Production data real-time transmission method of 5G network

Abstract

The invention discloses a production data real-time transmission method of a 5G network, and relates to the technical field of industrial Internet of Things and 5G communication. The timeliness, precision and relevance of multi-modal data are analyzed in real time by implementing adaptive aggregation and intelligent coding based on feature analysis on an edge side; according to an analysis result, small-byte state data generated periodically are intelligently packaged into an aggregation state frame in a time window, meanwhile, continuous large-flow waveform data are compressed in a segmented mode, compressed data segments are generated, metadata representing key features of the compressed data segments are extracted, and the compressed data segments are stored in a database; the state frame, the process parameter and the metadata are subjected to space-time association packaging in the packet header information, and an original and disordered mixed data stream is converted into a structured and internally associated adaptive aggregation data packet, so that the transmission load is optimized, a data foundation is laid for subsequent efficient transmission, the signaling overhead and the total data volume are greatly reduced, and the transmission efficiency is improved. And the channel utilization rate is improved.

Description

Technical Field

[0001] This invention relates to the fields of industrial Internet of Things and 5G communication technology, specifically a method for real-time transmission of production data over a 5G network. Background Technology

[0002] With the deep integration of the Industrial Internet and intelligent manufacturing technologies, it has become commonplace for industrial equipment to simultaneously generate a variety of production data with distinct characteristics during operation. This data primarily includes: equipment status signals with small data volumes but requiring extremely low latency; process parameters with moderate data volumes and specific timeliness requirements; and continuous waveform or streaming media data with massive data volumes. In the context of 5G networks enabling the Industrial Internet, the efficient and reliable real-time transmission of this heterogeneous, multimodal production data is a crucial foundation for achieving remote equipment monitoring, predictive maintenance, and process optimization.

[0003] In the existing technology, there are two main solutions for the transmission of the aforementioned multimodal data, but both have significant drawbacks: The first approach is unified encapsulation and transmission, which involves packaging all data indiscriminately and sending it through a single data stream. The disadvantage of this approach is that high-priority small-byte status information is integrated into a massive waveform data stream, resulting in a sharp increase in end-to-end latency and a lack of reliability. At the same time, the unified data packet, due to its huge size and single QoS requirement, is difficult to flexibly adapt to differentiated resources such as 5G network slicing, leading to a decline in the experience of critical services and rigidity in the utilization of network resources.

[0004] The second approach is independent parallel transmission, which establishes an independent data channel for each type of data. Frequent connection establishment for small data streams leads to high signaling overhead and low transmission efficiency. There is a lack of coordination between multiple data streams, making it impossible to utilize the spatiotemporal correlation between data for compression optimization. At the same time, managing numerous independent streams also increases system complexity and operation and maintenance costs.

[0005] Therefore, there is an urgent need for a new transmission method that can intelligently identify data characteristics, perform collaborative processing before transmission, and fully utilize the differentiated service capabilities of 5G networks, so as to fundamentally solve the problems of efficiency, reliability, and resource adaptation faced by mixed transmission of multimodal production data in industry. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a real-time transmission method for production data in 5G networks. This method enables real-time analysis of the timeliness, accuracy, and correlation of multimodal data by implementing adaptive aggregation and intelligent coding based on feature analysis at the edge. Based on the analysis results, periodically generated small-byte status data is intelligently packaged into aggregated status frames within a time window. Simultaneously, continuous high-volume waveform data is segmented and compressed to generate compressed data segments and extract metadata characterizing their key features. By spatiotemporally associating and encapsulating status frames, process parameters, and metadata in the packet header information, the original, disordered mixed data stream is transformed into structured, internally correlated adaptive aggregated data packets. This optimizes the transmission load, lays a data foundation for subsequent efficient transmission, significantly reduces signaling overhead and total data volume, and improves channel utilization.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for real-time transmission of production data in a 5G network, the specific steps of which are as follows: S100. Collect multimodal production data generated simultaneously by the same industrial equipment source at the edge computing node, and perform real-time feature analysis on the multimodal production data to identify the transmission feature requirements of each data mode. S200. Based on the transmission characteristic requirements, the multimodal production data is adaptively aggregated and intelligently encoded to generate a structured adaptive aggregated data packet containing aggregated header information and aggregated data body. S300. Based on the requirements of the differentiated service quality tags defined in the aggregate packet header information, the adaptive aggregate data packet is split into three data sub-streams with different transmission characteristics, including a high reliability sub-stream, a high bandwidth sub-stream, and a relay guarantee sub-stream. S400: Based on the transmission characteristics of each data sub-stream, dynamically map it to the corresponding network slice in the 5G network that provides different quality of service guarantees; S500: The data sub-streams are transmitted in parallel through mapped network slices, and at the receiving end, all data sub-streams are parsed, sorted and reassembled according to the aggregated packet header information to restore the complete spatiotemporally synchronized multimodal production data.

