific difference between each verification data sample and the benchmark data point through simple numerical calculations, clearly showing the deviation of samples in different intervals relative to the benchmark point. The calculation process is simplified to adapt to the weak computing power of edge devices, while accurately recording the numerical difference relationship corresponding to each sample.
[0034] Step 4.3: Based on the calculated numerical differences, a unified data correction value is generated by fusing the results using a preset statistical algorithm. Specifically, this includes: collecting all the calculated numerical differences, calling a preset lightweight statistical algorithm to fuse these differences, and the statistical algorithm fully considering the low power consumption and low resource overhead requirements of edge devices. The algorithm has a simple and efficient operation logic. By comprehensively analyzing and integrating the differences between samples in different intervals, a unified data correction value that dynamically adapts to the data distribution characteristics is finally generated.
[0035] In this embodiment of the invention, because a technical means is adopted to select representative verification data samples from each reference data range, calculate the numerical difference relationship between the samples and the benchmark data points, and then fuse the differences to generate a unified data correction value through a preset statistical algorithm, the technical problem in the traditional scheme that the error exceeds the warning accuracy requirement after the compressed data is decrypted and restored when the data distribution shifts due to the lack of a dynamic correction mechanism based on data distribution characteristics is overcome. Thus, the technical effect of providing a dynamically adapted data correction basis for data transmission is achieved, effectively reducing data recovery errors to meet the accuracy requirements of edge device monitoring data, and improving the overall data transmission reliability is achieved.
[0036] In a preferred embodiment of the present invention, step 5 above may include: Step 5.1: Based on the original data set and the data correction value, perform a data combination operation. The data correction value is used as metadata and concatenated with the original data set to obtain a combined data block. Specifically, this includes: identifying two types of core data to be processed: first, the original data set after preprocessing and time-series sorting, which contains valid motor vibration acceleration data arranged chronologically, with each data point reflecting the motor's vibration state at a specific moment; second, the calculated single data correction value, generated based on the distribution characteristics of the motor vibration acceleration data, which can dynamically adapt to data distribution shifts caused by changes in motor load. Due to the limited computing power and storage resources of edge devices, to avoid complex... To avoid excessive resource consumption from complex computations, edge devices employ a lightweight sequential concatenation logic to perform data combination operations. First, the data correction value is placed at the beginning of the data sequence as metadata. Then, each motor vibration acceleration data point arranged in chronological order in the original data set is sequentially appended to the metadata according to the order of acquisition time. This concatenation method does not require additional complex data format conversions; it can be completed simply through sequential integration, thus controlling the computational overhead and power consumption of the edge devices. The resulting combined data block has a clear structure and coherent logic. The arrangement of metadata first and original vibration acceleration data last ensures that the edge gateway can quickly and preferentially extract the data correction value after receiving and decrypting the data block.
[0037] Step 5.2: Perform lossless compression on the combined data block. Apply a preset lossless compression algorithm to process the data, resulting in a compressed data packet with a reduced size. Specifically, this includes: performing a preset lightweight lossless compression operation on the formed combined data block. The selected lossless compression algorithm is the improved LZW algorithm. Compared with traditional general lossless compression algorithms, this algorithm is more adaptable to the resource constraints of edge devices and can improve compression efficiency by combining the distribution characteristics of motor vibration acceleration data. Before the compression operation begins, the edge device will initialize and optimize the dictionary of the improved LZW algorithm based on the distribution characteristics of the motor vibration acceleration data obtained from the previous analysis, such as the values being concentrated in a specific range and having a certain continuity in time sequence. The vibration acceleration data will be pre-added to the algorithm dictionary. For frequently occurring numerical combinations, such as continuous numerical sequences of motor vibration acceleration under a certain load, there is no need to rebuild a large number of dictionary entries for basic numerical combinations during the compression process. This reduces the computational load during compression and avoids excessive power consumption on edge devices due to complex dictionary construction. During compression, the improved LZW algorithm encodes the combined data blocks segment by segment: for metadata, since it is a single value, it is directly encoded using the corresponding basic entry in the dictionary; for the original vibration acceleration data, the algorithm identifies repeated numerical combinations in the data sequence and replaces the repeated sequences with the corresponding indices already in the dictionary, thereby reducing the number of bits stored in the data. After compression, the size of the combined data blocks is reduced, ultimately forming a compressed data packet with reduced volume.
