5G network system and method for industrial control requirements

Abstract

The invention provides a 5G network system and method for industrial control requirements. The system comprises a first 5G terminal, a second 5G terminal, a timestamp generator, a PLC master station, a PLC slave station and a network data analysis computer. Two 5G terminals configured to transmit and receive industrial control data messages in a 5G network; the timestamp generator is respectively connected with the two 5G terminals, the PLC master station, the PLC slave station and the network data analysis computer through a plurality of Ethernet ports, and is configured to load timestamps and mirror data in an industrial control data message transmission process; the PLC master station and the PLC slave station are configured to generate and receive control instructions and state data; and the network data analysis computer is configured to collect mirror image data and execute time delay calculation, data storage, multi-dimensional performance analysis and industrial control system influence evaluation. According to the system, high-precision 5G network performance analysis oriented to industrial control is realized.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a 5G network system and method for industrial control needs. Background Technology

[0002] Existing technologies mainly involve 5G network performance analysis, including network probing-based and simulation-based approaches. Network probing-based approaches deploy probing devices in the 5G network to send test data packets (such as ping packets) and collect network performance parameters (such as packet loss rate, latency, and jitter). However, the measurement accuracy is limited to milliseconds, which cannot meet the nanosecond latency accuracy requirements of industrial control systems. Furthermore, the test packets do not match the characteristics of actual industrial data streams, lack bidirectional latency separation capabilities, and the statistical methods are limited (e.g., only providing average or maximum values, unable to perform CCDF and PDF analysis). Simulation-based approaches simulate industrial scenarios by establishing 5G network simulation models. However, the simulation environment differs from the real network, resulting in insufficient model accuracy. They cannot accurately model the interaction and dynamic characteristics of industrial equipment, have low time synchronization accuracy, and are difficult to reflect long-term performance fluctuations.

[0003] It is evident that existing technical solutions cannot effectively address the matching problem between 5G network performance analysis and actual industrial needs. Specifically, this manifests as low measurement accuracy, data loss, lack of ability to combine and analyze industrial parameters, and insufficient dynamic performance evaluation. Summary of the Invention

[0004] The purpose of this application is to provide a 5G network system and method for industrial control needs, which achieves high-precision 5G network performance analysis for industrial control through nanosecond-level timestamp measurement, sliding window dynamic analysis and industrial parameter coupling evaluation, and solves the problem of insufficient measurement accuracy and disconnect between existing technology and industrial needs.

[0005] In the first aspect, a 5G network analysis system for industrial control needs is provided, which may include: a first 5G terminal, a second 5G terminal, a timestamp generator, a PLC master station, a PLC slave station, and a network data analysis computer; The first 5G terminal and the second 5G terminal are configured to send and receive industrial control data packets in a 5G network. The timestamp generator is connected to the first 5G terminal, the second 5G terminal, the PLC master station, the PLC slave station, and the network data analysis computer through multiple Ethernet ports. It is configured to load timestamps and mirror data during the transmission of industrial control data messages. The PLC master station and PLC slave station are configured to generate and receive control commands and status data. The network data analysis computer is configured to acquire mirrored data and perform latency calculations, data storage, multi-dimensional performance analysis, and impact assessments on industrial control systems.

[0006] In one possible implementation, the timestamp generator includes at least five Ethernet ports, wherein: The first Ethernet port is connected to the first 5G terminal and is responsible for forwarding data from the PLC master station to the 5G network; The second Ethernet port is connected to the PLC master station to receive control commands and status query messages; The third Ethernet port is connected to the second 5G terminal and is responsible for forwarding 5G network data to the PLC slave station; The fourth Ethernet port is connected to the PLC slave station, forwarding 5G network messages and receiving status feedback data; The fifth Ethernet port is connected to the network data analysis computer.

[0007] In one possible implementation, the fifth Ethernet port is configured as a bidirectional mirror port, fully replicating the bidirectional data streams between the first and second Ethernet ports, and between the third and fourth Ethernet ports, and connecting to a network data analysis computer.

[0008] In one possible implementation, the PLC master station and the PLC slave station support Profinet or EtherNet / IP industrial Ethernet protocols, with an instruction execution cycle of no more than 1 millisecond; and are directly connected to the corresponding port of the timestamp generator via physical cables.

[0009] In one possible implementation, the system supports both 5G private network architecture and 5G public network architecture; The 5G private network architecture adopts local deployment of user plane functions and 5G LAN technology; The 5G public network architecture adopts a dual-layer secure transmission architecture of VPN and VxLAN. The first 5G terminal and the second 5G terminal are directly connected via LAN under the 5G private network architecture, and configured with VPN tunnel and VxLAN virtual interface binding under the 5G public network architecture.

[0010] Secondly, a 5G network analysis method for industrial control needs is provided, applied to the 5G network analysis system for industrial control needs described in the first aspect. This method may include: The timestamp generator loads a first timestamp before the data packet enters the 5G network and a second timestamp after the data packet leaves the 5G network. The network data analysis computer then performs validity screening based on the collected mirror data and calculates the 5G network latency data based on the timestamp difference of the matched data packet pairs. The network data analysis computer stores the latency data and related feature information into the database according to a preset field structure, and constructs the original measurement data table and the statistical analysis data table. The network data analysis computer performs multi-dimensional statistical analysis on the stored latency data to obtain statistical analysis results. By combining the parameters of the industrial control system and using statistical analysis results, the impact of 5G networks on the industrial control system is assessed, and a quantitative assessment report is generated.

