An Image Communication Control Method and System for Cloud-Edge Transmission

Abstract

The present invention provides an image communication control method and system for cloud-edge transmission, which relates to the technical field of cloud-edge image data transmission. The specific steps are as follows: Through merge sorting, the image communication data sent by the network port is merged into a complete data packet sequence according to the memory size, an edge node threshold is set, the data packets are divided into simple and complex sub-sequences, and then transmitted to the edge node and the cloud server. The network latency, bandwidth utilization rate, and real-time bandwidth usage data of the edge node and the cloud server are collected to generate an edge node coefficient and a bandwidth coefficient. The two types of data packets are unitized and calibrated, and the priority is set according to the bandwidth occupancy ratio of the complex data packets in the server, and then sent to the data buffer. A screening mechanism is set up to screen the data packets, and the two types of data packets after screening and processing form a data chain according to the serial number and identification. The present invention effectively combines the optimized utilization of edge computing and cloud resources, providing reliable technical support and guarantee for real-time image transmission.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud-edge image data transmission, and specifically to an image communication control method and system for cloud-edge transmission. Background Art

[0002] At present, with the advent of the big data era, image communication plays an increasingly important role in work and life. With the application of image communication in various fields, it has greatly facilitated people's lives. For example, through real-time video calls and conferences, image communication enables people to communicate face-to-face remotely, helping to reduce communication barriers caused by geographical distances, improve work efficiency and quality. Moreover, image communication technology combined with big data and artificial intelligence has promoted the development of smart cities and industrial automation. For example, in fields such as traffic management systems, smart grids, and intelligent manufacturing, system operation efficiency is optimized through real-time data collection and image analysis, enhancing the intelligent level of urban management and industrial production.

[0003] In the existing publicly disclosed technology, the publication number is CN112019940B. The original image data collected by an image acquisition device is received by an image uploading device, and after adding first marking information to the original image data, it is uploaded to a cloud server. The first marking information includes the number of a specified image receiving device. The image receiving device obtains the image data belonging to it from the cloud server through its number, and then completes the upload of the image data.

[0004] Deficiencies of the prior art:

[0005] Currently, the existing technology mainly performs information marking processing on the original image data by a receiving device, and then the cloud server judges the received image data sent by the device. According to whether the received image data has compliant marking information, if so, the image data is temporarily stored. In the transportation of image data, common situations where marking information is lost include increased bandwidth utilization and increased network latency. The loss of the marking signal will cause the cloud server to be unable to identify the transmitted image data, resulting in incomplete uploaded image content.

[0006] The processing capacity and bandwidth of the cloud server are not infinite. When a large amount of image transmission data is involved, if all the data packets included are processed by the cloud server, it may lead to performance bottlenecks and response delays, which will affect the real-time performance of the service and the user experience. Especially for applications that require high-speed data processing, such as real-time video streams and image data transmission, if the entire service depends on the cloud server to process a large number of data packets, once the server fails or the service is interrupted, it may cause the entire system to shut down and the service to be unavailable. This kind of dependence increases the risk of the system, especially in the case where there is no effective backup of image data.

[0007] Therefore, it is necessary to provide an image communication control method and system for cloud-edge transmission to solve the above problems.

[0008] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0009] The purpose of the present invention is to provide an image communication control method and system for cloud-edge transmission to solve the problems raised in the above background art.

[0010] To achieve the above purpose, the present invention provides the following technical solutions:

[0011] An image communication control method for cloud-edge transmission, the specific steps include:

[0012] Step 1: Collect real-time image communication data sent by the network port, merge and sort the received data packets according to their memory sizes, and merge the received scattered data packets into a complete data packet sequence;

[0013] Step 2: Set up an edge node threshold to perform segmentation preprocessing on the merged data packet sequence, divide the data packet sequence into a simple data packet sub-sequence and a complex data packet sub-sequence according to the number of bytes occupied, and transmit the simple data packet sub-sequence into the edge node, and transmit the complex data packet sub-sequence into the cloud server;

[0014] Step 3: Collect the time required for the simple data packet to be transmitted into the edge node, calculate the bandwidth utilization rate, generate an edge node coefficient using the network delay data and the bandwidth utilization rate, calibrate and unitize the simple data packet sub-sequence using the edge node coefficient, and send the generated standard simple data packet to the data processing buffer;

[0015] Step 4: Generate a bandwidth coefficient according to the bandwidth used by the network port to transmit data and the time required for the complex data packet to be transmitted into the cloud server, calibrate and unitize the complex data packet sub-sequence using the bandwidth coefficient, and generate a standard complex data packet;