[0038] In this embodiment of the invention, because a technical means is adopted to combine the data correction value as metadata with the original data set to form a combined data block, and then apply a preset lossless compression algorithm to the combined data block to perform compression operation, the technical problems of low compression ratio caused by the lack of dynamic data correction information in traditional lossless compression, lack of accurate correction basis during data recovery, and difficulty in adapting to the low resource overhead requirements of edge devices are overcome. Thus, the technical effect of improving the data compression ratio to reduce transmission bandwidth waste is achieved, while the compression process adapts to the computing and storage resource constraints of edge devices.
[0039] In a preferred embodiment of the present invention, step 6 above may include: Step 6.1: Perform encryption on the compressed data packet. Process it using a preset lightweight encryption algorithm to obtain an encrypted data stream. Specifically, this includes: First, determining that the preset lightweight encryption algorithm is an improved AES-128 encryption algorithm. The algorithm is optimized for the characteristics of edge devices with weak computing power and limited storage resources. On the one hand, it simplifies the round function calculation logic of traditional AES-128, reduces the number of multiplication operations, and reduces resource consumption during the calculation process. On the other hand, it integrates a side-channel attack protection design. By adding power smoothing processing during the encryption operation, it avoids the characteristic differences caused by the power consumption fluctuations of the device during different data operations, and prevents attackers from leaking keys by analyzing the device power consumption changes in the strong electromagnetic environment of the workshop. When performing the encryption operation, the edge device first retrieves the generated compressed data packet from local storage. This data packet contains the spliced metadata and the compressed original data of motor vibration acceleration. Subsequently, the edge device initializes the encryption algorithm parameters: it loads the initial key pre-stored locally and sets the data block length according to the size of the compressed data packet. The block length is controlled within 128 bytes, which not only meets the requirements of block encryption of improved AES-128, but also avoids excessive memory usage of the edge device due to excessive data processing volume in a single process.
[0040] The compressed data packet is divided into multiple data blocks according to the set block length. An improved AES-128 encryption operation is performed on each data block in sequence. First, the data block is padded to ensure that the data block length meets the algorithm requirements. Then, multiple rounds of iterative encryption are completed through an optimized round function. During each round of iteration, power consumption is smoothed to keep the device power consumption stable and eliminate the correlation between power consumption and data content. Finally, the XOR operation and permutation of each round of encryption results are performed to generate encrypted data blocks. After all data blocks are encrypted, the edge device splices the encrypted data blocks in their original order and adds an encryption identifier and data length information to the header of the spliced data stream to finally form a complete encrypted data stream.
[0041] Step 6.2: Perform wireless transmission of the encrypted data stream. Send the encrypted data stream to the edge gateway via a wireless communication protocol. Specifically, this includes: the preset wireless communication protocol is Bluetooth LE5.3, which features low power consumption, low bandwidth usage, and strong anti-electromagnetic interference capabilities. It is compatible with the battery-powered mode of the edge vibration sensor and can maintain stable communication in the complex electromagnetic environment where multiple devices work simultaneously in the workshop, meeting the requirement of real-time transmission of motor vibration data to the edge gateway. Before performing the wireless transmission operation, the edge device first completes the transmission parameter preparation: extract the length information of the encrypted data stream, combine it with the maximum transmission size of Bluetooth LE5.3, and divide the encrypted data stream into multiple transmission frames; at the same time, read the device's own unique identifier and the timestamp of the current data acquisition period, and add the information to the frame header of each transmission frame; scan the preset edge gateway Bluetooth signals in the workshop, and establish a low-power Bluetooth connection through the pre-configured pairing information. After the connection is established, the edge device sends data to the edge gateway sequentially according to the transmission frame order. After sending each frame, it waits for an acknowledgment frame from the gateway. The acknowledgment frame contains the verification result of the gateway's received data. If the verification passes, the next frame is sent; if the verification fails, the current frame is resent to avoid data loss or damage due to electromagnetic interference in the workshop. After all transmission frames have been sent, the edge device waits for a data reception completion signal from the gateway. After confirming that the signal reception is correct, it actively disconnects the Bluetooth connection to reduce power consumption. Throughout the transmission process, the frequency hopping technology of the Bluetooth LE5.3 protocol can effectively avoid electromagnetic interference from other devices in the workshop, ensuring stable transmission of encrypted data streams.