[0011] In one possible implementation, the methods for loading timestamps include: This is achieved by inserting a timestamp field between the preamble and the destination MAC address of the Ethernet frame using a custom field insertion method. The timestamp loading node is precisely located as a key node before the data enters the 5G network and after it leaves the 5G network.

[0012] In one possible implementation, the network data analysis computer performs validity screening based on the collected mirror data, including: Message matching is performed based on a dual-stack mechanism, creating an uplink data stack and a downlink data stack to store uplink messages sent from the PLC master station to the PLC slave station and downlink messages sent from the PLC slave station to the PLC master station, respectively. By comparing the source MAC address, destination MAC address, and data payload of data packets, matching data packet pairs are identified, and unmatched data packets are considered invalid and discarded. A unique sequence number is assigned to each matching data packet pair, and they are sorted by timestamp to ensure consistency in the data processing sequence.

[0013] In one possible implementation, the network data analysis computer performs multi-dimensional statistical analysis on the stored latency data to obtain statistical analysis results, including: Read latency data within the most recent time window from the stored data in seconds, and calculate the real-time average latency, maximum latency, minimum latency, and number of latency exceedances; Configure a variable data volume time window, and calculate the complementary cumulative distribution function and probability density function for the time delay data in each window; A continuous analysis window is generated based on the user-defined sliding step size and basic data volume. The average round-trip time, latency fluctuation coefficient, and industrial adaptability score within each window are calculated to evaluate the long-term trend of network performance.

[0014] Thirdly, an electronic device is provided, which includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements any of the steps described in the second aspect above.

[0015] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the steps of any of the methods described in the second aspect above.

[0016] This application provides a 5G network analysis method and apparatus for industrial control needs. The system includes a first 5G terminal, a second 5G terminal, a timestamp generator, a PLC master station, a PLC slave station, and a network data analysis computer. The first and second 5G terminals are configured to send and receive industrial control data packets in a 5G network. The timestamp generator is connected to the first 5G terminal, the second 5G terminal, the PLC master station, the PLC slave station, and the network data analysis computer via multiple Ethernet ports, and is configured to load timestamps and mirror data during industrial control data packet transmission. The PLC master station and the PLC slave station are configured to generate and receive control commands and status data. The network data analysis computer is configured to collect mirrored data and perform latency calculations, data storage, multi-dimensional performance analysis, and industrial control system impact assessment. This system achieves high-precision, industrial control-oriented 5G network performance analysis through nanosecond-level timestamp measurement, sliding window dynamic analysis, and coupled industrial parameter evaluation, solving the problem of insufficient measurement accuracy and disconnect between existing technologies and industrial needs. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of the structure of a 5G network analysis system for industrial control needs provided in this application embodiment; Figure 2 A flowchart illustrating a 5G network analysis method for industrial control needs provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application. Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art. The terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are only used to distinguish different components. The terms "comprising" or "including," etc., mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but does not exclude other elements or objects. The terms "connection," "coupled," or "linked," etc., are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0020] Example 1 The 5G network analysis system for industrial control needs provided in this application embodiment can be applied to... Figure 1 In the system architecture shown, such as Figure 1 As shown, the system may include: a first 5G terminal (UE1), a second 5G terminal (UE2), a timestamp generator, a PLC master station (PLC1), a PLC slave station (PLC2), and a network data analysis computer.

[0021] The first 5G terminal and the second 5G terminal are configured to send and receive industrial control data packets in a 5G network. The timestamp generator is connected to a first 5G terminal, a second 5G terminal, a PLC master station, a PLC slave station, and a network data analysis computer via multiple Ethernet ports. It is configured to load timestamps and mirror data during the transmission of industrial control data messages. The mirrored data may include timestamped industrial control messages, original control messages, and 5G network control signaling.

[0022] The PLC master station and PLC slave station are configured to generate and receive control commands and status data. The network data analysis computer is configured to acquire mirrored data and perform latency calculations, data storage, multi-dimensional performance analysis, and impact assessments on industrial control systems.

[0023] The system supports both 5G private network architecture and 5G public network architecture. The 5G private network architecture adopts local deployment of user plane functions to shorten data transmission links; it introduces 5G LAN technology to enable direct interconnection between industrial equipment such as PLCs and 5G terminals, avoids the forwarding delay of the public core network, and meets the requirements of low latency, localized data and high security within the factory area.

[0024] The 5G public network architecture employs a dual-layer secure transmission architecture of VPN and VxLAN to achieve logical interconnection of industrial equipment across geographical regions. Specifically, VPN ensures security through encrypted data transmission, while VxLAN technology encapsulates industrial data packets to achieve logical interconnection of industrial equipment across physical networks, ensuring transmission stability and compatibility in public network scenarios. After completing the above two network architecture designs, the configuration of subsequent terminal devices, timestamp generators, and PLCs differs depending on the network mode: In the 5G private network architecture, each device is deployed in the same industrial control domain via direct LAN connection; while in the 5G public network architecture, cross-network logical interconnection is achieved through VPN tunnels and VxLAN encapsulation, and the timestamp generator needs to be configured with public network address mapping to ensure synchronization of latency data collection.

[0025] (1) For the first 5G terminal (UE1) and the second 5G terminal (UE2), UE1 and UE2 adopt industrial-grade 5G CPE equipment, which supports gigabit Ethernet interface and PTPv2 nanosecond-level time synchronization protocol. The first 5G terminal and the second 5G terminal are directly connected via LAN under the 5G private network architecture, and VPN tunnel and VxLAN virtual interface binding are configured under the 5G public network architecture.

