Bar code scanning data rapid transmission method based on edge calculation

Abstract

The invention discloses a bar code scanning data rapid transmission method based on edge calculation, and the method comprises the following steps: S1, collecting bar code data, and packaging the bar code data into a bar code data package; s2, sending the barcode data packet to an edge computing node; s3, carrying out preprocessing by the edge computing node; s4, constructing and managing a multi-task scheduling queue in the edge computing node; s5, the edge computing node collects network state parameters and generates a network score value; s6, when the network score value is smaller than the network score threshold value, performing compression and encapsulation by adopting a content awareness compression coding mechanism; s7, when the network score value is greater than or equal to the network score threshold value, starting a delay synchronization mechanism; and S8, the cloud server receives, executes processing and returns to the edge computing node. According to the method, edge calculation and content perception compression are fused, efficient scheduling and reliable transmission of bar code data are realized, and the method has the advantages of high real-time performance, weak network resistance and efficient transmission.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data communication and edge computing, and particularly relates to a bar code scanning data fast transmission method based on edge computing. BACKGROUND

[0002] Under the background of the rapid development of the Internet of Things and intelligent manufacturing, bar code scanning technology has been widely used in logistics, warehousing, retail, manufacturing and other industries to realize commodity tracking, inventory management and production process control. Traditional bar code scanning systems usually rely on terminal devices to upload the collected bar code data to the cloud server through wireless networks for centralized processing and storage. However, in actual application, due to the limitations of the processing capacity of terminal devices and the uncertainty of network environment, problems such as high data transmission delay, high packet loss rate, and slow system response often occur, especially in unstable network state, channel quality fluctuation or short-term interruption scenarios, which seriously affect the real-time and reliability of the system.

[0003] In order to solve the above problems, some researches introduce edge computing architecture to sink part of the computing and processing tasks from the cloud to the gateway or node on the network edge, aiming to realize the on-site processing and optimized scheduling of bar code data. Existing edge computing solutions mainly focus on data preprocessing and protocol forwarding level, usually only realize basic field analysis and cache forwarding function, lack of network state perception and dynamic response ability, and also lack of perfect task scheduling mechanism and data compression transmission strategy. In addition, some technologies use simple cache retransmission method when the network is interrupted, which does not fully combine the structural characteristics, redundancy characteristics and time attributes of bar code data, resulting in low cache management efficiency, inflexible retransmission strategy, and still low overall system stability and transmission efficiency.

[0004] Therefore, the existing technology still has the following shortcomings: first, there is a lack of intelligent compression mechanism that integrates the structural characteristics of bar code data, resulting in low bandwidth utilization rate in the transmission process; second, the response mechanism to network state changes is not sensitive, making it difficult to dynamically switch transmission strategies to adapt to complex network environments; third, in the multi-terminal scanning scene, the edge node lacks fine-grained multi-task scheduling capability, which easily causes data congestion and scheduling imbalance. Therefore, it is urgent to propose a bar code data fast transmission method combining edge computing, intelligent compression and adaptive transmission to solve the real-time, robustness and transmission efficiency problems of bar code scanning systems in complex network environments. SUMMARY

[0005] One purpose of the present application is to provide a bar code scanning data fast transmission method based on edge computing, which integrates edge computing and content-aware compression mechanism, constructs a multi-task scheduling queue and adapts to network state, realizes efficient processing and stable uploading of bar code scanning data in weak network environment. Through dynamic selection of compression strategy and delay synchronization mechanism, the real-time response ability and data transmission efficiency of the system are improved, which has the advantages of strong transmission reliability, high resource utilization and good adaptability, and is especially suitable for logistics, e-commerce and manufacturing scenes with high real-time and stability requirements.

[0006] According to the bar code scanning data fast transmission method based on edge computing, the method comprises the following steps:

[0007] S1, a bar code scanning terminal acquires bar code data, and encapsulates the acquired bar code data into a bar code data packet;

[0008] S2, the bar code data packet is sent to an edge computing node deployed on the edge side;

[0009] S3, the edge computing node pre-processes the received bar code data packet;

[0010] S4, according to the timestamp, the bar code scanning terminal identifier and the priority value generated by the system rule in the bar code data packet, a multi-task scheduling queue is constructed and managed in the edge computing node, and parallel processing and load balancing of the bar code data are realized;

[0011] S5, the edge computing node acquires network state parameters, performs weighted calculation on the network state parameters to generate a network score value, and compares the network score value with a network score threshold value;

[0012] S6, when the network score value is less than the network score threshold value, the edge computing node adopts a content-aware compression encoding mechanism to compress and encapsulate the pre-processed bar code data packet, and uploads it to a cloud server through a real-time transmission path;

[0013] S7, when the network score value is greater than or equal to the network score threshold value, a delay synchronization mechanism is started, the pre-processed bar code data is stored in a local cache area, and an asynchronous uploading task queue is constructed, and when the subsequent network score value is lower than the network score threshold value, the content-aware compression encoding is triggered and the data uploading operation is performed;

[0014] S8, the cloud server receives, performs centralized archiving and business processing, and returns the processing result to the edge computing node.

[0015] Optionally, the bar code data packet comprises a timestamp field, a bar code scanning terminal identifier field and a bar code content field.

[0016] Optionally, the S2 specifically comprises:

[0017] S21, the barcode scanning terminal establishes a communication connection to the edge computing node through a pre-configured network access method after completing the barcode data packet encapsulation, the network access method includes short distance wireless communication and local area wired network access, and the communication protocol includes Wi-Fi, Bluetooth, ZigBee and Ethernet;

[0018] S22, the barcode scanning terminal initiates a data packet transmission request to the edge computing node after establishing the connection and attaches device authentication information, and the edge computing node returns a transmission authorization response after verifying the device identity;

[0019] S23, after completing the connection and authentication, the barcode scanning terminal selects the optimal communication channel for barcode data packet sending, the optimal communication channel is selected by real-time measurement of signal-to-noise ratio, and satisfies:

[0020]

[0021] Wherein, C opt represents the optimal communication channel, represents a set of available channels, S j represents the signal power of channel j, N j is the noise power of channel j, C j represents the jth channel in the set of available channels, argmax represents the variable value when the function maximum is taken, and ∈ represents "belongs to";

[0022] S24, after the barcode scanning terminal selects the optimal communication channel, the barcode data packet is sent in fragments according to the set data frame structure, each data frame includes a start flag, a barcode data load, a frame number and a cyclic redundancy check field;

[0023] S25, during the sending process, the barcode scanning terminal monitors the link state in real time, and dynamically adjusts the transmission rate according to the Shannon theory:

[0024]

[0025] Wherein, R t represents the transmission rate, B represents the channel bandwidth, S represents the current signal power, N represents the noise power, and R t represents the current theoretical maximum transmission rate;

[0026] S26, after each data frame of the barcode data packet is sent, the barcode scanning terminal listens to the acknowledgement ACK returned by the edge computing node, if no ACK is received within the set timeout time, the data frame is retransmitted until all frames are successfully sent and the acknowledgement is received;

[0027] S27. After all barcode data packets are successfully transmitted to the edge computing node, the barcode scanning terminal ends the data sending process and releases the communication channel resources.