[0008] Furthermore, in S100, the multimodal production data includes: Small-byte status data with periodicity, consisting of device input and output status signals; Process parameter data consisting of temperature, pressure, and speed parameters during equipment operation; Waveform data generated by vibration sensors, acoustic sensors, or vision sensors.

[0009] Furthermore, in S100, the feature requirements include timeliness requirements, accuracy requirements, and spatiotemporal correlations between different data modalities; Edge computing nodes are industrial gateways with 5G communication capabilities or edge servers deployed locally in the factory.

[0010] Furthermore, in S200, the steps of adaptive aggregation and intelligent encoding are as follows: S210. Within a preset aggregation time window, package the small-byte state data generated in multiple cycles to generate an aggregated state frame. S220. Perform segmentation and lossless compression on continuous waveform data to generate at least one compressed data segment, and extract the key features of each compressed data segment as metadata. S230. The aggregated status frame, the metadata, and the process parameter data are spatiotemporally correlated and encapsulated to form the aggregated data body, and the aggregated header information describing the internal structure, correlation, and service quality requirements of the aggregated data body is generated.

[0011] Furthermore, in S210, the process of packaging small-byte state data generated in multiple cycles to generate an aggregated state frame employs a dynamic packaging algorithm based on timeliness weights. The working principle of this dynamic packaging algorithm is as follows: In time Upon receiving the first small byte of status data, a maximum tolerable packet delay threshold is preset. Initialize an aggregated state frame and record its initial timestamp. ; For subsequent small-byte status data originating from the same device source, calculate their timestamps. and The difference ; according to Calculate the first Packing weight of each data point ,in, For the first The amount of data in each data point This represents the historical average data volume. and Let be the harmonic coefficient, and ; Used to balance the impact of data volume Used to reflect the decay of time-related effects; When the cumulative weights and Exceeding the preset threshold, or Reaching the maximum time difference At that time, all currently cached small-byte state data are packaged into a final aggregated state frame and sent to S230.

[0012] Furthermore, in S200, the generated aggregate header information includes: a globally unique identifier for the data packet, start and end timestamps of the aggregation time window, a list of data substreams and quality of service labels, and an index table describing the position and relationship of each data segment within the aggregated data body.

[0013] Furthermore, in S300, splitting the adaptive aggregated data packet into three data sub-streams with distinct transmission characteristics specifically involves: Based on the list of service quality tags contained in the aggregated packet header information, identify the service quality level corresponding to each part of the data in the adaptive aggregated data packet; Generate the high-reliability substream: Combine the data content marked with the highest reliability requirement in the service quality label list, i.e., aggregate the status frame and metadata, to generate the first data substream; Generate the high-bandwidth substream: Combine the data content marked as having high bandwidth requirements in the Quality of Service (QoS) tag list, i.e., compressed data segments, to generate a second data substream; Generate the relay assurance sub-stream: Combine the data content marked as standard assurance requirements in the service quality label list, i.e., process parameter data, to generate a third data sub-stream.