[0026] (2) For the timestamp generator: As a core module, it uses Beckhoff ET2000 industrial-grade equipment or a high-performance industrial computer (CPU frequency ≥ 3.0 GHz, memory ≥ 16 GB) with at least five Gigabit Ethernet ports. The port definitions are as follows: The first Ethernet port ETH1 is connected to the first 5G terminal UE1 and is responsible for forwarding data from the PLC master station to the 5G network. The second Ethernet port ETH2 is connected to the PLC master station to receive control commands and status query messages; The third Ethernet port ETH3 is connected to the second 5G terminal UE2, which is responsible for forwarding 5G network data to the PLC slave station; The fourth Ethernet port ETH4 is connected to the PLC slave station, forwarding 5G network messages and receiving status feedback data; The fifth Ethernet port, ETH5, is connected to the network data analysis computer, and is configured as a bidirectional mirroring port, fully replicating ETH1. ETH2 and ETH3 The bidirectional data stream between ETH4 devices includes messages with t1 / t2 timestamps, raw control messages between PLC devices, and a small amount of 5G network control signaling, ensuring that the mirrored data covers all transmission content related to industrial control and connects to the network data analysis computer.

[0027] The timestamp generator allocates an independent cache space (≥1GB) for the mirror link. When the peak data traffic of the main link exceeds gigabit bandwidth, the mirror data can be temporarily stored in the cache to avoid packet loss due to insufficient cache. At the same time, the transmission rate of the mirror link is consistent with that of the main link (1Gbps) to ensure the timing consistency between the mirror data and the main link data, and to provide timing basis for subsequent packet matching. The network data analysis computer's network card is configured in promiscuous mode and connected to the fifth Ethernet port ETH5 of the timestamp generator. It can receive all mirrored data transmitted through the ETH5 port and process the mirrored data through a dedicated data receiving thread (with higher priority than other threads) to avoid loss of mirrored data due to insufficient computer resources.

[0028] The connections between UE1 and timestamp generator ETH1, and between UE2 and ETH3, utilize CAT6 shielded gigabit Ethernet cables to resist electromagnetic interference in the industrial environment. The connections between the PLC master (PLC1) / PLC slave (PLC2) and timestamp generators ETH2 / ETH4 utilize unshielded gigabit Ethernet cables, simplifying the design due to their close proximity. ETH5 connects to the network data analysis computer via a gigabit Ethernet cable, ensuring unblocked transmission of mirrored data. The specific connection methods are as follows: UE1 connects directly to the ETH1 port of the timestamp generator via a shielded gigabit Ethernet cable to avoid increased latency caused by intermediate transfers; The ETH2 port of the timestamp generator is connected to the PLC1 via an unshielded gigabit Ethernet cable. Since the PLC1 and the timestamp generator are usually deployed in the same control cabinet, the electromagnetic environment is stable, which simplifies the link design. The connection method between UE2 and the timestamp generator ETH3 is the same as that between UE1, ensuring the symmetry of the bidirectional transmission link and avoiding delay deviations caused by link differences; The ETH4 port of the timestamp generator is connected to PLC2 via an unshielded gigabit Ethernet cable, which similarly ensures the stability of short-distance transmission. The ETH5 port of the timestamp generator is connected to the network data analysis computer via a gigabit Ethernet cable to transmit full mirror data. This link needs to ensure sufficient bandwidth to avoid packet loss caused by mirror data backlog.

[0029] (3) For PLC devices: PLC master station (PLC1) is responsible for issuing control commands such as device start / stop and parameter adjustment; PLC slave station (PLC2) is responsible for transmitting device status data back. Among them, PLC master station and PLC slave station support Profinet or EtherNet / IP industrial Ethernet protocol, and the command execution cycle is no more than 1 millisecond; and they are directly connected to the corresponding port of the timestamp generator through physical cables.

[0030] Furthermore, the end-to-end data transmission path formed by this system includes: Uplink path (PLC1→PLC2): The control command message generated by PLC1 is first transmitted to the ETH2 port of the timestamp generator via Ethernet cable. After being timestamped, it is forwarded from the ETH1 port to UE1. UE1 encapsulates the message into a 5G data frame and transmits it to UE2 via the 5G network (private network / public network). After receiving the data frame, UE2 decapsulates it into an Ethernet message and forwards it to the ETH3 port of the timestamp generator. After being timestamped again, it is transmitted from the ETH4 port to PLC2, completing the downlink transmission of the control command. Downlink path (PLC2→PLC1): The status feedback message generated by PLC2 is transmitted to the ETH4 port of the timestamp generator via Ethernet cable. After loading the timestamp, it is forwarded to UE2 from the ETH3 port. UE2 encapsulates the message into a 5G data frame and transmits it to UE1 via the 5G network. UE1 decapsulates the message and forwards it to the ETH1 port of the timestamp generator. After loading the timestamp, it is transmitted to PLC1 from the ETH2 port, completing the uplink transmission of status data.

[0031] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application. Furthermore, the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0032] Example 2 Figure 2 This is a flowchart illustrating a 5G network analysis method for industrial control needs, provided as an embodiment of this application. Figure 2 As shown, this method is applied to the aforementioned 5G network analysis system for industrial control needs, and the method may include: Step S210: Load the first timestamp before the data packet enters the 5G network using a timestamp generator, and load the second timestamp after the data packet leaves the 5G network. Then, use a network data analysis computer to perform validity screening based on the collected mirror data and calculate the latency data of the 5G network based on the timestamp difference of the matched data packet pairs.