[0028] Optionally, the S3 specifically includes:

[0029] S31. After receiving the barcode data packet sent by the barcode scanning terminal, the edge computing node writes the barcode data packet into the receiving buffer queue;

[0030] S32. Perform format standardization on each field in the barcode data packet, converting the timestamp field into UTC format;

[0031] S33. The barcode content field is uniformly encoded in UTF-8, and the barcode scanning terminal identification field is standardized in hexadecimal format and mapped to the device unique identification code;

[0032] S34. Perform content integrity check. If the barcode data packet is marked as structurally incomplete data, record it in the abnormal data log and terminate subsequent processing.

[0033] S35, performing outlier removal on the barcode data packet, wherein the judgment rules include the presence of illegal characters in the barcode content field and the field length exceeding the set range. If any of the conditions is met, the barcode data packet is marked as an outlier and removed;

[0034] S36, filtering redundant data for the barcode data packet;

[0035] S37. The barcode data packets that have passed integrity verification, exception elimination, and redundancy filtering are stored in a processing buffer.

[0036] Optionally, the S4 specifically includes:

[0037] S41. After completing the preprocessing of the barcode data packet, the edge computing node extracts the timestamp field, the barcode scanning terminal identification field, and the priority value from the barcode data packet. The three fields form a scheduling tuple. The priority value is generated by the rule engine of the edge node by weighted combination of the normalized score values ​​mapped from the barcode scanning terminal type, the barcode data timestamp, and the barcode content field.

[0038] S42. To achieve parallel processing of barcode data, the edge computing node initializes a multi-task scheduling queue set. Each queue in the multi-task scheduling queue set corresponds to a scheduling processing thread. The number of queues is dynamically set according to the current computing resources and number of tasks of the edge computing node:

[0039]

[0040] Among them, k represents the number of queues, K max Indicates the maximum number of scheduling queues supported by the edge node, Nactive represents the number of tasks currently to be scheduled, C unit represents the number of tasks processed per unit time of each queue, min represents the minimum value between the two, represents the floor value;

[0041] S43, the edge node performs task priority scoring by comprehensively considering time urgency and priority value of the scheduling tuple;

[0042] S44, all tasks to be scheduled are sorted in descending order of scoring value, and are sequentially distributed to the set of multi-task scheduling queues, and high-scoring tasks are preferentially distributed to low-load queues to achieve scheduling load balancing at the task level;

[0043] S45, during task execution, the edge computing node periodically evaluates the running state of each task queue, calculates the current queue length and the average waiting time of the task, and automatically transfers low-priority tasks from high-load queues to queues with high idle degree if the current queue length is greater than the queue length threshold or the average waiting time of the task is greater than the average waiting time threshold.

[0044] S46, when all barcode data tasks to be processed are successfully distributed to each scheduling queue and the scheduling sorting is completed, the system marks the task state as scheduling completed.

[0045] Optionally, the S5 specifically includes:

[0046] S51, the edge computing node periodically collects network state parameters related to data transmission after task scheduling is completed, and the network state parameters include network delay, bandwidth utilization, and packet loss rate;

[0047] S52, the edge node normalizes the network state parameters and maps them to a unified scoring dimension, and calculates the network score value in a weighted manner;

[0048] S53, the edge node compares the network score value with the preset network score threshold, and if the network score value is less than the network score threshold, the current network state is determined to be good, otherwise, it is determined to be an abnormal state;

[0049] S54, the edge node adds the network state judgment result as a scheduling label to each barcode data task to be transmitted, and the scheduling label serves as a trigger basis for selecting a data transmission strategy;

[0050] S55, the network score and judgment process is run in a loop task manner in the edge node to ensure that the network score value reflects the current network actual situation in real time and drives the transmission strategy to change in real time.

[0051] Optionally, the S6 specifically includes:

[0052] S61. When the edge computing node determines that the current network status is good based on the network status score, the edge computing node starts a content-aware compression coding mechanism to perform compression processing on the pre-processed barcode data. The compression strategy is selected based on the characteristics of the barcode content field. The content-aware compression coding mechanism dynamically selects field difference compression, structured field compression, or static dictionary compression strategy based on the structural characteristics and repetition rate of the barcode content field.

[0053] S62. The edge computing node extracts the barcode content field in each barcode data packet and calculates its structural complexity and historical repetition rate. The structural complexity represents the average character length of the current barcode content field, and the historical repetition rate represents the ratio of the current barcode content field to the historical field in the edge node cache set:

[0054]

[0055] Among them, ρ i Represents the historical repetition rate, N match (β i ) indicates the number of cache fields that match the current barcode content field, N total Indicates the total number of cache fields, β i Indicates the current barcode content field;

[0056] S63. Based on the structural complexity and historical repetition rate, the edge node selects a compression strategy:

[0057] If the structural complexity is less than the complexity threshold and the historical repetition rate is greater than the repetition rate threshold, the current field is determined to be simple in structure and highly redundant. A field difference compression strategy is adopted to retrieve the historical barcode content field with the highest similarity from the edge node cache set as a reference field. A difference list between the current barcode content field and the historical barcode content field is generated by character-by-character comparison. Only the difference parts are marked with position and difference records. The edge node cache set caches the barcode content fields that have been received and processed before.