[0014] Furthermore, in S400, the dynamic mapping employs a slice selection algorithm based on multi-factor decision-making. The working principle of this slice selection algorithm is as follows: For the first Each data sub-stream predefines a demand vector. ,in, Indicates the first The maximum latency of each data substream Indicates the first The minimum reliability required for each data substream Indicates the first The bandwidth required for each data substream; For the first Each available network slice predefines a capability vector. ,in, Represented as the first The latency that an available network slice can provide. Represented as the first The reliability that an available network slice can provide. Represented as the first The available network slices can allocate bandwidth; Calculate the first The data substream and the first Matching score of each network slice ,in, As a weighting factor, and ,function Used to evaluate network slice capability vector For data substream demand vector The degree of satisfaction; For each data substream Select Match Score The network slice with the highest value is used as the mapping target.

[0015] Furthermore, the specific steps of S500 are as follows: Analysis: Receive and parse the data sub-streams transmitted through different network slices, and extract the aggregate packet header information carried in each data sub-stream; Sorting: Based on the globally unique identifier of the data packet in the aggregation header information, all data sub-streams belonging to the same adaptive aggregation data packet are classified, and the processing sequence of each data sub-stream and its internal data segments is determined according to the start and end timestamps of the aggregation time window and the internal index order in the aggregation header information. Reassembly: Based on the index table in the aggregated packet header information, the data in the sorted high-reliability substream, high-bandwidth substream, and relay guarantee substream are aligned and spliced. The compressed data segments in the high-bandwidth substream are decompressed, and a complete data sequence containing small-byte status data, process parameter data, and waveform data, synchronized with the original acquisition time, is restored and output.

[0016] Compared with existing technologies, this method for real-time transmission of production data in a 5G network has the following advantages: I. This invention implements adaptive aggregation and intelligent coding based on feature analysis at the edge side to analyze the timeliness, accuracy, and correlation of multimodal data in real time. Based on the analysis results, periodically generated small-byte status data is intelligently packaged into aggregated status frames within a time window. At the same time, continuous high-volume waveform data is segmented and compressed to generate compressed data segments and extract metadata that characterizes their key features. By encapsulating the status frames, process parameters, and metadata in the packet header information through spatiotemporal correlation, the original, disordered mixed data stream is transformed into a structured, internally correlated adaptive aggregated data packet, thereby optimizing the transmission load and laying a data foundation for subsequent efficient transmission. This greatly reduces signaling overhead and total data volume, and improves channel utilization.

[0017] Second, this invention solves the problem of balancing resource mismatch and reliability under a unified transmission strategy by splitting data packets into sub-streams with different characteristics based on the aggregate packet header information and performing dynamic divide-and-conquer transmission for 5G network slicing. After encoding, based on the service quality tag list in the aggregate packet header information, the data content in the aggregate packet is accurately split into high-reliability sub-streams, high-bandwidth sub-streams, and relay guarantee sub-streams. A multi-factor decision algorithm is used to calculate the matching degree between the demand vector of each sub-stream and the capability vector of the 5G network slice, thereby dynamically mapping the high-reliability sub-stream to uRLLC slices, the high-bandwidth sub-stream to eMBB slices, and the relay guarantee sub-stream to dedicated or standard queues. This strategy enables critical status information to be transmitted in a dedicated high-reliability channel without interference from large data streams, ensuring deterministic latency and ultra-high reliability.

[0018] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating the steps of a real-time transmission method for production data in a 5G network; Figure 2 This is a flowchart of the steps in S500 of this embodiment of the invention for parsing, sorting and reassembling the data substream; Figure 3 This is an operation flowchart for a real-time transmission method of production data in a 5G network. Detailed Implementation

[0021] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0022] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a method for real-time transmission of production data in a 5G network,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plurality forms, unless the context clearly indicates otherwise; “plural” generally includes at least two.

[0023] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0024] To address the shortcomings of existing industrial IoT technologies, such as low efficiency, unreliability, and rigid network resource adaptation in the mixed transmission of multimodal production data, this invention provides a real-time production data transmission method using a 5G network. This method aims to achieve efficient, reliable, and resource-optimized real-time transmission by intelligently and collaboratively processing heterogeneous data generated from the same device source at the network edge, and by fully utilizing the differentiated Quality of Service (QoS) guarantees provided by 5G network slicing technology.