[0033] In specific implementation, the timestamp loading method includes: inserting a timestamp field between the Ethernet frame preamble and the destination MAC address using a custom field insertion method. The timestamp loading node is precisely located at key nodes before and after data enters the 5G network. Specifically: The timestamp generator monitors the data stream on the ETH2 port in real time. When a message is detected from PLC1, the timestamp generation logic is triggered. Based on a local high-precision real-time clock (or an external clock synchronized via PTPv2), the current nanosecond-level time (accuracy ≤ 10ns) is obtained. Using the "custom field insertion" method, a t1 timestamp field (8 bytes long, formatted as UTC time + nanosecond offset) is inserted between the Ethernet frame preamble and the destination MAC address. This insertion method does not modify the original message's protocol structure, avoiding impact on subsequent 5G network transmission. After timestamp loading is complete, the timestamp generator forwards the message from the ETH1 port to UE1 for 5G network transmission. The timestamp generator listens to the data stream on the ETH3 port. When it detects a 5G network packet forwarded by UE2, it triggers the t2 timestamp generation logic. It also obtains nanosecond-level time based on a high-precision clock and inserts a t2 timestamp field (8 bytes in length, with the same format as t1) after the t1 field of the Ethernet frame to ensure that the two timestamps can be quickly identified by the subsequent data analysis module. After t2 loading is completed, the packet is forwarded from the ETH4 port to PLC2.

[0034] Here, t1 corresponds to the last node before the data enters the 5G network, and t2 corresponds to the first node after the data leaves the 5G network. The time difference between the two directly reflects the one-way transmission latency of the 5G network. At the same time, the unified timestamp format ensures that subsequent data processing can be quickly parsed, avoiding the time-consuming format conversion. Furthermore, if the timestamp generator uses PTPv2 synchronization, the latency value needs to be corrected according to the clock synchronization deviation (≤5ns).

[0035] (1) The specific implementation steps for filtering the validity of collected mirror data using network data analysis computers are the core link in converting the original mirror data into usable analytical data. The goal is to eliminate invalid data, clarify data links, and quantify network latency through a three-step processing flow of validity filtering, flow classification, and latency calculation, providing a structured and accurate data foundation for subsequent data storage and performance analysis. Among them, data validity matching ensures that the analysis object is only industrial control-related data, flow judgment provides a basis for bidirectional latency calculation, and latency calculation directly outputs core indicators reflecting the performance of 5G networks. The three are progressively linked, forming a complete closed loop of data processing, including: A dual-stack mechanism is used for message matching, creating an uplink data stack and a downlink data stack. The uplink data stack stores messages sent from the PLC master to the PLC slave, and the downlink data stack stores messages sent from the PLC slave to the PLC master. Matching data message pairs are identified by comparing the source MAC address, destination MAC address, and data payload of the data messages. Unmatched data messages are considered invalid and discarded. A unique sequence number is assigned to each matching data message pair, and they are sorted by timestamp to ensure the consistency of data processing timing.

[0036] Specifically, in industrial control scenarios, mirrored data contains a large amount of invalid information (such as 5G network signaling and device heartbeat messages), and a dual-stack matching mechanism is needed to filter out the valid data related to PLC master-slave communication.

[0037] After receiving mirrored data through the ETH port, the network data analysis computer starts a message parsing thread to extract key information from each message, including the source MAC address, destination MAC address, data payload (i.e., PLC control instructions or status data), and t1 / t2 timestamp fields, storing these in a temporary data buffer. It also creates uplink and downlink data stacks to store uplink messages from PLC1 to PLC2 and downlink messages from PLC2 to PLC1, respectively. For each message in the temporary data buffer, it initially determines its stack based on the source MAC address (messages with source MAC_PLC1 are placed in the uplink stack, and those with source MAC_PLC2 are placed in the downlink stack). The system searches the corresponding stack for message pairs with identical payloads and interchanged source / destination MAC addresses (e.g., if a message in the uplink stack has a device start / stop command payload and a destination MAC address of MAC_PLC2, a matching acknowledgment message with the same payload and source MAC address of MAC_PLC2 must be found in the downlink stack). When a matching message pair is found, it is determined to be valid industrial control data transmitted via the 5G network and marked as matched. Messages without a matching pair (such as isolated heartbeat messages or incomplete messages) are determined to be invalid data and deleted from the buffer. Simultaneously, a unique sequence number is assigned to each successfully matched message pair, and the message pairs are sorted according to the t1 timestamp order to ensure the timing consistency of subsequent data processing and analysis.

[0038] (2) The specific implementation steps for calculating the latency data of the 5G network based on the timestamp difference of matched data packet pairs include: First, the data flow direction is determined based on the MAC address: Clearly defining the data flow direction is a prerequisite for distinguishing uplink / downlink latency and calculating round-trip latency. Accurate flow classification is achieved by establishing a MAC address mapping relationship. MAC address library initialization: When the data analysis module starts, it automatically reads the MAC addresses of PLC1 and PLC2 (which can be actively queried through the industrial Ethernet protocol or manually entered by the user) and records them as MAC_PLC1 (PLC master station) and MAC_PLC2 (PLC slave station) respectively. It also establishes a mapping table of MAC address-device role-data flow direction (e.g., MAC_PLC1→PLC master station→uplink data sender, MAC_PLC2→PLC slave station→downlink data sender).

[0039] Flow determination logic: For matched message pairs, extract the source MAC address and destination MAC address of the uplink message: if source MAC = MAC_PLC1 and destination MAC = MAC_PLC2, it is determined to be uplink data (PLC1→PLC2); extract the source MAC address and destination MAC address of the downlink message: if source MAC = MAC_PLC2 and destination MAC = MAC_PLC1, it is determined to be downlink data (PLC2→PLC1); if the source / destination MAC address of the message does not belong to MAC_PLC1 or MAC_PLC2 (such as the base station MAC address of a 5G network), it is determined to be non-industrial control data and excluded from subsequent latency calculation and performance analysis to avoid interfering with the analysis results.