[0016] Step 5: Arrange the priorities according to the bandwidth sizes occupied by the standard complex data packets in the cloud server, and the priority of the standard complex data packet with a smaller bandwidth occupancy is higher than that of the standard complex data packet with a larger bandwidth occupancy, and send each standard complex data packet to the data processing buffer in turn according to the priority;

[0017] Step 6: Set up a screening mechanism to calculate the bandwidth ratio of standard simple data packets and standard complex data packets, which is used to identify and screen the headers of the unitized data packets entering the data processing buffer subsequently. Screen out the data packets with incomplete information and missing data and resend them to the network port for data supplement processing. Form a data chain by combining the remaining data packets in the buffer according to the sequence number and identification, and send them to the cloud server to achieve image communication at the terminal.

[0018] Further, in Step 1, perform merge sort on the received data packets according to their memory sizes, and integrate the received scattered data packets into a complete data packet sequence. The specific method is as follows: Decompose the list of data packets to be sorted into several small subsequences until each subsequence contains only one unit data packet. Then gradually merge the adjacent subsequences to obtain larger ordered subsequences. During the merging process, it is necessary to compare the byte sizes of the unit data packet elements in the two subsequences and merge them into a new ordered sequence in ascending order of the occupied bytes. Repeat the above steps until all the subsequences are merged into a complete ordered sequence.

[0019] Further, in Step 2, set up an edge node byte threshold to perform segmentation preprocessing on the merged data packet sequence. Divide the data packet sequence into a simple data packet subsequence and a complex data packet subsequence according to the occupied bytes, and transmit the simple data packet subsequence into the edge node and the complex data packet subsequence into the cloud server. The specific method is as follows: Consider the data packets with bytes greater than the node byte threshold as complex data packets, and the data packets with bytes not greater than the node byte threshold as simple data packets. All the simple data packets form a simple data packet subsequence, and all the complex data packets form a complex data packet subsequence. The specific logic is:

[0020]

[0021] if S i ≤T, S i ∈X jd

[0022] if S i >T, S i ∈X fz

[0023] where S i represents the number of bytes occupied by the i-th data packet in the data packet sequence, T represents the preset byte threshold, X jd represents the simple data packet subsequence, X fz represents the complex data packet subsequence, i is a positive integer, i = 1, 2, 3... N, and N represents the total number of all data packets in the data packet sequence.

[0024] Further, in step 3, collect the network latency data in the edge computing node, calculate the bandwidth utilization rate, and generate an edge node coefficient using the network latency data and the bandwidth utilization rate. The specific logic is as follows:

[0025]

[0026] Among them, P kd represents the bandwidth utilization rate in data transportation, M u represents the bandwidth used by the network port to transmit data, M o represents the total bandwidth of the network port. represents the time required for the j-th simple data packet to be transmitted into the edge node. j represents the index of the simple data packet, and j = 1, 2, 3... n. C by represents the edge node coefficient, and n represents the number of simple data packets in the simple data packet subsequence;

[0027] Further, in step 3, use the generated edge node coefficient to calibrate the simple data packet subsequence and perform unitization processing to generate a standard simple data packet. The specific logic is as follows:

[0028]

[0029] Among them, represents the number of bytes occupied by the j-th standard simple data packet, represents the number of bytes occupied by the j-th simple data packet in the simple data packet subsequence.

[0030] Further, in step 4, use the bandwidth used by the network port to transmit data and the monitored network latency data to generate a bandwidth coefficient. The specific logic is as follows:

[0031]

[0032] Among them, represents the time required for the k-th complex data packet to be transmitted into the cloud server. k represents the index of the complex data packet, and k = 1, 2, 3... N - n. G dk represents the bandwidth coefficient, and N - n represents the number of complex data packets in the complex data packet subsequence.

[0033] Further, in step 4, use the bandwidth coefficient to calibrate the complex data packet subsequence and perform unitization processing to generate a standard complex data packet. The specific logic is as follows:

[0034] Among them, represents the number of bytes occupied by the k-th standard complex data packet, represents the number of bytes occupied by the k-th complex data packet in the complex data packet subsequence.