[0025] This invention is primarily applied to industrial internet scenarios such as intelligent manufacturing, remote monitoring, and predictive maintenance. In a 5G network environment, traditional unified transmission or independent parallel transmission schemes have significant shortcomings when dealing with small-byte periodic status data, medium-volume process parameters, and large-volume continuous waveform data simultaneously generated by industrial equipment. This invention constructs an end-to-end transmission technology system encompassing data generation, collaborative processing, network adaptation, and complete reconstruction through multimodal data feature analysis at the edge, adaptive aggregation coding, stream splitting based on quality of service tags, dynamic mapping for 5G network slicing, and intelligent reassembly at the receiving end.

[0026] Specifically, such as Figure 1 As shown, a method for real-time transmission of production data in a 5G network includes the following steps: S100. Collect multimodal production data generated simultaneously by the same industrial equipment source at the edge computing node, and perform real-time feature analysis on the multimodal production data to identify the transmission feature requirements of each data mode. S200. Based on the transmission characteristic requirements, the multimodal production data is adaptively aggregated and intelligently encoded to generate a structured adaptive aggregated data packet containing aggregated header information and aggregated data body. S300. Based on the requirements of the differentiated service quality tags defined in the aggregate packet header information, the adaptive aggregate data packet is split into three data sub-streams with different transmission characteristics, including a high reliability sub-stream, a high bandwidth sub-stream, and a relay guarantee sub-stream. S400: Based on the transmission characteristics of each data sub-stream, dynamically map it to the corresponding network slice in the 5G network that provides different quality of service guarantees; S500: The data sub-streams are transmitted in parallel through mapped network slices, and at the receiving end, all data sub-streams are parsed, sorted and reassembled according to the aggregated packet header information to restore the complete spatiotemporally synchronized multimodal production data.

[0027] In the specific implementation process, the edge computing nodes deployed on the factory site start to work. The edge computing nodes are specifically industrial gateways with 5G communication modules and certain computing capabilities, or dedicated edge servers deployed locally in the factory. The nodes collect multimodal production data from the same industrial equipment, such as welding robots, CNC machine tools, and industrial robots, in real time through industrial buses, IO modules, or dedicated sensor interfaces.

[0028] The collected multimodal production data includes: Small byte status data: Consists of the device's digital input / output (DI / DO) signals and the controller's internal status bits. In this embodiment, the number of bytes sent at one time is small, but the generation has a strict periodicity, and the timeliness and reliability requirements for transmission are high. Any loss or delay may lead to misjudgment of the control logic.

[0029] Process parameter data: generated by various analog sensors during equipment operation, such as voltage signals, current signals and video image signals of welding robots. In this embodiment, the amount of such data is moderate, and the timeliness and accuracy requirements for transmission are high. It is necessary to ensure that the data arrives within an acceptable time window and that the values ​​are accurate.

[0030] Waveform data: generated by high-frequency sampling vibration sensors, acoustic sensors or multi-channel vision sensors. In this embodiment, this type of data is represented as continuous waveforms or streaming media. It has a large data flow and high bandwidth requirements for transmission, but its tolerance for latency is relatively higher than that for state data. However, it is necessary to ensure the integrity and continuity of the data stream.

[0031] While collecting data, edge computing nodes initiate real-time feature analysis to analyze the inherent characteristics and transmission requirements of each data modality: Timeliness requirements: the period during which the analysis data is generated and the maximum allowable end-to-end delay. For example, in this embodiment, the status data requirement is <10ms, the process parameter requirement is <100ms, and the waveform data requirement is <500ms.

[0032] Accuracy requirements: Analyze whether lossy compression is allowed for the data, and the acceptable error range. Status data and key process parameters require lossless or extremely low distortion, while waveform data can be lossily compressed while retaining key features.

[0033] Spatiotemporal correlation: Analyzing the time synchronization and logical relationships between data from different modes. For example, the vibration waveform at a certain moment needs to be correlated with the equipment speed and load status at the corresponding moment in order to be used for fault diagnosis.