[0040] Flow direction identification and storage: Add a flow direction identifier (such as "U" for uplink data and "D" for downlink data) to the sequence number of the matching message pair, and store the flow direction information in a data structure to provide a directional basis for subsequent latency calculation.

[0041] Next, calculate the end-to-end delay and round-trip time (RTT): Extract the t1 timestamp (t1_u) of the uplink message and the t1 timestamp (t1_d) of the downlink message from the matched message pairs, as well as the t2 timestamp (t2_u) of the uplink message and the t2 timestamp (t2_d) of the downlink message; at the same time, verify the validity of the timestamps: check whether the timestamp format conforms to the preset standard (UTC time + nanosecond offset) and whether the t2 time is later than the t1 time (to avoid negative delay caused by clock synchronization deviation). If the verification fails, mark the message pair as invalid and do not participate in the delay calculation. Uplink latency (T_uplink) reflects the 5G network transmission latency from PLC1 to PLC2. The calculation formula is T_uplink=t2_u-t1_u, where t1_u is the timestamp before the uplink message enters the 5G network, and t2_u is the timestamp after the uplink message leaves the 5G network. The difference between the two is the one-way transmission latency of the 5G network. Downlink latency (T_downlink) reflects the 5G network transmission latency from PLC2 to PLC1. The calculation formula is T_downlink=t2_d-t1_d, and the logic is the same as that of uplink latency. Round-trip time (RTT) calculation reflects the key indicator for evaluating network stability in industrial control scenarios. It reflects the complete cycle delay of master station command issuance and slave station confirmation transmission. The calculation formula is RTT = T_uplink + T_downlink. RTT is calculated only for uplink / downlink packets within the same data interaction cycle to ensure that the result reflects the round-trip transmission time of a single control command.

[0042] If the timestamp generator uses PTPv2 to synchronize an external clock, a clock synchronization deviation calibration step needs to be added after the delay calculation. The calculation result is corrected according to the synchronization deviation value of the PTP protocol (usually ≤5ns) to further improve the accuracy of delay measurement.

[0043] After the aforementioned latency calculations and clock skew calibration, a structured latency result dataset is obtained, including three categories of indicators: uplink latency, downlink latency, and round-trip latency, as well as the original timestamps. These indicators, as the final measurement output, along with the corresponding timestamps, MAC addresses, and flow information of the packets, are written into the data structure to be stored. To ensure data traceability and the continuity of subsequent statistical analysis, this dataset will be imported into a database as raw input for structured storage and hierarchical management.

[0044] Step S220: The latency data and related feature information are stored in the database according to the preset field structure by the network data analysis computer, and the original measurement data table and statistical analysis data table are constructed.

[0045] Relevant characteristic information may include timestamps, data flow direction, latency value, message size, and sequence number.

[0046] By designing a RawData table for raw measurement data and a StatisticsData table for statistical analysis data, the system stores the raw measurement data and the subsequently generated statistical data respectively, achieving hierarchical data storage and on-demand retrieval: the RawData table retains all key measurement information, ensuring data traceability and secondary analysis; the StatisticsData table stores aggregated statistical indicators, improving the efficiency of subsequent analysis. The storage architecture design must balance data integrity and query efficiency, employing a relational database (such as MySQL or PostgreSQL) for structured data management, while optimizing data retrieval speed through indexing.

[0047] Table 1: Raw Measurement Data

[0048] Table 2: Statistical Analysis Data

[0049] Step S230: Perform multi-dimensional statistical analysis on the stored latency data using a network data analysis computer to obtain the statistical analysis results.

[0050] Based on preprocessed structured data, a three-layer analysis model—real-time monitoring, statistical analysis, and industrial linkage—is constructed. The core objective is to comprehensively evaluate the performance of 5G networks in industrial control scenarios from three dimensions: short-term performance fluctuations, long-term distribution characteristics, and industrial scenario adaptation. Real-time performance monitoring meets the real-time requirements of industrial control, quickly reporting short-term network fluctuations; variable window statistical analysis uncovers latency distribution patterns and determines network reliability boundaries; and sliding window industrial linkage analysis combines network performance with industrial control parameters to ensure the analysis results align with actual industrial needs. The three-layer model progresses progressively, from general performance evaluation to industry-specific evaluation, forming a complete analytical system.

[0051] In practice, latency data within the most recent time window is read from the stored data in seconds, and the real-time average latency, maximum latency, minimum latency, and number of latency exceedances are calculated. A variable data volume time window is configured, and the complementary cumulative distribution function (CCDF) and probability density function (PDF) are calculated for the latency data within each window. A continuous analysis window is generated based on the user-defined sliding step size and basic data volume, and the average round-trip latency, latency fluctuation coefficient, and industrial adaptability score within each window are calculated to evaluate the long-term trend of network performance.

[0052] Specifically: (1) Read the latency data within the most recent time window from the stored data in seconds, and calculate the real-time average latency, maximum latency, minimum latency, and number of latency exceedances, including: To address the need for real-time network status awareness in industrial control scenarios, a real-time performance monitoring model with second-level updates is designed to quickly respond to short-term performance fluctuations in the 5G network. Data acquisition mechanism: The data analysis module starts a real-time monitoring thread, which reads the raw data of the Timestamp within the most recent second from the RawData table every second (the default is to read the most recent 1000 sets, and if the data volume is insufficient, it will read all). Real-time data of different network modes and different flow directions are obtained by filtering the NetworkMode+Direction conditions.