[0035] Further, in step 5, according to the bandwidth size occupied by each complex data packet in the cloud server, the arrangement priority is set from the complex data packet with a smaller bandwidth ratio to the complex data packet with a larger bandwidth ratio. The specific method is as follows: measure the bandwidth of the data packets to be entered into the buffer, sort them according to their bandwidth ratios, and the data packets consuming less bandwidth resources will be given higher priorities. The data packets with higher priorities will be processed first to ensure that they quickly pass through the system;

[0036] A screening mechanism is established to calculate the bandwidth ratio between the standard simple data packet and the standard complex data packet. The specific logic is as follows:

[0037]

[0038] where F dkb represents the bandwidth ratio between the standard simple data packet and the standard complex data packet, represents the average bandwidth value of the standard simple data packet, represents the average bandwidth value of the standard complex data packet;

[0039] Further, in step 6, the bandwidth ratio is used to screen the simple data packet subsequence and the complex data packet subsequence, and the data packets with incomplete information and missing data are screened out and resent to the network port for data supplement processing. The specific method is as follows: use the header recognition to determine the type of the data packet, compare its bandwidth with the average bandwidth value of the data packets of its corresponding type. If both are less than the corresponding average bandwidth value, it is determined as a data packet with incomplete information and missing data, and the missing and damaged parts are recorded. These data packets are sent back to the original network port to request data repair. During the waiting for retransmission, these data packets are temporarily reserved in a dedicated waiting area to ensure that all the data packets after supplementary processing re-enter the data processing buffer completely and correctly. After screening, the two types of data packets are sent to the cloud server, and a data chain is formed according to the combination of the sequence number and the identifier of each data packet to realize image communication at the terminal.

[0040] The present invention further provides an image communication control system for cloud-edge transmission. The above control system is used to execute the above image communication control method and includes:

[0041] A data packet acquisition and sorting module, which is used to acquire the real-time image communication data sent by the network port, merge and sort the received data packets according to their memory sizes, and merge the received scattered data packets into a complete data packet sequence;

[0042] The threshold node segmentation processing module is used to set up edge node thresholds to perform segmentation preprocessing on the merged data packet sequence, divide the data packet sequence into simple data packet sub-sequences and complex data packet sub-sequences according to the occupied bytes, transmit the simple data packet sub-sequences into the edge nodes, and transmit the complex data packet sub-sequences into the cloud server;

[0043] The unitary calibration processing module is used to collect network delay data in the edge computing nodes, calculate the bandwidth utilization rate, generate edge node coefficients using the network delay data and the bandwidth utilization rate, calibrate and perform unitary processing on the simple data packet sub-sequences using the edge node coefficients, and send the generated standard simple data packets to the data processing buffer;

[0044] The coefficient optimization module is used to generate a bandwidth coefficient according to the bandwidth used for data transmission by the network port and the time required for the complex data packets to be transmitted into the cloud server, calibrate and perform unitary processing on the complex data packet sub-sequences using the bandwidth coefficient, and generate standard complex data packets;

[0045] The bandwidth arrangement module is used to arrange the priorities according to the bandwidth sizes occupied by the standard complex data packets in the cloud server, and the priority of the standard complex data packets with a smaller bandwidth occupancy ratio is higher than that of the standard complex data packets with a larger bandwidth occupancy ratio, and each standard complex data packet is sequentially sent to the data processing buffer according to the priority;

[0046] The data packet screening and serialization module is used to set up a screening mechanism, calculate the bandwidth ratio of the standard simple data packets and the standard complex data packets, identify and screen the headers of the unitary data packets entering the data processing buffer subsequently, screen out the data packets with incomplete information and missing data and resend them to the network port for data supplement processing, form a data chain by combining the remaining data packets in the buffer according to the sequence number and the identifier, and send them to the cloud server to achieve image communication at the terminal.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] By merging and sorting the received data packets, the present invention can effectively reduce the impact of network congestion caused by data packet fragmentation, divide the data packet sequence into simple data packet sub-sequences and complex data packet sub-sequences according to their byte counts, quickly process the simple data packets at the edge nodes to reduce the data transmission delay in the network, and perform bandwidth optimization processing on the complex data packets through the cloud server to balance the bandwidth utilization rate and ensure the efficient transmission and processing of data;

[0049] The present invention also ensures that important data can be preferentially processed and transmitted by establishing a data packet priority management and screening mechanism. The establishment of the screening mechanism helps to identify and process data packets with incomplete information or missing data, improving the integrity and accuracy of the data. The processed and optimized standard data packets are formed into a data chain according to the serial number and identification, providing a continuous and orderly data stream for the terminal device, which helps to achieve stable and high-quality real-time image communication, significantly improving the efficiency, reliability and real-time performance of the real-time image communication system, and providing users with a better usage experience and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the overall method flow of the present invention.