[0034] The results of feature analysis are quantified into a series of transmission feature requirement tags, providing a basis for subsequent intelligent aggregation and differentiated transmission.

[0035] The S200 process transforms the unordered and diverse raw data streams into a structured data packet with a clear internal structure and rich control information, namely an adaptive aggregated data packet. This adaptive aggregated data packet consists of two parts: an aggregated header and an aggregated data body.

[0036] S210. For periodically generated small-byte status data, a dynamic packetization algorithm based on timeliness weight is adopted to avoid the huge signaling overhead caused by sending each tiny data packet individually. The dynamic packetization algorithm works within a preset aggregation time window, and its working principle is implemented as follows: In time When the first small-byte status data packet is received from the target device source, a preset maximum tolerable packetization delay threshold is established. It also initializes an empty aggregate frame buffer and records the start timestamp. ; For subsequent arrivals belonging to the same equipment source... Data packets (timestamps) Data volume ),calculate ; At the same time, the packet weight of the data packet is calculated. ,in, For the first The amount of data in each data point This represents the historical average data volume. and Let be the harmonic coefficient, and , Used to balance the impact of data volume Used to reflect the decay of time-related effects; Add data packets to the buffer and accumulate their weights. ; When satisfied When the time is reached, a packing operation is triggered, which packs all small byte state data in the buffer into an aggregated state frame in chronological order. This frame contains not only the data content, but also the offset of each original data packet within the frame and the original timestamp.

[0037] Send the aggregation status frame into the S230 process, reset the buffer and timer, and begin aggregation for the next window.

[0038] S220. For continuously arriving high-volume waveform data, it is first divided into data segments of fixed duration. Then, based on the feature analysis results, lossless compression or high-fidelity lossy compression is performed on each data segment to generate compressed data segments. At the same time, metadata that can characterize its key features is extracted from each original data segment or compressed data segment, such as the main peak frequency of the spectrum, amplitude RMS, zero-crossing rate, specific frequency band energy, etc. The amount of this metadata is smaller than that of the original waveform, but it contains core information for rapid state judgment.

[0039] S230. The outputs of the first two steps, including the aggregated status frame, compressed data segment and its metadata, and the directly received process parameter data, are fused and encapsulated, specifically as follows: Constructing an aggregated data body: Based on the time axis, all data within an aggregated time window are aligned and assembled. For example, the aggregated status frame, several process parameter data points generated within the window, and waveform compressed data segments and corresponding metadata belonging to the time window are spliced ​​together according to a predefined structure to form an aggregated data body. During splicing, the starting position and length of each data segment in the aggregated data body are recorded.

[0040] Generate aggregate header information: This includes the following information: Globally unique identifier for data packets: used to uniquely identify and associate the split sub-streams at the receiving end.

[0041] The start and end timestamps of the aggregation time window precisely define the original time range of all data within this data packet.

[0042] Number of data sub-streams and Quality of Service (QoS) tag list: This explicitly indicates that this data packet will be split into several sub-streams and tags the QoS requirement level for each piece of data (such as status frames, metadata, compressed waveform segments, process parameters) in the aggregated data body.

[0043] Index table: Describes in tabular form the location, type, original timestamp, and relationships between each data segment within the aggregated data body.

[0044] Finally, the aggregate header information is appended before the aggregate data body to generate a complete adaptive aggregate data packet.

[0045] Based on the Quality of Service (QoS) tag list in the aggregate packet header information generated by S200, the system splits the structured aggregated data packet into three data sub-streams with different transmission characteristics: High-reliability sub-stream (first data sub-stream): Extracts data content marked with the highest reliability requirement by QoS tags, including: the entire aggregated status frame, containing all critical device status signals; metadata of all waveform data, containing waveform feature summaries; these data are combined and encapsulated into a high-reliability sub-stream.

[0046] High-bandwidth sub-stream (second data sub-stream): Extract data content marked with high bandwidth requirements by QoS tags. This data is combined into a high-bandwidth sub-stream. This sub-stream has a large data volume, and the core requirement is stable high throughput to ensure stream continuity.