[0053] Real-time average latency: The arithmetic mean of all Delay fields in this group within the current second; Real-time maximum delay: The maximum value of the Delay field for this group within the current second; Real-time minimum latency: The minimum value of the Delay field for this group within the current second; Delay Exceedance Count: Counts the number of times Delay ≥ preset threshold (default equal to the sampling period of the industrial control system) within the current 1 second, reflecting the availability of real-time data.

[0054] (2) Configure a variable data volume time window, and calculate the complementary cumulative distribution function (CCDF) and probability density function (PDF) for the time-delayed data within each window, specifically including: To uncover the long-term distribution characteristics of 5G network latency and determine latency boundaries at different reliability levels, a statistical analysis algorithm based on a variable data volume window is designed: 1) Data window configuration: Set the variable data volume time window length: [100,1000,10000,100000,1000000]; The user specifies the data starting point x, and the system obtains the corresponding data volume of the dataset [x:x+window length].

[0055] 2) The latency distribution calculation includes the computational logic for the Complementary Cumulative Distribution Function (CCDF) and the Probability Density Function (PDF). CCDF calculation generates a CCDF curve by statistically analyzing the proportion of data packets with latency values ​​exceeding a preset threshold, visually displaying the probability of latency exceeding the threshold. PDF calculation uses a kernel density estimation method to smooth the latency distribution, generating a PDF curve and identifying the peak intervals of the latency distribution. Specifically: Complementary Cumulative Distribution Function (CCDF) Calculation: For each data subset, count the number of data packets with a delay greater than the upper limit of the delay interval (e.g., 0~1ms, 1~2ms...), and calculate the CCDF value (CCDF(t) = number of data packets with a delay greater than t / total number of data packets). The CCDF curve can intuitively reflect the probability of delay exceeding the threshold. For example, CCDF(10ms) = 0.001 means that only 0.1% of the data packets have a delay greater than 10ms. Next, the probability density function (PDF) is calculated: the kernel density estimation (KDE) method is used to smooth the latency distribution of the data subset and generate a PDF curve, which reflects the probability density of different latency values ​​and helps maintenance personnel identify the peak range of latency distribution (such as the latency of most data packets being concentrated in 2~3ms).

[0056] 3) Reliability Boundary Calculation: Based on the CCDF calculation results, the delay boundary values ​​corresponding to different reliability levels are determined: The reliability level is defined as the percentage of data packets with delay ≤ boundary value, supporting five levels: 99%, 99.9%, 99.99%, 99.999%, and 99.9999%. For example, when the reliability requirement is 99.9%, find the t value corresponding to CCDF(t) = 0.001 (i.e., only 0.1% of data packets have a delay exceeding t). This t value is the delay boundary under the 99.9% reliability level, which can provide a basis for setting the sampling period in industrial control systems (e.g., the sampling period should be greater than this boundary value to ensure data availability).

[0057] (3) Based on the user-defined sliding step size and basic data volume, a continuous analysis window is generated. The average round-trip time, latency fluctuation coefficient, and industrial adaptability score within each window are calculated to evaluate the long-term trend of network performance, specifically including: First, configure the sliding window parameters. Specifically: the user sets the batch data interaction cycle according to the industrial PLC, with a default of 10,000 sets of data (corresponding to approximately 10 minutes of batch data transmission for the PLC); the amount of data moved by each window is 100 sets of data by default (sliding step size) to ensure the continuity of the analysis results; continuous analysis windows are generated according to the basic data volume + sliding step size, such as the first window being [0:10000] (sets 1 to 10000 of data), the second window being [100:10100] (sets 101 to 10100 of data), until all original data is covered.

[0058] Next, the average value of all RTT values ​​within each sliding window is calculated, which reflects the round-trip transmission efficiency of industrial control commands within that time period; and the ratio of the standard deviation to the average value of the Delay field within the window is calculated (fluctuation coefficient = StdDelay / AvgDelay). The smaller the coefficient, the more stable the delay and the better the industrial control effect. Based on industrial control requirements, scoring rules were set (e.g., AvgDelay ≤ 80% of sampling period earns 30 points, fluctuation coefficient ≤ 0.1 earns 30 points, CCDF (sampling period) ≤ 0.001 earns 40 points), with a total score of 100 points. A score ≥ 80 points indicates suitability for industrial scenarios, 60-80 points indicates basic suitability, and < 60 points indicates need for optimization. Analysis results are presented in the format of window number - industrial suitability score - key indicators to help maintenance personnel identify long-term trends in network performance and optimize 5G network configuration accordingly.

[0059] Step S240: Combining the parameters of the industrial control system, assess the impact of the 5G network on the industrial control system based on the statistical analysis results, and generate a quantitative assessment report.

[0060] The specific implementation includes the following processing logic: (1) Control System Reliability Analysis: Based on the user-input sampling period (Deadline), the proportion of data packets with delay values ​​less than the sampling period is statistically analyzed to calculate reliability indices. Windowing of unreliable data is then marked for root cause analysis. Specifically: In the evaluation module, users input the core parameter of the industrial control system—the sampling period (Deadline). This parameter is determined by the industrial control process (e.g., 10ms for motor control scenarios and 1ms for precision instrument control scenarios). It means that the data must be transmitted within the Deadline before it can be effectively used by the PLC. For the RawData data within each sliding window, filter the available data based on the condition that Delay < Deadline: Available data quantity: The number of data packets within the window that satisfy Delay < Deadline; Total data count: The number of all valid data (TimestampValid=1) in the window; The reliability of the control system within the sliding window = (Available data quantity / Total data quantity) × 100%. This indicator directly reflects the availability of data transmitted by the 5G network to the industrial control system. Evaluation criteria: Industrial control scenarios typically require reliability ≥ 99.99% (i.e., data timeout of 0.01%). If the reliability of a certain window is < 99.99%, it is marked as unreliable and further analysis of the reasons for the latency exceedance is required (such as 5G network congestion or UE device failure). For windows that do not meet reliability standards, extract the data packets with a delay greater than or equal to the deadline, analyze their SequenceNum, NetworkMode, Timestamp and other information to locate the time period or network mode in the timeout data, and assist maintenance personnel in quickly troubleshooting the fault.