[0058] If the structural complexity is not less than the complexity threshold and the historical repetition rate is not greater than the repetition rate threshold, it is determined to be structurally complex but not repetitive. A structured field compression strategy is adopted to decompose the current barcode content field into a set of subfields. Each subfield corresponds to the field name and field value in the structure definition table, and they are spliced ​​into a compressed field. The structure definition table is a field specification configuration table preset in the edge node, including the identifier, field name, type, maximum length and field value of each subfield in the barcode content field, which is used to support structured parsing and reconstruction of the barcode content field. The edge node performs segmentation, identification and splicing operations on the barcode field content according to the structure definition table:

[0059]

[0060] wherein, denotes the structured field compression result, b j denotes the jthsubfield content, I j denotes the field identifier corresponding thereto, ||denotes the splicing operation in the byte stream, N denotes the number of subfields, denotes the combination operation performed on all subfields, and the structure definition table is preexisting in the edge node configuration file;

[0061] If the current barcode content field matches an existing entry in the local static compression dictionary, a dictionary mapping compression strategy is adopted, the edge node looks up the compression code representation corresponding to the current barcode content field, the local static compression dictionary represents a set of key-value mapping tables predefined and saved in the local storage of the edge computing node, the key represents the barcode content field, and the value represents the compression code representation corresponding to the barcode content field;

[0062] S64, the edge node generates a standardized compression data frame according to the selected compression strategy, the standardized compression data frame includes a field type label, a barcode scanning terminal identifier, a timestamp and a compression content field;

[0063] S65, after the completion of the compression data frame, the edge node adjusts the rate according to the bandwidth condition of the real-time transmission path, and uploads the data frame to the cloud server one by one.

[0064] Optionally, the S7 specifically comprises:

[0065] S71, when the edge computing node according to the real-time computing network score value is not lower than the network score threshold value, it indicates that the current network state is abnormal, the edge node suspends the real-time uploading process, and starts a delay synchronization mechanism to ensure data integrity;

[0066] S72, the edge computing node extracts the timestamp, barcode scanning terminal identifier and collection sequence number in the terminal of each preprocessed barcode data packet, constructs a sortable index key value according to the above information, and writes the barcode data packet into the local cache area in the index mode, the local cache area is a temporary storage module set in the edge node, supports the data organization structure of first-in first-out, and is provided with an upper limit of cache capacity, when the cache area is full, the system automatically deletes the earliest cached but not uploaded data, and records the deletion event in the system exception log at the same time;

[0067] S74, the edge node continuously monitors the network environment in the delay synchronization state, updates the network state score according to the set period, and when the score results of two consecutive times are both lower than the network score threshold value, it is judged that the network is restored to a good state, the system removes the delay synchronization state and prepares to enter the data recovery uploading process;

[0068] S75, the edge node sequentially schedules the buffered data packets into an asynchronous upload task queue according to the time sequence, and the asynchronous upload task queue is sorted in priority of data generation time, ensuring that early generated data is preferentially processed;

[0069] S76, the barcode data packets in the asynchronous upload task queue first perform content-aware compression encoding operation, and the compression mode is consistent with the compression strategy adopted in the normal network state, and after compression, they are re-encapsulated into data frames conforming to the upload protocol;

[0070] S77, the edge node sends the encapsulated data frames to the cloud server one by one, and listens to the upload confirmation response, if the cloud successfully confirms, the corresponding data in the local cache area is cleared, if the upload fails, retry operation is performed, after the number of retries exceeds the maximum retry upper limit set by the system, the data will be marked as "upload failed", and written into the abnormal upload record;

[0071] S78, all data packets completing the delayed upload are uniformly marked with a "delayed synchronization completion" state label and synchronously written into the local upload record log table.

[0072] The beneficial effects of the present application are:

[0073] Firstly, the present application realizes efficient transmission and intelligent scheduling of barcode scanning data in a complex network environment by introducing an edge computing architecture and a content-aware data processing mechanism, overcomes the problems of dependence on cloud processing, large transmission delay and poor robustness in the prior art, and establishes an optimized data interaction channel between the barcode scanning terminal and the edge computing node, so that the present application not only supports real-time acquisition and standardized preprocessing of barcode data, but also realizes multi-task scheduling, network state awareness, intelligent compression and adaptive uploading and other core capabilities on the edge side, so that the system has real-time and high reliability for actual scenes.

[0074] Secondly, the present application realizes high-concurrency scheduling and load balancing of multi-terminal scanning tasks by introducing a scheduling tuple construction mechanism and a multi-task scheduling queue set in the edge computing node, effectively alleviates the processing bottleneck problem caused by dense deployment of terminals, generates a network score value by collecting network delay, bandwidth utilization and packet loss rate and other parameters, the system can dynamically judge the network state, automatically switch real-time uploading and delayed synchronization strategy, enhance the fault tolerance and continuous operation ability in weak network or offline environment, and the content-aware compression mechanism designed based on the content structure characteristics of barcode data enables the edge node to adaptively select differential compression, structured compression or dictionary mapping compression mode, greatly improves the compression rate and transmission efficiency, and reduces the network load.

[0075] Finally, the application improves the data processing capability and uploading efficiency of the barcode scanning system under variable network conditions on the basis of ensuring the integrity and accuracy of the barcode data, has beneficial effects of quick response, flexible structure, intelligent compression, reliable transmission and the like, is suitable for application scenes such as logistics, e-commerce and manufacturing which have high requirements for real-time performance and data integrity, and has good engineering application value. BRIEF DESCRIPTION OF DRAWINGS

[0076] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application, and do not limit the application. In the drawings:

[0077] Fig. 1 A flowchart of a barcode scanning data fast transmission method based on edge computing proposed by the application;

[0078] Fig. 2 An adaptation judgment logic diagram of three compression strategies in a content-aware compression encoding mechanism of a barcode scanning data fast transmission method based on edge computing proposed by the application;

[0079] Fig. 3 A flowchart of edge node execution network state scoring and transmission strategy judgment of a barcode scanning data fast transmission method based on edge computing proposed by the application. DETAILED DESCRIPTION

[0080] The application will now be described in further detail with reference to the drawings. These drawings are simplified schematic diagrams, and only schematically show the basic structure of the application, and thus only show the configurations related to the application.