[0051] Figure 2 It is a schematic diagram of the system module flow of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] To make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to specific embodiments.

[0053] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those of ordinary skill in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before the term cover the elements or objects listed after the term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0054] Embodiment:

[0055] Please refer to Figure 1 : An image communication control method for cloud-edge transmission, including:

[0056] Step 1: Collect real-time image communication data sent from the network port, merge and sort the received data packets according to their memory sizes, and merge the received scattered data packets into a complete data packet sequence;

[0057] Step 2: Set the edge node threshold to perform segmentation preprocessing on the merged data packet sequence. Divide the data packet sequence into a simple data packet subsequence and a complex data packet subsequence according to the number of bytes occupied, and transmit the simple data packet subsequence into the edge node and the complex data packet subsequence into the cloud server;

[0058] Step 3: Collect the time required for the simple data packet to be transmitted into the edge node, calculate the bandwidth utilization rate, generate an edge node coefficient using the network delay data and the bandwidth utilization rate, calibrate and unitize the simple data packet subsequence using the edge node coefficient, and send the generated standard simple data packet to the data processing buffer;

[0059] Step 4: Generate a bandwidth coefficient based on the bandwidth used by the network port to transmit data and the time required for the complex data packet to be transmitted into the cloud server, calibrate and unitize the complex data packet subsequence using the bandwidth coefficient, and generate a standard complex data packet;

[0060] Step 5: Arrange the priorities according to the bandwidth size occupied by the standard complex data packet in the cloud server, and the priority of the standard complex data packet with a smaller bandwidth occupancy is higher than that of the standard complex data packet with a larger bandwidth occupancy. Send each standard complex data packet to the data processing buffer in sequence according to the priority;

[0061] Step 6: Set up a screening mechanism, calculate the bandwidth ratio of the standard simple data packet and the standard complex data packet, which is used to identify and screen the headers of the unitized data packets entering the data processing buffer subsequently, screen out the data packets with incomplete information and missing data and resend them to the network port for data supplementation processing, form a data chain by combining the remaining data packets in the buffer according to the sequence number and the identifier, and send them to the cloud server to achieve image communication at the terminal.

[0062] It should be noted that during the transmission process, data packets are often divided into multiple smaller segments. These segments may arrive at the receiving end in different orders, and there is a phenomenon of data packet fragmentation, which may affect the real-time performance of image data. Using the merge sort method to process data packets can reduce data packet fragmentation, and merging them into a complete data packet sequence can reduce the network delay that may occur during processing and transmission.

[0063] Therefore, it is necessary to collect the real-time image communication data sent by the network port, perform merge sort on the received data packets according to their memory sizes, and merge the received scattered data packets into a complete data packet sequence. The specific method is as follows:

[0064] Decompose the list of data packets to be sorted into several small subsequences until each subsequence contains only one unit data packet. Then, gradually merge adjacent subsequences to obtain larger ordered subsequences. During the merging process, it is necessary to compare the byte sizes of the unit data packet elements in the two subsequences and merge them into a new ordered sequence in ascending order of the number of bytes occupied. Repeat the above steps until all subsequences are merged into a complete ordered sequence.

[0065] It should be noted that edge nodes are usually closer to end-users and have faster response speeds. Since simple data packets are smaller and have lower processing requirements, using edge nodes for preprocessing can reduce the need to transfer data to the cloud, reduce the load and processing pressure on cloud servers, optimize the network bandwidth utilization rate of the entire system, and set a threshold and divide subsequences according to the data packet size, enabling the system to dynamically adjust the data processing and transmission strategies according to the current network load and performance requirements. This flexibility can better adapt to network conditions at different times and locations, improving the overall stability and reliability of the system.

[0066] Therefore, it is necessary to set an edge node threshold to perform segmentation preprocessing on the merged data packet sequence, divide the data packet sequence into simple data packet subsequences and complex data packet subsequences according to the number of bytes occupied, and transmit the simple data packet subsequences into the edge nodes and the complex data packet subsequences into the cloud server. The specific logic is as follows:

[0067]

[0068] if S i ≤T, S i ∈X jd

[0069] if S i >T, S i ∈X fz

[0070] Among them, S i represents the number of bytes occupied by the i-th data packet in the data packet sequence, T represents the preset byte threshold, X jd represents the simple data packet subsequence, X fz represents the complex data packet subsequence, i takes positive integers, i = 1, 2, 3... N, N represents the total number of all data packets in the data packet sequence, and the average number of bytes of all received data packets is used as the edge node threshold to divide the data packet sequence into simple data packets and complex data packets, which not only optimizes the utilization efficiency of system resources and the use of network bandwidth, but also significantly improves the response speed, flexibility and security of the real-time image communication system.