[0042] In this embodiment of the invention, by employing a technique of encrypting compressed data packets using a preset lightweight encryption algorithm to obtain an encrypted data stream, and then transmitting the encrypted data stream wirelessly to the edge gateway via a wireless communication protocol, the technical problems of traditional solutions—such as lack of side-channel attack protection, easy leakage of keys through power consumption analysis in strong electromagnetic interference environments, leading to data tampering or theft risks, and the encryption and transmission process not being adapted to the weak computing power and power-sensitive characteristics of edge devices—are overcome. This achieves the technical effect of controlling the computing and power consumption of edge devices, adapting to their resource constraints, enhancing the anti-interference and anti-attack capabilities of data transmission, ensuring the security and integrity of data transmission from the edge device to the edge gateway, and realizing end-to-end secure data transmission.
[0043] like Figure 2 As shown, embodiments of the present invention also provide a lightweight network data encryption transmission system for edge devices, comprising: The acquisition module is used to collect raw monitoring data of the physical environment and obtain a raw data set arranged in time series. The analysis module is used to define the data range based on the original data set, and obtain the geometric center parameters and the numerical distribution direction of the data space; The calculation module is used to determine the geometric center parameter as the reference data point; divide the data space into at least two non-overlapping numerical intervals based on the numerical distribution direction of the data space, so as to determine each numerical interval as the reference data range; select verification data samples located in different numerical intervals based on the reference data range; and calculate a data correction value according to the numerical relationship between each verification data sample and the reference data point. A compression module is used to combine the original data set with the data correction value to obtain the combined data; and to perform lossless compression processing on the combined data to obtain a compressed data packet with a reduced size. The processing module is used to encrypt the compressed data packet to obtain an encrypted data stream; and to send the encrypted data stream to the edge gateway via a wireless communication protocol.
[0044] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A lightweight network data encryption transmission method for edge devices, characterized in that, The method includes: Raw monitoring data of the physical environment are collected to obtain a raw data set arranged in time series; Based on the original dataset, the data range is defined to obtain the geometric center parameter and the numerical distribution direction of the data space; The geometric center parameter is determined as the reference data point; based on the numerical distribution direction of the data space, the data space is divided into at least two non-overlapping numerical intervals, so that each numerical interval is determined as the reference data range. Based on the reference data range, verification data samples located in different numerical intervals are selected; a data correction value is calculated based on the numerical relationship between each verification data sample and the benchmark data point. The original data set is combined with the data correction value to obtain the combined data; the combined data is then subjected to lossless compression to obtain a compressed data packet with a reduced size. The compressed data packet is encrypted to obtain an encrypted data stream; the encrypted data stream is then sent to the edge gateway via a wireless communication protocol.
2. The lightweight network data encryption transmission method for edge devices according to claim 1, characterized in that, Raw monitoring data of the physical environment were collected, resulting in a raw dataset arranged in time series, including: The raw monitoring data stream containing invalid and noisy data is collected; the raw monitoring data stream is preprocessed to remove invalid data and filter noisy data, thereby obtaining a set of purified valid monitoring data. The purified and valid monitoring data is used as input, and sorting and caching operations are performed by timestamp to finally obtain the original data set arranged in time series.