[0081] REFERENCE Figs. 1-3 A barcode scanning data fast transmission method based on edge computing, comprising the following steps:

[0082] S1, collecting barcode data by a barcode scanning terminal, and packaging the collected barcode data into a barcode data packet;

[0083] S2, sending the barcode data packet to an edge computing node deployed at an edge side;

[0084] S3, pre-processing the received barcode data packet by the edge computing node;

[0085] S4, constructing and managing a multi-task scheduling queue in the edge computing node according to the timestamp, barcode scanning terminal identifier and priority value generated by the system rule in the barcode data packet, to realize parallel processing and load balancing of the barcode data;

[0086] S5, the edge computing node collects network state parameters, performs weighted calculation on the network state parameters to generate a network score value, and compares the network score value with a network score threshold value;

[0087] S6, when the network score value is less than the network score threshold value, the edge computing node adopts a content-aware compression encoding mechanism to compress and encapsulate the preprocessed barcode data packet, and uploads it to the cloud server through a real-time transmission path;

[0088] S7, when the network score value is greater than or equal to the network score threshold value, a delay synchronization mechanism is started, the preprocessed barcode data is stored in a local cache area, and an asynchronous upload task queue is constructed, and when the subsequent network score value is lower than the network score threshold value, the content-aware compression encoding is triggered and the data upload operation is performed;

[0089] S8, the cloud server receives, performs centralized archiving and business processing, and returns the processing result to the edge computing node.

[0090] The application introduces an edge computing architecture in the barcode scanning system, combines key steps such as barcode data acquisition, edge preprocessing, network state judgment, content-aware compression, delay synchronization and asynchronous upload, and constructs a complete data acquisition and transmission closed loop process, which improves the reliability and real-time performance of the system in complex network environment. The method not only improves the success rate and transmission efficiency of barcode data upload, but also enhances the processing capacity and network adaptive ability of the edge node, reduces the dependence on the cloud, is suitable for complex application scenarios such as multi-terminal concurrency, high-frequency scanning, weak network environment, has good expansibility and engineering application value.

[0091] In the embodiment, the barcode data packet includes a timestamp field, a barcode scanning terminal identification field and a barcode content field.

[0092] The application encapsulates the barcode data packet into a structured data packet containing a timestamp field, a terminal identification field and a barcode content field, so that the barcode data has the properties of orderable, identifiable and classifiable before entering the edge computing node, providing clear basis for subsequent scheduling, compression, caching and uploading operations. The structure design improves the standardization and format uniformity of barcode data processing, which is beneficial to the edge node to quickly complete field analysis and rule judgment, reduces the data preprocessing burden, improves the overall data flow efficiency, and enhances the compatibility of barcode data in multi-terminal and multi-type devices, ensuring the stable operation of the system in complex device deployment environment.

[0093] In the embodiment, the S2 specifically includes:

[0094] S21, after the bar code scanning terminal completes the bar code data packet encapsulation, a communication connection to the edge computing node is established through a pre-configured network access mode, the network access mode includes short distance wireless communication and local area wired network access, and the communication protocol includes Wi-Fi, Bluetooth, ZigBee and Ethernet;

[0095] S22, after the bar code scanning terminal establishes the connection, initiates a data packet transmission request to the edge computing node and attaches device authentication information, and the edge computing node returns a transmission authorization response after verifying the device identity;

[0096] S23, after completing the connection and authentication, the bar code scanning terminal selects the optimal communication channel for bar code data packet sending, the optimal communication channel is selected through the signal-to-noise ratio measured in real time, and the following conditions are met:

[0097]

[0098] Wherein, C opt represents the optimal communication channel, represents a set of available channels, S j represents the signal power of channel j, N j is the noise power of channel j, C j represents the jth channel in the set of available channels, and argmax represents the variable value when the function maximum is taken, and ∈ represents “belongs to”;

[0099] S24, after the bar code scanning terminal selects the optimal communication channel, the bar code data packet is sent in fragments according to the set data frame structure, each data frame includes a start flag, a bar code data load, a frame number and a cyclic redundancy check field;

[0100] S25, during the sending process, the bar code scanning terminal monitors the link state in real time, and dynamically adjusts the transmission rate according to the Shannon theory:

[0101]

[0102] Wherein, R t represents the transmission rate, B represents the channel bandwidth, S represents the current signal power, N represents the noise power, and R t represents the current theoretical maximum transmission rate;

[0103] S26, after each data frame of the bar code data packet is sent, the bar code scanning terminal listens to the acknowledgement ACK returned by the edge computing node, if no ACK is received within the set timeout time, the data frame is retransmitted until all frames are successfully sent and the confirmation is received;

[0104] S27, after all the barcode data packets are successfully transmitted to the edge computing node, the barcode scanning terminal ends the data sending process and releases the communication channel resource.

[0105] The application improves the stability and adaptive ability of the data transmission link by designing the complete communication process between the barcode scanning terminal and the edge node, including network access, device authentication, optimal channel selection, dynamic rate adjustment, data frame sending and confirmation mechanism, introduces signal-to-noise ratio judgment and Shannon rate regulation mechanism, effectively reduces the influence of channel interference, improves the success rate of scanning code data upload, frame-level confirmation and retransmission strategy ensures the data integrity, avoids the overall task failure caused by single frame loss, makes the barcode data upload process more reliable, safe and real-time, effectively supports the application demand in weak network, local area network or channel congestion scene.

[0106] In the embodiment, the S3 specifically includes:

[0107] S31, after the edge computing node receives the barcode data packet sent by the barcode scanning terminal, the barcode data packet is written into the receiving buffer queue;

[0108] S32, the format standardization operation is performed on each field in the barcode data packet, and the time stamp field is converted into UTC format;

[0109] S33, the barcode content field is uniformly encoded in UTF-8, and the barcode scanning terminal identification field is standardized into hexadecimal format and mapped into a unique device identification code;

[0110] S34, content integrity verification is performed, if the barcode data packet is marked as incomplete structure data, it is recorded in the abnormal data log and the subsequent processing is stopped;

[0111] S35, the barcode data packet is subjected to outlier rejection, the determination rules include that there are illegal characters in the barcode content field and the field length exceeds the set range, if any condition is met, it is marked as abnormal and rejected;

[0112] S36, the barcode data packet is subjected to redundant data filtering;

[0113] S37, the barcode data packet that passes the integrity verification, abnormal rejection and redundant filtering is stored in the processing buffer.