[0071] It should be noted that by collecting and analyzing the network latency data and bandwidth utilization in the edge computing nodes, the edge node coefficient can be dynamically generated. The edge node coefficient reflects the workload and network status of the current edge node, and can guide the system on how to adjust the processing priority and speed when processing simple data packets, which can effectively reduce the data processing latency caused by network congestion or high node load, and improve the response speed and stability of the system.

[0072] Therefore, it is necessary to collect the network latency data in the edge computing nodes, calculate the bandwidth utilization in the edge nodes, generate the edge node coefficient using the network latency data and bandwidth utilization, calibrate and unitize the simple data packets using the edge node coefficient, and send the generated standard simple data packets to the data processing buffer. The specific logic is as follows:

[0073]

[0074] Among them, P kd represents the bandwidth utilization in data transportation, M u represents the bandwidth used by the network port to transmit data, M o represents the total bandwidth of the network port. represents the time required for the j-th simple data packet to be transmitted into the edge node, where j represents the index of the simple data packet, and j = 1, 2, 3... n, C by represents the edge node coefficient, n represents the number of simple data packets in the simple data packet sub-column, and the bandwidth utilization. represents the number of bytes occupied by the j-th standard simple data packet. represents the number of bytes occupied by the j-th simple data packet in the simple data packet sub-column. It can be seen from the above formula that P kd is proportional to the used bandwidth M u When the number of data packet transmissions increases, it will lead to an increase in bandwidth utilization and network latency time. Collecting the network data in the edge computing points and generating the edge node coefficient based on it to optimize and adjust the processing flow of simple data packets is a key optimization step in the real-time image communication system, which improves the performance and efficiency of the system.

[0075] It should be noted that when transmitting the complex data packet sub-column to the cloud server, it often causes problems such as increased bandwidth usage and increased network latency. Therefore, by monitoring and analyzing the bandwidth usage of the cloud server, generating the bandwidth coefficient to help the system optimize the bandwidth utilization when processing complex data packets, and ensuring the stability of data transmission, this is crucial for the overall system to still be able to operate effectively under high load and network congestion conditions. By calibrating and unitizing the complex data packet sub-column using the generated bandwidth coefficient, the situation of data loss or damage can be greatly reduced, further improving the security and integrity of the data.

[0076] Therefore, it is necessary to collect the bandwidth used for network transmission data in the cloud server and the monitored real-time network latency data to generate a bandwidth coefficient. The specific logic is as follows:

[0077]

[0078] Among them, represents the time required for the k-th complex data packet to be transmitted into the cloud server. k represents the index of the complex data packet, and k = 1, 2, 3... N - n, G dk represents the bandwidth coefficient, and N - n represents the number of complex data packets in the complex data packet subsequence. It can be seen from the above formula that the bandwidth coefficient G dk is proportional to the bandwidth usage M u and the total network latency time is proportional. By real-time monitoring of the network latency data in the cloud server, it can help identify and address factors that may affect data transmission speed and latency. The bandwidth coefficient generated based on the latency data can adjust the processing order and priority of complex data packets to minimize latency and improve the real-time performance of data processing. This is crucial for real-time image communication and other latency-sensitive applications, and can significantly enhance the user experience and system performance.

[0079] Use the bandwidth coefficient to calibrate and unitize the complex data packet subsequence to generate standard complex data packets. The specific logic is as follows:

[0080]

[0081] Among them, represents the number of bytes occupied by the j-th standard complex data packet, represents the number of bytes occupied by the j-th complex data packet in the complex data packet subsequence. By calibrating and unitizing the complex data packet subsequence using the generated bandwidth coefficient, the system can process a large amount of complex data more effectively, thus supporting a wider range of complex real-time image communication application requirements.

[0082] It should be noted that when the unitized complex data packets enter the data buffer, if they enter disorderly, it will cause network node congestion and network congestion. Processing the data packets with a relatively small bandwidth ratio first can effectively reduce the bandwidth occupancy rate of the system under high load and reduce the queue congestion caused by the processing of large data packets, thereby improving the overall data processing efficiency and speed.