[0061] (2) Network Availability Analysis: Based on the watchdog timer count input by the user, calculate the maximum allowable data interruption time, detect whether the time interval between adjacent data packets exceeds the limit, count the number of network disconnection events, and calculate availability metrics. Specifically: Industrial control systems have extremely high requirements for network continuity; even a brief network outage can lead to communication interruptions between PLC master and slave stations, causing equipment downtime. Based on the watchdog mechanism commonly used in industrial scenarios, a network availability assessment mechanism is designed: 1) Watchdog parameter configuration: The user inputs the watchdog count of the industrial control system. This parameter represents the maximum number of consecutive interrupt responses allowed by the slave station by the PLC master station (e.g., 5 times). Based on the sampling period (Deadline), the maximum allowable interrupt time of data is calculated as Deadline × WatchdogCount (e.g., 10ms × 5 = 50ms). That is, if the time interval between two adjacent data packets exceeds 50ms, the PLC master station will determine that the network is disconnected and trigger the protection mechanism (e.g., equipment shutdown).

[0062] 2) Network Disconnection Detection: For the RawData data (sorted by Timestamp) within each sliding window, calculate the time interval between two adjacent data packets (Δt = Timestamp of the next data packet - Timestamp of the previous data packet): If Δt > the maximum allowed data interruption time, it is determined as a network disconnection event, and the start time of the disconnection (Timestamp of the previous data packet) and the recovery time (Timestamp of the next data packet) are recorded; if there are multiple consecutive Δt > the threshold within the window, it is considered as a consecutive network disconnection event (only counted once) to avoid duplicate counting; 3) Availability calculation: Network availability within the sliding window = (Total data volume - Number of network disconnection events) / Total data volume × 100%. This indicator reflects the network's continuous operation capability. Evaluation criteria: For industrial control scenarios, network availability is required to be ≥99.999%. If the availability of the window does not meet the standard, it is necessary to analyze the time period and network mode of the disconnection event and investigate whether it is due to weak 5G network signal or UE device disconnection.

[0063] 4) Disconnection Impact Assessment: For network disconnection events, calculate the disconnection duration (recovery time - start time) and associate it with the industrial control commands within that time period (using SequenceNum) to assess whether the disconnection caused the loss of control commands (e.g., if 3 control commands are not transmitted within the disconnection duration, it may lead to the equipment parameters not being adjusted in time), providing a basis for subsequent optimization.

[0064] (3) Long-term runtime analysis: The continuous runtime within each sliding window is statistically analyzed, and the average continuous runtime, maximum continuous runtime, and variance are calculated to assess the long-term stability of the network. Specifically: Industrial control systems typically require 24 / 7 continuous operation, and the long-term stability of 5G networks directly determines the continuity of industrial production. By statistically analyzing the network's continuous operating time, a long-term performance evaluation mechanism is designed. 1) Continuous running time calculation: For each sliding window, the effective continuous running time within the window is calculated based on the network disconnection detection results: If there is no network disconnection event within the window, the continuous running time = Timestamp of the last data packet within the window - Timestamp of the first data packet within the window (or Deadline × total number of data, with an error of ≤1%); If there is a network disconnection event within the window, the continuous running time = (total window duration - sum of the durations of all disconnection events), ensuring that only the duration of normal network operation is counted.

[0065] 2) Statistics on long-term operating indicators: Average continuous running time: The arithmetic mean of the continuous running times of all sliding windows, reflecting the average continuous duration of long-term network operation; Maximum continuous running time: The maximum continuous running time across all sliding windows, reflecting the network's optimal continuous running capability; Minimum continuous running time: The minimum continuous running time across all sliding windows, reflecting the worst-case scenario of network operation.

[0066] 3) Operational stability assessment: Calculate the variance of the continuous running time of all sliding windows. The smaller the variance, the smaller the difference in the continuous running time of each window, and the more stable the network is in long-term operation. Combine with industrial control requirements, set stability assessment standards (such as variance ≤ 10000ms², i.e., continuous running time fluctuation ≤ 100ms). If the variance exceeds the standard, it is necessary to analyze whether it is caused by 5G network signal fluctuations or periodic interference (such as equipment start-up and shutdown during factory shift changes).

[0067] This application also provides an electronic device, such as... Figure 3 As shown, it includes a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340.

[0068] Memory 330 is used to store computer programs; When the processor 310 executes the program stored in the memory 330, it performs the following steps: The timestamp generator loads a first timestamp before the data packet enters the 5G network and a second timestamp after the data packet leaves the 5G network. The network data analysis computer then performs validity screening based on the collected mirror data and calculates the 5G network latency data based on the timestamp difference of the matched data packet pairs. The network data analysis computer stores the latency data and related feature information into the database according to a preset field structure, and constructs the original measurement data table and the statistical analysis data table. The network data analysis computer performs multi-dimensional statistical analysis on the stored latency data to obtain statistical analysis results. By combining the parameters of the industrial control system and using statistical analysis results, the impact of 5G networks on the industrial control system is assessed, and a quantitative assessment report is generated.