[0114] The application greatly improves the data quality management capability of the system by setting up a preprocessing procedure in the edge node to perform field standardization, encoding format conversion, integrity check, exception rejection and redundancy filtering on the barcode data packet. The unified data structure processing procedure enables the system to have high fault tolerance and adaptation capability when dealing with data from different terminals and formats. The preprocessing mechanism cleans up redundant and error data before formal data scheduling, reduces resource waste in subsequent task scheduling and uploading phase, improves system stability and overall operation efficiency, and provides a solid data foundation for high-frequency code scanning and large-scale concurrent scenarios.

[0115] In the embodiment, the S4 specifically includes:

[0116] S41, after the edge computing node completes the preprocessing of the barcode data packet, the timestamp field, the barcode scanning terminal identifier field and the priority value in the barcode data packet are extracted, which form a scheduling tuple, and the priority value is generated by weighting and combining the normalized score values respectively mapped by the rule engine of the edge node according to the barcode scanning terminal type, the barcode data timestamp and the barcode content field;

[0117] S42, to realize parallel processing of barcode data, the edge computing node initializes a multi-task scheduling queue set, each queue in the multi-task scheduling queue set corresponds to a scheduling processing thread, and the number of queues is dynamically set according to the current computing resources and the number of tasks of the edge computing node:

[0118]

[0119] Wherein, k represents the number of queues, K max represents the maximum number of scheduling queues supported by the edge node, N active represents the number of tasks to be scheduled at present, C unit represents the number of processing tasks per unit time of each queue, min represents the minimum value between the two, represents rounding down;

[0120] S43, the edge node performs task priority scoring on the scheduling tuple according to the time urgency and the priority value;

[0121] S44, all tasks to be scheduled are sorted in descending order of scoring value, and are sequentially distributed to the multi-task scheduling queue set, and high-scoring tasks are preferentially distributed to low-load queues to realize scheduling load balancing at the task level;

[0122] S45, during the task execution process, the edge computing node periodically evaluates the running state of each task queue, calculates the current queue length and the task average waiting time, and automatically transfers the low-priority task from the high-load queue to the queue with high idle degree if the current queue length is greater than the queue length threshold or the task average waiting time is greater than the average waiting time threshold;

[0123] S46, when all the to-be-processed barcode data tasks are successfully allocated to the scheduling queues and the scheduling sorting is completed, the system marks the task state as scheduling completion.

[0124] The application realizes efficient data parallel processing and resource load balancing by constructing a scheduling tuple composed of a timestamp, a terminal identifier and a priority value on an edge node, combining a multi-task scheduling queue and a scoring allocation mechanism, and dynamically adjusting the queue number and the allocation strategy according to the task urgency and the current computing resource state, thereby relieving the system pressure during high-concurrency code scanning, preventing processing blockage or resource congestion, further optimizing the task execution efficiency and response speed through the priority evaluation and dynamic migration strategy, enhancing the scheduling robustness of the edge node in the face of complex data flow, and ensuring the order and efficiency of the overall data transmission chain.

[0125] In the embodiment, the S5 specifically includes:

[0126] S51, after the task scheduling is completed, the edge computing node periodically collects network state parameters related to data transmission, and the network state parameters include network delay, bandwidth utilization and packet loss rate;

[0127] S52, the edge node normalizes the network state parameters and maps them to a unified scoring dimension, and calculates the network score value in a weighted manner;

[0128] S53, the edge node compares the network score value with a preset network score threshold, if the network score value is less than the network score threshold, the current network state is determined to be good, otherwise, it is determined to be an abnormal state;

[0129] S54, the edge node adds the network state judgment result as a scheduling label to each to-be-transmitted barcode data task, and the scheduling label serves as a trigger basis for selecting a data transmission strategy;

[0130] S55, the network score and judgment process is run in a loop task mode in the edge node, so as to ensure that the network score value reflects the current network actual condition in real time, and drives the transmission strategy to change in real time.

[0131] The present invention introduces a network status parameter monitoring and scoring mechanism in the edge node, which can perceive key network indicators including delay, bandwidth utilization and packet loss rate in real time, and judge the quality of the network through weighted scoring, thereby driving the intelligent switching of subsequent transmission strategies. This mechanism realizes dynamic adaptation of data upload paths, avoids transmission failures caused by blind uploading in weak network environments, and improves the system's response ability to sudden network changes. The combination of network scoring results and scheduling tags enables each data task to have a recognizable network context state, providing technical support for compression strategy selection and upload path judgment, and enhancing the intelligence level of the system.

[0132] In this embodiment, S6 specifically includes:

[0133] S61. When the edge computing node determines that the current network status is good based on the network status score, the edge computing node starts a content-aware compression coding mechanism to perform compression processing on the pre-processed barcode data. The compression strategy is selected based on the characteristics of the barcode content field. The content-aware compression coding mechanism dynamically selects field difference compression, structured field compression, or static dictionary compression strategy based on the structural characteristics and repetition rate of the barcode content field.

[0134] S62. The edge computing node extracts the barcode content field in each barcode data packet and calculates its structural complexity and historical repetition rate. The structural complexity represents the average character length of the current barcode content field, and the historical repetition rate represents the ratio of the current barcode content field to the historical field in the edge node cache set:

[0135]

[0136] Among them, ρ i Represents the historical repetition rate, N match (β i ) indicates the number of cache fields that match the current barcode content field, N total Indicates the total number of cache fields, β i Indicates the current barcode content field;

[0137] S63. Based on the structural complexity and historical repetition rate, the edge node selects a compression strategy:

[0138] If the structural complexity is less than the complexity threshold and the historical repetition rate is greater than the repetition rate threshold, the current field is determined to be simple in structure and highly redundant. A field difference compression strategy is adopted to retrieve the historical barcode content field with the highest similarity from the edge node cache set as a reference field. A difference list between the current barcode content field and the historical barcode content field is generated by character-by-character comparison. Only the difference parts are marked with position and difference records. The edge node cache set caches the barcode content fields that have been received and processed before.