[0083] Therefore, it is necessary to set the arrangement priority from the complex data packet with a small bandwidth ratio to the complex data packet with a large bandwidth ratio according to the bandwidth size occupied by each complex data packet in the cloud server, and send them to the data processing buffer in turn. The specific method steps are as follows:

[0084] Perform bandwidth measurement on the data packets to enter the buffer, sort them according to their bandwidth ratios, and the data packets consuming less bandwidth resources will be assigned higher priorities. The high-priority data packets will be processed first to ensure they pass through the system quickly.

[0085] It should be noted that data loss and incomplete information may occur during the transportation of the two types of data packets, which may lead to incomplete image display on the terminal. However, through the screening and retransmission mechanism, the system can effectively utilize network resources and processing capabilities on the premise of ensuring data integrity.

[0086] Therefore, a screening mechanism needs to be established to calculate the bandwidth ratio of standard simple data packets and standard complex data packets, which is used to identify and screen the headers of the unitized data packets entering the data processing buffer subsequently, screen out the data packets with incomplete information and missing data, and retransmit them to the network port for data supplementation processing. The specific logic is as follows:

[0087]

[0088] Among them, F dkb represents the bandwidth ratio of standard simple data packets to standard complex data packets, represents the average bandwidth of standard simple data packets, represents the average bandwidth of standard complex data packets.

[0089] It should be noted that forming a data chain with the remaining data packets according to the serial numbers and identifiers, and processing them through the cloud server, and finally realizing image communication at the terminal can not only improve the efficiency and accuracy of data transmission, but also bring better user experience and service quality to users, enhance the system's ability to identify error data and security threats, thereby ensuring the stability and reliability of the overall system.

[0090] Therefore, the two types of data packets after screening need to be sent to the cloud server, form a data chain according to the combination of the serial number and identifier of each data packet, and the specific method steps for realizing image communication at the terminal are as follows:

[0091] Determine the type of data packet using header recognition, compare its bandwidth with the average bandwidth of data packets of its corresponding type. If both are less than the corresponding average bandwidth, it is determined as a data packet with incomplete information and missing data, record the missing and damaged parts, send these data packets back to the original network port to request data repair. During the waiting for retransmission, temporarily retain these data packets in a dedicated waiting area to ensure that all supplemented and processed data packets re-enter the data processing buffer intact. After screening, the two types of data packets are sent to the cloud server, and a data chain is formed by combining the sequence number and identifier of each data packet to achieve image communication at the terminal.

[0092] The present invention also provides an image control system for cloud-edge transmission, including:

[0093] A data packet acquisition and sorting module, which is used to acquire real-time image communication data sent by the network port, merge and sort the received data packets according to their memory sizes, and merge the received scattered data packets into a complete data packet sequence;

[0094] A threshold node segmentation and processing module, which is used to set an edge node threshold to perform segmentation preprocessing on the merged data packet sequence, divide the data packet sequence into a simple data packet sub-sequence and a complex data packet sub-sequence according to the number of bytes occupied, and transmit the simple data packet sub-sequence into the edge node, and transmit the complex data packet sub-sequence into the cloud server;

[0095] A unitary calibration and processing module, which is used to acquire network delay data in the edge computing node, calculate the bandwidth utilization rate, generate an edge node coefficient using the network delay data and the bandwidth utilization rate, calibrate and perform unitary processing on the simple data packet sub-sequence using the edge node coefficient, and send the generated standard simple data packet to the data processing buffer;

[0096] A coefficient optimization module, which is used to acquire the bandwidth used for network port transmission data in the cloud server and the real-time network delay data, generate a bandwidth coefficient, calibrate and perform unitary processing on the complex data packet sub-sequence using the bandwidth coefficient, and generate a standard complex data packet;

[0097] A bandwidth arrangement module, which is used to arrange the priorities according to the bandwidth sizes occupied by the standard complex data packets in the cloud server, and the priority of the standard complex data packet with a smaller bandwidth occupancy ratio is higher than that of the standard complex data packet with a larger bandwidth occupancy ratio, and send each standard complex data packet to the data processing buffer in turn according to the priority;

[0098] The data packet screening and serialization module sets up a screening mechanism to calculate the bandwidth ratio of standard simple data packets and standard complex data packets, which is used to identify the headers of the subsequent unitized data packets entering the data processing buffer and screen them. The data packets with incomplete information and missing data are screened out and resent to the network port for data supplementation processing. The remaining data packets in the buffer are combined according to the sequence number and identification to form a data chain and sent to the cloud server to achieve image communication at the terminal.

[0099] All the above formulas are dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0100] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0101] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0102] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application.