3. The lightweight network data encryption transmission method for edge devices according to claim 2, characterized in that, Based on the original dataset, the data range is defined to obtain the geometric center parameter and the numerical distribution direction of the data space, including: The original dataset arranged in time series is used as input data; based on the input data, the geometric center parameter is obtained by processing it through a central statistical algorithm to characterize the central position of the entire dataset in the numerical space. Using the geometric center parameter as the reference point for distribution analysis, statistical analysis is performed on the distribution characteristics of the input data around the reference point to determine the main numerical distribution direction of the obtained data space.
4. The lightweight network data encryption transmission method for edge devices according to claim 3, characterized in that, The geometric center parameter is determined as the reference data point; based on the numerical distribution direction of the data space, the data space is divided into at least two non-overlapping numerical intervals, each of which is determined as a reference data range, including: Receive the geometric center parameters and the numerical distribution direction of the data space; directly set the geometric center parameters as the reference data points for data processing; Based on the numerical distribution direction of the data space, numerical space partitioning parameters are obtained to define different data regions; By using numerical space partitioning parameters, the data space is divided into at least two non-overlapping numerical intervals, and each of the partitioned numerical intervals is ultimately determined as the reference data range.
5. The lightweight network data encryption transmission method for edge devices according to claim 4, characterized in that, Based on the reference data range, validation data samples were selected that were located in different numerical intervals. A data correction value is calculated based on the numerical relationship between each verification data sample and the baseline data point, including: Receive the output reference data ranges; based on the interval boundaries of each reference data range, select representative verification data samples from each reference data range; Each selected validation data sample is compared with the baseline data point, and the numerical difference between each validation data sample and the baseline data point is calculated. Based on the calculated differences between the various values, a unified data correction value is generated through a pre-defined statistical algorithm.
6. The lightweight network data encryption transmission method for edge devices according to claim 5, characterized in that, The original dataset is combined with the data correction value to obtain the combined data; The combined data is subjected to lossless compression to obtain a compressed data packet with a reduced size, including: Based on the original data set and the data correction value, a data combination operation is performed, and the data correction value is used as metadata to be concatenated with the original data set to obtain a combined data block; Lossless compression is performed on the combined data blocks, and a preset lossless compression algorithm is applied to process them to obtain compressed data packets with reduced size.
7. The lightweight network data encryption transmission method for edge devices according to claim 6, characterized in that, The compressed data packet is encrypted to obtain an encrypted data stream; Sending the encrypted data stream to the edge gateway via a wireless communication protocol includes: Encryption is performed on the compressed data packet using a preset lightweight encryption algorithm to obtain an encrypted data stream; Perform wireless transmission operations on the encrypted data stream, and send the encrypted data stream to the edge gateway through a wireless communication protocol.
8. A lightweight network data encryption transmission system for edge devices, the system implementing the method as described in any one of claims 1 to 7, characterized in that, include: The acquisition module is used to collect raw monitoring data of the physical environment and obtain a raw data set arranged in time series. The analysis module is used to define the data range based on the original data set, and obtain the geometric center parameters and the numerical distribution direction of the data space; The calculation module is used to determine the geometric center parameter as the reference data point; divide the data space into at least two non-overlapping numerical intervals based on the numerical distribution direction of the data space, so as to determine each numerical interval as the reference data range; and select verification data samples located in different numerical intervals based on the reference data range. A data correction value is calculated based on the numerical relationship between each verification data sample and the baseline data point; A compression module is used to combine the original data set with the data correction value to obtain the combined data; The combined data is then subjected to lossless compression to obtain a compressed data packet with a reduced size. The processing module is used to encrypt the compressed data packet to obtain an encrypted data stream; and to send the encrypted data stream to the edge gateway via a wireless communication protocol.
9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.