[0047] Relay Assurance Substream (Third Data Substream): Extracts data content marked with QoS tags as required by standard assurance standards, i.e., process parameter data. This data is combined into a relay assurance substream.

[0048] After splitting, each sub-stream is accompanied by necessary control information related to the sub-stream extracted from the original aggregate header information to ensure that it can be correctly identified and reassembled at the receiving end.

[0049] Three data substreams with different characteristics are intelligently mapped to network slices in the 5G network that can provide corresponding quality of service guarantees. A slice selection algorithm based on multi-factor decision-making is adopted, and its working principle is implemented as follows: Define the demand and capability vector: for the first A demand vector is defined based on the characteristics of each data sub-stream to be transmitted. ,in, Indicates the first The maximum latency of each data substream Indicates the first The minimum reliability required for each data substream Indicates the first The bandwidth required for each data substream.

[0050] For the first Each available network slice predefines a capability vector. ,in, Represented as the first The latency that an available network slice can provide. Represented as the first The reliability that an available network slice can provide. Represented as the first The available network slices can allocate bandwidth.

[0051] Calculate the first The data substream and the first Matching score of each network slice ,in, As a weighting factor, and ,function Used to evaluate network slice capability vector For data substream demand vector In this embodiment, the degree of satisfaction is determined by the function. Including delay evaluation function Reliability assessment Bandwidth assessment These functions ensure that high scores are awarded when the slice capacity fully meets or exceeds the requirements, and scores are reduced when the capacity is insufficient.

[0052] For each data substream Calculate its matching score with all available network slices, and select the network slice with the highest score as its mapping target. In this embodiment: The high-reliability substream will be mapped to uRLLC (ultra-reliable low-latency communication) slices.

[0053] High-bandwidth substreams will be mapped to eMBB (enhanced mobile broadband) slices.

[0054] The relay guarantee substream will be mapped to a dedicated standard queue in an MIoT (Massive Internet of Things) slice or eMBB slice specifically designed for the Industrial Internet of Things.

[0055] The S500 process transmits each data sub-stream in parallel through mapped network slices, and at the receiving end, parses, sorts, and reassembles all data sub-streams according to the aggregated packet header information to reconstruct complete spatiotemporally synchronized multimodal production data, such as... Figure 2 As shown, the specific steps for S500 to parse, sort, and reorganize are as follows: Analysis: Receive and parse the data sub-streams transmitted through different network slices, and extract the aggregate packet header information carried in each data sub-stream; Sorting: Based on the globally unique identifier of the data packet in the aggregation header information, all data sub-streams belonging to the same adaptive aggregation data packet are classified, and the processing sequence of each data sub-stream and its internal data segments is determined according to the start and end timestamps of the aggregation time window and the internal index order in the aggregation header information. Reassembly: Based on the index table in the aggregated packet header information, the data in the sorted high-reliability substream, high-bandwidth substream, and relay guarantee substream are aligned and spliced. The compressed data segments in the high-bandwidth substream are decompressed, and a complete data sequence containing small-byte status data, process parameter data, and waveform data, synchronized with the original acquisition time, is restored and output.

[0056] like Figure 3 As shown, the specific steps of the real-time transmission method for production data in a 5G network provided by this invention are as follows: (1) Data acquisition and feature analysis Production data from different welding robots can be collected synchronously from the same device on edge computing nodes deployed in industrial sites.

[0057] The collected data is analyzed in real time to identify the transmission requirements for each type of data, including high reliability / high timeliness, medium timeliness / medium reliability, and high bandwidth / continuous stream.

[0058] (2) Intelligent aggregation and packaging Multiple device status signals generated in consecutive cycles, each with a small data volume, are packaged and merged into a single data packet (aggregated status frame) according to an adjustable strategy.

[0059] The system segments and compresses massive amounts of continuous waveform data to generate data blocks that are easy to transmit, and extracts summary information that represents their key features from these blocks.