[0139] If the structural complexity is not less than the complexity threshold and the historical repetition rate is not greater than the repetition rate threshold, it is determined that the structure is complex but not repetitive, the structured field compression strategy is adopted, the current barcode content field is disassembled into a subfield set, each subfield corresponds to the field name and field value in the structure definition table, and is spliced into a compressed field, the structure definition table is a preset field specification configuration table in the edge node, including the identifier, field name, type, maximum length and field value of each subfield in the barcode content field, used to support the structured analysis and reconstruction of the barcode content field, and the edge node performs segmentation, identification and splicing operations on the barcode field content according to the structure definition table:

[0140]

[0141] Wherein, The structured field compression result is represented by b j The jth subfield content is represented by I j The field identifier corresponding thereto is represented by ||, the splicing operation in the byte stream is represented by ||, and the number of subfields is represented by N, The combination operation is performed on all subfields, and the structure definition table is preexisting in the edge node configuration file;

[0142] If the current barcode content field matches an existing entry in the local static compression dictionary, the dictionary mapping compression strategy is adopted, the edge node finds the compressed encoding representation corresponding to the current barcode content field, the local static compression dictionary represents a set of key-value mapping tables predefined and saved in the local storage of the edge computing node, the key represents the barcode content field, and the value represents the compressed encoding representation corresponding to the barcode content field;

[0143] S64, the edge node generates a standardized compression data frame according to the selected compression strategy, the standardized compression data frame includes a field type label, a barcode scanning terminal identifier, a timestamp and a compressed content field;

[0144] S65, after the compression data frame is completed, the edge node adjusts the rate according to the bandwidth condition of the real-time transmission path, and uploads the data frame to the cloud server one by one.

[0145] The content-aware compression encoding mechanism proposed in the application is based on the structural features and historical repetition rate of the barcode content field, dynamically selects the field difference compression, structured field compression or local dictionary compression strategy, improves the compression efficiency of the barcode data before transmission, the mechanism can flexibly adapt to the compression method according to the data complexity and redundancy degree, while ensuring the integrity of the data reconstruction, effectively reduces the data amount, reduces the bandwidth occupation during uploading, through the cooperative use of the local structure definition table and the compression dictionary, the edge node has stronger on-site self-processing capacity, so that the system can maintain efficient data flow and stable performance in multiple scenes.

[0146] In this embodiment, S7 specifically includes:

[0147] S71, when the edge computing node determines that the real-time network score is not lower than the network score threshold, it means that the current network state is abnormal, the edge node suspends the real-time uploading process, and starts a delayed synchronization mechanism to ensure data integrity;

[0148] S72, the edge computing node extracts the timestamp, barcode scanning terminal identifier and collection sequence number in the terminal for each pre-processed barcode data packet, constructs a sortable index key value according to the above information, and writes the barcode data packet into the local cache area in the index mode, the local cache area is a temporary storage module set in the edge node, supports first-in first-out data organization structure, and is provided with an upper limit of cache capacity, when the cache area is full, the system automatically deletes the earliest cached but not uploaded data, and records the deletion event in the system exception log at the same time;

[0149] S74, the edge node continuously monitors the network environment in the delayed synchronization state, updates the network state score according to the set period, and when the score results of two consecutive times are both lower than the network score threshold, it is judged that the network returns to good state, the system removes the delayed synchronization state and prepares to enter the data recovery uploading process;

[0150] S75, the edge node schedules the cache data packet into the asynchronous uploading task queue in sequence according to the time sequence, the asynchronous uploading task queue is sorted according to the data generation time as the priority, ensuring that the early generated data is processed first;

[0151] S76, the barcode data packet entering the asynchronous uploading task queue first performs content-aware compression encoding operation, the compression mode is consistent with the compression strategy adopted in the normal network state, and after compression, it is re-encapsulated into a data frame conforming to the uploading protocol;

[0152] S77, the edge node sends the encapsulated data frame to the cloud server one by one, and listens to the uploading confirmation response, if the cloud end successfully confirms, the corresponding data in the local cache area is cleared, if the uploading fails, the retry operation is performed, after the number of retries exceeds the maximum retry upper limit set by the system, the data will be marked as "uploading failed", and written into the abnormal uploading record;

[0153] S78, all data packets completing delayed uploading are uniformly marked with "delayed synchronization completion" state label, and written into the local uploading record log table synchronously.

[0154] The application automatically triggers a delay synchronization mechanism when the network state is abnormal, buffers barcode data locally at the edge node, and through the construction of a cache index and a task queue, a periodic monitoring and recovery upload process, realizes stable storage and intelligent retransmission of temporarily interrupted data, and an asynchronous upload task queue ensures the order and integrity of the data after the network is restored. The upload state monitoring and retry mechanism further enhances the robustness and self-healing ability of the system. Through unified log recording and exception marking management, the system can realize full-process tracing and error backtracking of delayed upload tasks, effectively solve the problem of data loss and upload failure under weak network conditions, and improve the overall system reliability.

[0155] Embodiment 1

[0156] In order to verify the feasibility of the application in implementation, the application is applied to the barcode scanning data transmission scene of a certain intelligent warehouse logistics center. The logistics center processes more than 100,000 in-out warehouse barcode data per day, and deploys more than 120 barcode scanning terminals. The terminals are distributed in various picking areas, sorting areas and shipping areas. Some areas have weak signal coverage, and there are problems such as unstable network, high delay and occasional interruption. The traditional barcode scanning system directly uploads data to the cloud, which often causes data upload delay, scanning terminal lag, data loss and other problems during peak periods, seriously affecting operation efficiency and data accuracy.

[0157] In this embodiment, edge computing nodes are deployed at the edge of the warehouse, and a lightweight client module that communicates with the edge node is embedded in all barcode scanning terminals. After each terminal collects a barcode, it encapsulates the data into a standardized data packet in real time, which contains timestamp, terminal number and barcode content fields. The data packet is sent to the edge computing node through the local area network.

[0158] After receiving the data, the edge node first preprocesses the data packet, including timestamp unification (converted to UTC format), field legality check, content format standardization and abnormal data elimination. The system processes an average of more than 7500 data per hour, with a processing delay of less than 3ms.

[0159] Subsequently, the edge node establishes a multi-task scheduling queue based on terminal identification, timestamp and dynamic priority value, realizes parallel data processing, and dynamically expands the number of parallel queues from 4 to 10 during peak periods (e.g. 9:30-11:00 am and 15:00-17:00 pm). The average scheduling waiting time is reduced by about 41%, ensuring the orderly processing of high-concurrency scanning tasks.

[0160] To address the problem of network status fluctuations, the edge node collects network status parameters every 30 seconds and performs scoring. When the network score value is lower than the set threshold, the edge node automatically enables the content-aware compression mechanism and dynamically selects the appropriate compression strategy (field difference compression, structured field compression or dictionary compression) based on the structural complexity and repetition rate of the barcode content field.