[0069] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0070] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0071] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0072] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0073] The implementation methods and beneficial effects of the various components of the electronic device in the above embodiments for solving the problem can be found in [reference needed]. Figure 2 The steps in the illustrated embodiments are used to implement the electronic device. Therefore, the specific working process and beneficial effects of the electronic device provided in this application will not be repeated here.

[0074] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform any of the 5G network analysis methods for industrial control needs described in the above embodiments.

[0075] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the 5G network analysis methods for industrial control needs described in the above embodiments.

[0076] Those skilled in the art will understand that the embodiments in this application can be provided as methods, systems, or computer program products. Therefore, the embodiments in this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments in this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] This application describes embodiments of methods, apparatus (systems), and computer program products according to embodiments of this application with reference to flowchart illustrations and / or block diagrams. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0078] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0079] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0080] Although preferred embodiments have been described in this application, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of this application.

[0081] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims in this application and their equivalents, then this application also intends to include these modifications and variations.

Claims

1. A 5G network analysis system for industrial control needs, characterized in that, The system includes: a first 5G terminal, a second 5G terminal, a timestamp generator, a PLC master station, a PLC slave station, and a network data analysis computer; The first 5G terminal and the second 5G terminal are configured to send and receive industrial control data packets in a 5G network. The timestamp generator is connected to the first 5G terminal, the second 5G terminal, the PLC master station, the PLC slave station, and the network data analysis computer through multiple Ethernet ports. It is configured to load timestamps and mirror data during the transmission of industrial control data messages. The PLC master station and PLC slave station are configured to generate and receive control commands and status data. The network data analysis computer is configured to acquire mirrored data and perform latency calculations, data storage, multi-dimensional performance analysis, and impact assessments on industrial control systems.

2. The system as described in claim 1, characterized in that, The timestamp generator includes at least five Ethernet ports, wherein: The first Ethernet port is connected to the first 5G terminal and is responsible for forwarding data from the PLC master station to the 5G network; The second Ethernet port is connected to the PLC master station to receive control commands and status query messages; The third Ethernet port is connected to the second 5G terminal and is responsible for forwarding 5G network data to the PLC slave station; The fourth Ethernet port is connected to the PLC slave station, forwarding 5G network messages and receiving status feedback data; The fifth Ethernet port is connected to the network data analysis computer.

3. The system as described in claim 2, characterized in that, The fifth Ethernet port is configured as a bidirectional mirror port, fully replicating the bidirectional data streams between the first and second Ethernet ports, and between the third and fourth Ethernet ports, and connecting to the network data analysis computer.

4. The system as described in claim 1, characterized in that, The PLC master station and the PLC slave station support Profinet or EtherNet / IP industrial Ethernet protocols, with an instruction execution cycle of no more than 1 millisecond; and are directly connected to the corresponding port of the timestamp generator via physical cables.

5. The system as described in claim 1, characterized in that, The system supports both 5G private network architecture and 5G public network architecture. The 5G private network architecture adopts local deployment of user plane functions and 5G LAN technology; The 5G public network architecture adopts a dual-layer secure transmission architecture of VPN and VxLAN. The first 5G terminal and the second 5G terminal are directly connected via LAN under the 5G private network architecture, and configured with VPN tunnel and VxLAN virtual interface binding under the 5G public network architecture.

6. A 5G network analysis method for industrial control needs, characterized in that, The method, applied to the 5G network analysis system for industrial control needs as described in any one of claims 1-5, comprises: The timestamp generator loads a first timestamp before the data packet enters the 5G network and a second timestamp after the data packet leaves the 5G network. The network data analysis computer then performs validity screening based on the collected mirror data and calculates the 5G network latency data based on the timestamp difference of the matched data packet pairs. The network data analysis computer stores the latency data and related feature information into the database according to a preset field structure, and constructs the original measurement data table and the statistical analysis data table. The network data analysis computer performs multi-dimensional statistical analysis on the stored latency data to obtain statistical analysis results. By combining the parameters of the industrial control system and using statistical analysis results, the impact of 5G networks on the industrial control system is assessed, and a quantitative assessment report is generated.

7. The method as described in claim 6, characterized in that, Methods for loading timestamps include: This is achieved by inserting a timestamp field between the preamble and the destination MAC address of the Ethernet frame using a custom field insertion method. The timestamp loading node is precisely located as a key node before the data enters the 5G network and after it leaves the 5G network.

8. The method as described in claim 6, characterized in that, The network data analysis computer performs validity screening based on the collected mirror data, including: Message matching is performed based on a dual-stack mechanism, creating an uplink data stack and a downlink data stack to store uplink messages sent from the PLC master station to the PLC slave station and downlink messages sent from the PLC slave station to the PLC master station, respectively. By comparing the source MAC address, destination MAC address, and data payload of data packets, matching data packet pairs are identified, and unmatched data packets are considered invalid and discarded. A unique sequence number is assigned to each matching data packet pair, and they are sorted by timestamp to ensure consistency in the data processing sequence.

9. The method as described in claim 6, characterized in that, The network data analysis computer performs multi-dimensional statistical analysis on the stored latency data to obtain statistical analysis results, including: Read latency data within the most recent time window from the stored data in seconds, and calculate the real-time average latency, maximum latency, minimum latency, and number of latency exceedances; Configure a variable data volume time window, and calculate the complementary cumulative distribution function and probability density function for the time delay data in each window; A continuous analysis window is generated based on the user-defined sliding step size and basic data volume. The average round-trip time, latency fluctuation coefficient, and industrial adaptability score within each window are calculated to evaluate the long-term trend of network performance.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 6-9.

Images