[0060] Generate a detailed packet header, including the packet's unique number, time range, internal structure diagram, and label the network requirement level for different types of data within the packet.

[0061] The packaged status frames, compressed waveform data blocks and their feature summaries, and process parameters are integrated according to time sequence and correlation to form a structured adaptive aggregated data package.

[0062] (3) Split data streams according to transmission requirements Read the requirement level marked in the data packet header.

[0063] Data with the highest reliability requirement is combined to form a high-reliability substream.

[0064] Data content with high bandwidth requirements is combined to form a high bandwidth substream.

[0065] Data content that meets the standard guarantee requirement level is combined together to form a relay guarantee substream.

[0066] (4) Dynamically map to 5G network slices For each data sub-stream's transmission requirements, assess the service capabilities of each network slice in the current 5G network.

[0067] The decision-making mechanism selects the network slice that best matches the needs of each sub-flow.

[0068] (5) Parallel transmission and receiver reassembly The three data sub-streams are transmitted simultaneously and independently through their respective allocated 5G network slices.

[0069] At the receiving end in the cloud or remotely, the various sub-streams belonging to the same original data packet are first received and identified.

[0070] Arrange the data blocks in all substreams in the correct order based on the timestamps and internal structure in the packet header.

[0071] The arranged data blocks are spliced ​​together, and the compressed waveform data is decompressed and restored to finally recover a complete sequence that is completely synchronized with the original time and contains all types of data, for use by upper-layer applications.

[0072] In summary, this invention solves the challenge of real-time transmission of multimodal industrial production data through a full-link design encompassing intelligent data aggregation at the edge, precise resource matching on the network side, and lossless data restoration at the receiving end. In its implementation, this method significantly reduces signaling overhead and total data transmission volume through dynamic packaging and feature extraction compression. QoS-based intelligent splitting and dynamic mapping for 5G slicing ensure that critical status information is transmitted in dedicated, highly reliable channels, free from interference from large-volume data, guaranteeing deterministic low latency and ultra-high reliability while allowing large-volume data to enjoy sufficient bandwidth resources. Finally, at the receiving end, intelligent guidance from packet header information reassembles spatiotemporally synchronized complete data, providing a high-quality data foundation for advanced industrial applications. The entire process is adaptive and efficient, fully utilizing the capabilities of the 5G network and achieving refined matching and optimized utilization of network resources and industrial business needs.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for real-time transmission of production data in a 5G network, characterized in that, The specific steps of this method are as follows: S100. Collect multimodal production data generated simultaneously by the same industrial equipment source at the edge computing node, and perform real-time feature analysis on the multimodal production data to identify the transmission feature requirements of each data mode. S200. Based on the transmission characteristic requirements, the multimodal production data is adaptively aggregated and intelligently encoded to generate a structured adaptive aggregated data packet containing aggregated header information and aggregated data body. S300. Based on the requirements of the differentiated service quality tags defined in the aggregate packet header information, the adaptive aggregate data packet is split into three data sub-streams with different transmission characteristics, including a high reliability sub-stream, a high bandwidth sub-stream, and a relay guarantee sub-stream. S400: Based on the transmission characteristics of each data sub-stream, dynamically map it to the corresponding network slice in the 5G network that provides different quality of service guarantees; S500: The data sub-streams are transmitted in parallel through mapped network slices, and at the receiving end, all data sub-streams are parsed, sorted and reassembled according to the aggregated packet header information to restore the complete spatiotemporally synchronized multimodal production data.

2. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In S100, the multimodal production data includes: Small-byte status data with periodicity, consisting of device input and output status signals; Process parameter data consisting of temperature, pressure, and speed parameters during equipment operation; Waveform data generated by vibration sensors, acoustic sensors, or vision sensors.

3. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In S100, the feature requirements include timeliness requirements, accuracy requirements, and the spatiotemporal correlation between different data modalities; Edge computing nodes are industrial gateways with 5G communication capabilities or edge servers deployed locally in the factory.

4. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In step S200, the steps of adaptive aggregation and intelligent encoding are as follows: S210. Within a preset aggregation time window, package the small-byte state data generated in multiple cycles to generate an aggregated state frame. S220. Perform segmentation and lossless compression on continuous waveform data to generate at least one compressed data segment, and extract the key features of each compressed data segment as metadata. S230. The aggregated status frame, the metadata, and the process parameter data are spatiotemporally correlated and encapsulated to form the aggregated data body, and the aggregated header information describing the internal structure, correlation, and service quality requirements of the aggregated data body is generated.

5. The method for real-time transmission of production data in a 5G network according to claim 4, characterized in that, In step S210, the process of packaging small-byte state data generated in multiple cycles to generate an aggregated state frame employs a dynamic packaging algorithm based on timeliness weights. The working principle of this dynamic packaging algorithm is as follows: In time Upon receiving the first small byte of status data, a maximum tolerable packet delay threshold is preset. Initialize an aggregated state frame and record its initial timestamp. ; For subsequent small-byte status data originating from the same device source, calculate their timestamps. and The difference ; according to Calculate the first Packing weight of each data point ,in, For the first The amount of data in each data point This represents the historical average data volume. and Let be the harmonic coefficient, and ; When the cumulative weights and Exceeding the preset threshold, or Reaching the maximum time difference At that time, all currently cached small-byte state data are packaged into a final aggregated state frame and sent to S230.

6. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In step S200, the generated aggregate header information includes: a globally unique identifier for the data packet, start and end timestamps of the aggregation time window, a list of data substreams and quality of service labels, and an index table describing the position and relationship of each data segment within the aggregated data body.

7. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In S300, splitting the adaptive aggregated data packet into three data sub-streams with different transmission characteristics specifically involves: Based on the list of service quality tags contained in the aggregated packet header information, identify the service quality level corresponding to each part of the data in the adaptive aggregated data packet; Generate the high-reliability substream: Combine the data content marked with the highest reliability requirement in the service quality label list, i.e., aggregate the status frame and metadata, to generate the first data substream; Generate the high-bandwidth substream: Combine the data content marked as having high bandwidth requirements in the Quality of Service (QoS) tag list, i.e., compressed data segments, to generate a second data substream; Generate the relay assurance sub-stream: Combine the data content marked as standard assurance requirements in the service quality label list, i.e., process parameter data, to generate a third data sub-stream.

8. The method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, In S400, the dynamic mapping adopts a slice selection algorithm based on multi-factor decision-making. The working principle of this slice selection algorithm is as follows: For the first Each data sub-stream predefines a demand vector. ,in, Indicates the first The maximum latency of each data substream Indicates the first The minimum reliability required for each data substream Indicates the first The bandwidth required for each data substream; For the first Each available network slice predefines a capability vector. ,in, Represented as the first The latency that an available network slice can provide. Represented as the first The reliability that an available network slice can provide. Represented as the first The available network slices can allocate bandwidth; Calculate the first The data substream and the first Matching score of each network slice ,in, As a weighting factor, and ,function Used to evaluate network slice capability vector For data substream demand vector The degree of satisfaction; For each data substream Select Match Score The network slice with the highest value is used as the mapping target.

9. A method for real-time transmission of production data in a 5G network according to claim 1, characterized in that, The specific steps of S500 are as follows: Analysis: Receive and parse the data sub-streams transmitted through different network slices, and extract the aggregate packet header information carried in each data sub-stream; Sorting: Based on the globally unique identifier of the data packet in the aggregation header information, all data sub-streams belonging to the same adaptive aggregation data packet are classified, and the processing sequence of each data sub-stream and its internal data segments is determined according to the start and end timestamps of the aggregation time window and the internal index order in the aggregation header information. Reassembly: Based on the index table in the aggregated packet header information, the data in the sorted high-reliability substream, high-bandwidth substream, and relay guarantee substream are aligned and spliced. The compressed data segments in the high-bandwidth substream are decompressed, and a complete data sequence containing small-byte status data, process parameter data, and waveform data, synchronized with the original acquisition time, is restored and output.

Images