[0161] If the network score is higher than the threshold, the system automatically starts the delayed synchronization mechanism, caches the pre-processed data in the local cache, and builds an asynchronous upload queue.

[0162] In the cloud, all data is received and used for order archiving, inventory updates, and business data reflow. In order to verify the performance of the present invention in implementation, experimental tests were conducted and the performance was compared with traditional solutions in the intelligent warehousing scenario.

[0163] Table 1 Performance comparison of the present invention and traditional solutions in intelligent warehousing scenarios

[0164]

[0165]

[0166] In terms of transmission efficiency, the system of the present invention can process and upload 15,800 barcode data items per hour, an increase of approximately 41.1% compared to the 11,200 items of the original system. Thanks to the introduction of the content-aware compression mechanism, the average size of a single barcode data packet is compressed from 620 bytes to 358 bytes, with a compression rate increased by more than 42%, effectively reducing the network load. In addition, the response delay of the barcode terminal during peak hours is reduced from an average of 1,130 milliseconds in the original system to 625 milliseconds, indicating that the present invention improves the terminal user experience and ensures the real-time nature of the operation process.

[0167] In terms of system stability, the present invention achieves robust data protection in a weak network environment by introducing a network scoring mechanism and a delay synchronization strategy. The original system has a data packet loss rate of 0.37% when the network is unstable, while the present invention achieves 0 packet loss. At the same time, the upload failure rate is reduced from 3.2% of the original system to 1.58%, and the number of retries is also reduced by half, indicating that the robustness of the system is enhanced. Through the asynchronous upload queue and caching mechanism, the present invention can ensure that all data is ultimately reliably delivered to the cloud.

[0168] In terms of processing power, this invention deploys preprocessing and task scheduling modules at edge nodes, reducing the average data preprocessing time from 9.7ms to 5.3ms and optimizing the average task scheduling latency from 188ms to 106ms. These improvements significantly enhance the system's responsiveness and throughput when facing high-concurrency code scanning tasks.

[0169] In terms of network adaptability, the application realizes the function of automatically scoring the network state every 30 seconds and dynamically adjusting the uploading strategy, which is a key capability that the original system does not have. At the same time, the edge node is configured with a local buffer with a maximum cache of 60,000 data, which can still guarantee that all code scanning tasks are not lost during a 15-minute weak network duration. With the help of an asynchronous uploading mechanism, the uploading success rate of the cache task reaches 100%, realizing full data backhaul under weak network.

[0170] The above is only the preferred specific embodiment of the application, but the protection scope of the application is not limited thereto, any person skilled in the art can make equivalent replacement or change within the technical range disclosed by the application according to the technical solution and the inventive concept of the application, which should be covered within the protection scope of the application.

Claims

1. A method for rapid transmission of barcode scanning data based on edge computing, characterized in that: The steps include: S1. The barcode scanning terminal collects barcode data and encapsulates the collected barcode data into a barcode data packet; S2. Send the barcode data packet to the edge computing node deployed on the edge side; S3, the edge computing node pre-processes the received barcode data packet; S4. Build and manage a multi-task scheduling queue in the edge computing node based on the timestamp in the barcode data packet, the barcode scanning terminal identifier, and the priority value generated by the system rules; S5. The edge computing node collects network status parameters, performs weighted calculation on the network status parameters to generate a network score value, and compares the network score value with the network score threshold; S6. When the network score value is less than the network score threshold, the edge computing node compresses and encapsulates the pre-processed barcode data packet using a content-aware compression coding mechanism and uploads it to the cloud server through a real-time transmission path; S7. When the network score value is greater than or equal to the network score threshold, the delayed synchronization mechanism is activated to store the pre-processed barcode data in the local cache area and to build an asynchronous upload task queue. When the subsequent network score value is lower than the network score threshold, the content-aware compression encoding is triggered and the data upload operation is performed; S8. The cloud server receives the data, performs centralized archiving and business processing, and returns the processing results to the edge computing node.

2. A method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The barcode data packet includes a timestamp field, a barcode scanning terminal identification field and a barcode content field.

3. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S2 specifically includes: S21. After completing the barcode data packet encapsulation, the barcode scanning terminal establishes a communication connection to the edge computing node through a preconfigured network access method; S22. After establishing a connection, the barcode scanning terminal actively initiates a data packet transmission request to the edge computing node with device authentication information. The edge computing node returns a transmission authorization response after verifying the device identity. S23. After completing the connection and authentication, the barcode scanning terminal selects the optimal communication channel to send the barcode data packet; S24. After selecting the optimal communication channel, the barcode scanning terminal segments and sends the barcode data packet according to the set data frame structure. Each data frame includes a start flag, a barcode data payload, a frame sequence number, and a cyclic redundancy check field. S25. During the transmission process, the barcode scanning terminal monitors the link status in real time and dynamically adjusts the transmission rate based on Shannon's theory; S26. After each data frame of the barcode data packet is sent, the barcode scanning terminal monitors the confirmation response ACK returned by the edge computing node. If the ACK is not received within the set timeout period, the data frame is resent until all frames are successfully sent and confirmed; S27. After all barcode data packets are successfully transmitted to the edge computing node, the barcode scanning terminal ends the data sending process and releases the communication channel resources.

4. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S3 specifically includes: S31. After receiving the barcode data packet sent by the barcode scanning terminal, the edge computing node writes the barcode data packet into the receiving buffer queue; S32. Perform format standardization on each field in the barcode data packet, converting the timestamp field into UTC format; S33. The barcode content field is uniformly encoded in UTF-8, and the barcode scanning terminal identification field is standardized in hexadecimal format and mapped to the device unique identification code; S34. Perform content integrity check. If the barcode data packet is marked as structurally incomplete data, record it in the abnormal data log and terminate subsequent processing. S35, performing outlier removal on the barcode data packet, wherein the judgment rules include the presence of illegal characters in the barcode content field and the field length exceeding the set range. If any of the conditions is met, the barcode data packet is marked as an outlier and removed; S36, filtering redundant data for the barcode data packet; S37. The barcode data packets that have passed integrity verification, exception elimination, and redundancy filtering are stored in a processing buffer.

5. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S4 specifically includes: S41. After completing the preprocessing of the barcode data packet, the edge computing node extracts the timestamp field, the barcode scanning terminal identification field, and the priority value from the barcode data packet. The three fields form a scheduling tuple. The priority value is generated by the rule engine of the edge node by weighted combination of the normalized score values ​​mapped from the barcode scanning terminal type, the barcode data timestamp, and the barcode content field. S42. To implement parallel processing of barcode data, the edge computing node initializes a multi-task scheduling queue set, where each queue in the multi-task scheduling queue set corresponds to a scheduling processing thread, and the number of queues is dynamically set based on the current computing resources and number of tasks of the edge computing node; S43, the edge node performs task priority scoring based on the time urgency and priority value of the scheduling tuple; S44. Sort all tasks to be scheduled by their scores from high to low, and assign them to a multi-task scheduling queue set in order, with high-scoring tasks being assigned to low-load queues first, to achieve task-level scheduling load balancing. S45. During task execution, the edge computing node periodically evaluates the running status of each task queue, calculates the current queue length and the average waiting time of the task, and automatically transfers low-priority tasks from the high-load queue to the queue with high idleness if the current queue length is greater than the queue length threshold or the average waiting time of the task is greater than the average waiting time threshold. S46. When all pending barcode data tasks are successfully assigned to each scheduling queue and the scheduling order is completed, the system marks the task status as scheduling completed.

6. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S5 specifically includes: S51. After the task scheduling is completed, the edge computing node periodically collects network status parameters related to data transmission, wherein the network status parameters include network delay, bandwidth utilization, and packet loss rate; S52. The edge node normalizes the network status parameters, maps them to a unified scoring dimension, and calculates the network score value in a weighted manner; S53. The edge node compares the network score value with a preset network score threshold. If the network score value is less than the network score threshold, the current network state is determined to be good; otherwise, it is determined to be abnormal. S54. The edge node attaches the network status judgment result as a scheduling tag to each barcode data task to be transmitted, and the scheduling tag serves as a trigger basis for selecting a data transmission strategy; S55. In the edge node, the network scoring and judgment process is run in a cyclic task manner.

7. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S6 specifically includes: S61. When the edge computing node determines that the current network status is good based on the network status score, the edge computing node starts a content-aware compression coding mechanism to perform compression processing on the pre-processed barcode data. The compression strategy is selected based on the characteristics of the barcode content field. The content-aware compression coding mechanism dynamically selects field difference compression, structured field compression, or static dictionary compression strategy based on the structural characteristics and repetition rate of the barcode content field. S62. The edge computing node extracts the barcode content field from each barcode data packet and calculates its structural complexity and historical repetition rate. The structural complexity represents the average character length of the current barcode content field, and the historical repetition rate represents the ratio of the current barcode content field to the historical field in the edge node cache set. S63. Based on the structural complexity and historical repetition rate, the edge node selects a compression strategy: If the structural complexity is less than the complexity threshold and the historical repetition rate is greater than the repetition rate threshold, the current field is determined to be simple in structure and highly redundant. A field difference compression strategy is adopted to retrieve the historical barcode content field with the highest similarity from the edge node cache set as a reference field. A difference list between the current barcode content field and the historical barcode content field is generated by character-by-character comparison. Only the difference parts are marked with position and difference records. The edge node cache set caches the barcode content fields that have been received and processed before. If the structural complexity is not less than the complexity threshold and the historical repetition rate is not greater than the repetition rate threshold, it is determined to be structurally complex but not repetitive. A structured field compression strategy is adopted to decompose the current barcode content field into a set of subfields. Each subfield corresponds to a field name and field value in the structure definition table, and the subfields are spliced ​​into a compressed field. The structure definition table is a field specification configuration table preset in the edge node, including the identifier, field name, type, maximum length and field value of each subfield in the barcode content field, which is used to support structured parsing and reconstruction of the barcode content field. The edge node performs segmentation, identification and splicing operations on the barcode field content according to the structure definition table. If the current barcode content field matches an existing entry in the local static compression dictionary, a dictionary mapping compression strategy is adopted, and the edge node searches for the compressed encoding representation corresponding to the current barcode content field. The local static compression dictionary represents a set of key-value mapping tables predefined and stored in the local storage of the edge computing node. The key represents the barcode content field, and the value represents the compressed encoding representation corresponding to the barcode content field. S64. The edge node generates a standardized compressed data frame according to the selected compression strategy. The standardized compressed data frame includes a field type tag, a barcode scanning terminal identifier, a timestamp, and a compressed content field. S65. After the data frames are compressed, the edge node adjusts the rate according to the bandwidth status of the real-time transmission path and uploads the data frames one by one to the cloud server.

8. The method for rapid transmission of barcode scanning data based on edge computing according to claim 1, characterized in that: The S7 specifically includes: S71. When the edge computing node calculates the network score value based on real-time calculation and it is not lower than the network score threshold, it indicates that the current network status is abnormal. The edge node suspends the real-time upload process and starts the delayed synchronization mechanism to ensure data integrity. S72. The edge computing node extracts the timestamp, barcode scanning terminal identifier, and acquisition sequence number in the terminal from each pre-processed barcode data packet, constructs a sortable index key value based on the above information, and writes the barcode data packet into the local cache using this index; S74. The edge node continuously monitors the network environment in the delayed synchronization state and updates the network status score according to the set period. When the score results are lower than the network score threshold for two consecutive times, the network is judged to have recovered to a good state. The system cancels the delayed synchronization state and prepares to enter the data recovery upload process. S75. The edge node schedules the cached data packets into the asynchronous upload task queue in the order of their time. The asynchronous upload task queue is sorted by the time of data generation to ensure that data generated earlier is processed first. S76: The barcode data packet entering the asynchronous upload task queue first performs a content-aware compression encoding operation, using the same compression strategy as that used in a normal network state. After the compression is completed, the data packet is repackaged into a data frame that complies with the upload protocol. S77. The edge node sends the encapsulated data frames one by one to the cloud server and listens for the upload confirmation response. If a successful confirmation is received from the cloud, the corresponding data is cleared from the local cache. If the upload fails, a retry operation is performed. If the number of retries exceeds the maximum retry limit set by the system, the data will be marked as "upload failed" and an abnormal upload record will be written. S78. All data packets that have completed delayed upload will be uniformly marked with a "delayed synchronization completed" status label and synchronously written into the local upload record log table.

Images