Claims

1. A method for controlling image communication of cloud-edge transmission, characterized in that: The specific steps include: Step 1: Collect the real-time image communication data sent by the network port, merge and sort the received data packets according to their memory size, and merge the received scattered data packets into a complete data packet sequence; Step 2: Set up edge node thresholds to segment and pre-process the merged data packet sequence, divide the data packet sequence into simple data packet sub-sequences and complex data packet sub-sequences according to the bytes occupied, and transmit the simple data packet sub-sequences to the edge node, and transmit the complex data packet sub-sequences to the cloud server; Step 3: Collect the time required for simple data packets to enter the edge node, calculate the bandwidth utilization, generate edge node coefficients using network delay data and bandwidth utilization, calibrate and unitize the simple data packet sub-sequences using the edge node coefficients, and send the generated standard simple data packets to the data processing buffer; Step 4: Generate a bandwidth coefficient based on the bandwidth used by the network port to transmit data and the time required for the complex data packet to be transmitted to the cloud server, and use the bandwidth coefficient to calibrate and unitize the complex data packet sub-sequences to generate a standard complex data packet; Step 5: Prioritize the standard complex data packets according to the bandwidth occupied by the cloud server, and the priority of the standard complex data packets with a small bandwidth share is higher than that of the standard complex data packets with a large bandwidth share, and send each standard complex data packet to the data processing buffer in turn according to the priority; Step 6: Establish a screening mechanism to calculate the bandwidth ratio of standard simple data packets and standard complex data packets, which is used to identify the headers of the unitized data packets that subsequently enter the data processing buffer and screen them. Data packets with incomplete information or missing data are screened out and resent to the network port for data supplementation. The remaining data packets in the buffer are linked to the serial number and identifier to form a data link, which is sent to the cloud server to realize image communication at the terminal.

2. The image communication control method for cloud-edge transmission according to claim 1, characterized in that: The received data packets are merged and sorted according to their memory sizes, and the received scattered data packets are integrated into a complete data packet sequence. The specific method is: decompose the data packet list to be sorted into several small subsequences until each subsequence contains only one unit data packet, and then gradually merge adjacent subsequences to obtain a larger ordered subsequence. During the merging process, it is necessary to compare the byte sizes occupied by the unit data packet elements of the two subsequences, and merge them into a new ordered sequence from small to large in terms of the bytes occupied, and repeat the steps until all subsequences are merged into a complete ordered sequence.

3. The image communication control method for cloud-edge transmission according to claim 1, characterized in that: The edge node byte threshold is set to segment and pre-process the merged data packet sequence. The time required for simple data packets to enter the edge node is used to divide the data packet sequence into simple data packet sub-sequences and complex data packet sub-sequences according to the bytes occupied, and the simple data packet sub-sequences are transmitted to the edge node, and the complex data packet sub-sequences are transmitted to the cloud server. The specific method is: the data packets with bytes greater than the node byte threshold are regarded as complex data packets, and the data packets with bytes not greater than the node byte threshold are regarded as simple data packets. All simple data packets form simple data packet sub-sequences, and all complex data packets form complex data packet sub-sequences. The specific logic is: if S i ≤T,S i ∈X jd if S i >T,S i ∈X fz Among them, S i represents the number of bytes occupied by the ith data packet in the data packet sequence, T represents the preset byte threshold, and X jd Represents a simple packet sub-column, X fz Represents a complex data packet subsequence, i is a positive integer, i=1, 2, 3...N, N represents the total number of all data packets in the data packet sequence.

4. The image communication control method for cloud-edge transmission according to claim 1, characterized in that: Collect network delay data in edge computing nodes, calculate bandwidth utilization, and use network delay data and bandwidth utilization to generate edge node coefficients. The specific logic is as follows: Among them, P kd Indicates the bandwidth utilization in data transmission, M u Indicates the bandwidth used by the network port to transmit data, M o Indicates the total bandwidth of the network port. represents the time required for the jth simple data packet to be transmitted into the edge node, j represents the index of the simple data packet, and j = 1, 2, 3...n, C by represents the edge node coefficient, and n represents the number of simple packets in the simple packet sub-column.

5. The image communication control method for cloud-edge transmission according to claim 4, characterized in that: The generated edge node coefficients are used to calibrate and unitize the simple data packet sub-sequences to generate standard simple data packets. The specific logic is as follows: in, Represents the number of bytes occupied by the jth standard simple data packet, Represents the number of bytes occupied by the jth simple data packet in the simple data packet sub-column.

6. The image communication control method for cloud-edge transmission according to claim 1, characterized in that: The bandwidth coefficient is generated by using the bandwidth used by the network port to transmit data and the monitored network delay data. The specific logic is as follows: in, represents the time required for the kth complex data packet to be transmitted into the cloud server, k represents the index of the complex data packet, and k = 1, 2, 3...Nn, G dk represents the bandwidth factor, Nn represents the number of complex packets in the complex packet sub-column, M u Indicates the bandwidth used by the network port to transmit data.

7. The image communication control method for cloud-edge transmission according to claim 6, characterized in that: The complex data packet sub-sequences are calibrated and unitized using the bandwidth coefficients to generate standard complex data packets. The specific logic is as follows: in, Represents the number of bytes occupied by the kth standard complex data packet, Represents the number of bytes occupied by the kth complex data packet in the complex data packet sub-column.

8. The image communication control method for cloud-edge transmission according to claim 7, characterized in that: According to the bandwidth occupied by each complex data packet in the cloud server, the priority is set from complex data packets with small bandwidth share to complex data packets with large bandwidth share. The specific method is as follows: the bandwidth of the data packets that are about to enter the buffer is measured, and they are sorted according to their bandwidth share. Data packets that consume less bandwidth resources will be given higher priority, and high-priority data packets will be processed first to ensure that they pass through the system quickly; A screening mechanism is established to calculate the bandwidth ratio of standard simple data packets and standard complex data packets. The specific logic is as follows: Among them, F dkb Represents the bandwidth ratio of standard simple data packets to standard complex data packets, Represents the average standard simple packet bandwidth, Represents the average bandwidth of standard complex packets, Indicates the number of bytes occupied by the jth standard simple data packet.

9. The image communication control method for cloud-edge transmission according to claim 8, characterized in that: The simple data packet sub-sequence and the complex data packet sub-sequence are screened by bandwidth comparison, and the data packets with incomplete information and missing data are screened out and resent to the network port for data supplementation processing. The specific method is as follows: the type of data packet is determined by header recognition, and its bandwidth is compared with the average bandwidth of data packets of the same type. If both are smaller than the corresponding bandwidth average, it is determined to be a data packet with incomplete information and missing data, and the missing and damaged parts are recorded. These data packets are sent back to the original network port to request data repair. While waiting for retransmission, these data packets are temporarily retained in a special waiting area to ensure that all supplemented data packets re-enter the data processing buffer intactly and correctly. The two types of data packets after screening are sent to the cloud server, and the serial number and identification of each data packet are combined to form a data link to realize image communication at the terminal.

10. A cloud-edge transmitted image communication control system, the control system being used to execute the image communication control method according to any one of claims 1 to 9, comprising: The data packet collection and sorting module is used to collect the real-time image communication data sent by the network port, merge and sort the received data packets according to their memory size, and merge the received scattered data packets into a complete data packet sequence; The threshold node segmentation processing module is used to set the edge node threshold to perform segmentation preprocessing on the merged data packet sequence, divide the data packet sequence into simple data packet sub-sequences and complex data packet sub-sequences according to the bytes occupied, and transmit the simple data packet sub-sequences to the edge node, and transmit the complex data packet sub-sequences to the cloud server; A unitized calibration processing module is used to collect network delay data in edge computing nodes, calculate bandwidth utilization, generate edge node coefficients using network delay data and bandwidth utilization, calibrate and unitize simple data packet sub-sequences using edge node coefficients, and send the generated standard simple data packets to the data processing buffer; A coefficient optimization module is used to generate a bandwidth coefficient according to the bandwidth used by the network port to transmit data and the time required for the complex data packet to be transmitted into the cloud server, and to calibrate and unitize the complex data packet sub-sequences using the bandwidth coefficient to generate a standard complex data packet; The bandwidth arrangement module is used to arrange the priority of the standard complex data packets according to the bandwidth occupied by the standard complex data packets in the cloud server, and the priority of the standard complex data packets with a small bandwidth share is higher than that of the standard complex data packets with a large bandwidth share, and each standard complex data packet is sent to the data processing buffer in sequence according to the priority; The data packet screening and serialization module is used to establish a screening mechanism, calculate the bandwidth ratio of standard simple data packets and standard complex data packets, and perform header recognition and screening on the unitized data packets that subsequently enter the data processing buffer. The data packets with incomplete information and missing data are screened out and resent to the network port for data supplementation processing. The remaining data packets in the buffer are linked with the serial number and identification to form a data link, and sent to the cloud server to realize image communication at the terminal.

Images