Remote video monitoring data transmission method

Abstract

The invention discloses a remote video monitoring data transmission method, and the method comprises the steps: collecting visual video information through a robot camera, and transmitting the visual video information to a remote power plant monitoring center through a HDMI interface, an encoder, a 5G-CPE, a decoder and other links. The 5G-CPE uses MEC, 5G slicing and SRT technologies to adjust resource allocation according to network conditions and requirements, and defines configuration slices and priorities. A network monitoring component is deployed to monitor slice loads in real time, a load threshold value is set for each slice, and a resource adjustment strategy is triggered when the loads exceed the threshold values. And resources are dynamically reallocated after analysis and evaluation, so that sufficient and optimized utilization of high-load slice resources is ensured. Performance is continuously monitored and feedback adjustment is performed, network adaptability and flexibility are improved, key video streams are ensured to be transmitted preferentially, and power plant inspection requirements are met.

Description

Technical Field

[0001] The present application relates to the technical field of power system stability analysis, and in particular to a remote video monitoring data transmission method. Background Art

[0002] With the rapid development of the power industry, the continuous increase in power load and voltage levels, and the continuous expansion of the power grid, the surge in the number and types of power equipment has brought unprecedented challenges to the installation, detection, operation and maintenance of equipment. In this context, the integrated application of the Internet of Things and AI technology has injected new vitality into the construction of smart grids. Among them, digitalization and automation have become the key forces to promote the transformation and upgrading of the power industry. The widespread use of intelligent equipment such as power inspection robots and drones is gradually replacing manual tasks such as video monitoring, equipment detection and data collection, greatly improving work efficiency and safety.

[0003] However, there are still a series of technical challenges in the transmission of remote video surveillance data. In particular, the existing technical solutions are still insufficient in terms of processing massive data, meeting high bandwidth requirements, and achieving low-latency transmission. For remote real-time controlled video transmission, how to ensure low-latency transmission of data in remote public network communications has become an urgent problem to be solved. In order to achieve high-speed, stable and secure video transmission, provide smart devices with remote monitoring capabilities similar to the "first-person perspective", and thereby improve the safety and reliability of automation equipment, it is necessary to explore more advanced remote video surveillance data transmission methods.

[0004] At present, most drones and inspection robots still rely on remote controls for on-site operation, and real-time video transmission is mainly limited to wired Ethernet transmission within the local area network. It is necessary to break this limitation and realize remote monitoring. At the same time, in order to solve the delay problem caused by the video transmission module through the Internet, it is necessary to find a new solution to reduce the delay and meet the high requirements of the power industry for safety and stability.

[0005] Therefore, a new remote video monitoring data transmission method is developed by combining the Internet of Things, AI technology and advanced communication technology. This method should be able to achieve rapid transmission of massive data, meet high bandwidth requirements, and ensure low-latency transmission. At the same time, this method should also have a high level of intelligence, be able to support remote monitoring and data analysis, reduce labor costs, improve work efficiency, and enhance the reliability of diagnosis and the real-time nature of communication. The application of this innovative method will inject new impetus into the intelligent development of the power industry and promote the power industry to a higher level. Summary of the invention

[0006] The present application provides a remote video surveillance data transmission method and system, which can solve the problems of massive transmission delay of remote video data, remote video surveillance and the like existing in the prior art, and meet the needs of high bandwidth, low latency and remote video surveillance.

[0007] The present application provides a remote video surveillance data transmission method, comprising: S101, collecting robot visual video information data through a robot camera, and transmitting the visual video information data to an encoder through an HDMI interface; S102, the encoder encodes the visual video information data, and transmits the encoded visual video information data to the 5G-CPE through the network port; S103, 5G-CPE transmits the encoded visual video information data to the decoder, and uses MEC, 5G slicing technology and SRT network transmission technology, and adjusts 5G slice resource allocation according to real-time network conditions and transmission requirements; S104, the decoder decodes the received encoded visual video information data, and transmits the decoded visual video information data to a remote monitoring center of the power plant, and finally presents it on a display screen of the monitoring center.

[0008] Preferably, the S103, and adjusting the 5G slice resource allocation method according to the real-time network status and transmission requirements, includes: S201, collect and analyze specific business requirements in power plant inspection scenarios, define and configure multiple 5G slices according to the real-time, bandwidth, and reliability requirements of different application scenarios of power plant inspection, and assign priorities; S202, deploy network monitoring components to monitor the load of 5G slices in real time and collect KPIs of each slice; S203, setting a load threshold for each slice, and triggering a resource adjustment policy mechanism when the KPIs of a slice are monitored to exceed the set threshold; S204: Analyze and evaluate whether other 5G slice resources have available remaining resources for dynamic resource reallocation, increase resources for high-load 5G slices according to the evaluation results, and migrate low-priority video streams from high-load 5G slices to idle slices; S205: After the resources are adjusted, the transmission performance of the slices is continuously monitored, and the resource adjustment strategy is adjusted based on the monitoring results.

[0009] Preferably, the S201 defines and configures multiple slices according to the real-time, bandwidth, and reliability requirements of different application scenarios of power plant inspection, including: According to the different scene requirements of power plant inspection, the service type and performance indicators of each slice are clarified, and the network architecture, resource allocation strategy, isolation mechanism, etc. of the slice are designed; Create a new slice instance in the network management system, assign a unique identifier to each slice, and allocate network resources according to slice requirements, including wireless resources, core network resources, and transmission resources.

[0010] Preferably, the step S201, allocating priorities, includes: Each video stream is assigned a QCI-P priority level, which is set based on the content and importance of the video stream. Daily inspection video streams have a lower priority. Such video streams are mainly used to record the daily operation status of the equipment, and have relatively low requirements for real-time performance and bandwidth. Emergency inspection video streams are set to the highest priority. Such video streams are usually generated after an equipment anomaly is discovered or an alarm is received. Video information data transmission needs to be carried out immediately, and has high requirements for real-time performance, bandwidth, and reliability. Each element in the queue is a video stream request, including information such as the video stream identifier, priority, bandwidth requirement, and real-time requirement. When a new video stream request arrives, it is added to the priority queue and sorted according to its priority. For each video stream request, slice resources are dynamically allocated according to its bandwidth requirements and real-time requirements. According to the priority and bandwidth requirements of the video stream request, appropriate resources are allocated to the video stream from the available slice resources.

[0011] Preferably, the step S201, allocating priorities, further includes: S301, prioritize all devices in the 5G slice according to specific business requirements and network resource usage; S302, when deciding to adjust 5G slice resources, give priority to devices with lower priority for 5G slice switching; S303, after the communication status of the switched device is verified, set the device to send data to the original slice and the new slice simultaneously within x seconds; S304, decoder synchronous decoding and data comparison; S305, after the video data comparison and verification pass, confirming that the video data transmitted by the new slice is consistent with the video data of the original slice and time synchronized, issue a switching instruction to stop sending video data to the original slice and completely switch to the new slice for video data transmission; S306: After the switch is completed, the performance and stability of the new slice are continuously monitored, and the slice configuration and resource allocation are adjusted in a timely manner according to the monitoring results and feedback.

[0012] Preferably, the S301, prioritizing all devices in the 5G slice, includes: Collect key indicators of all devices, take into account factors such as device criticality, real-time requirements, and urgency based on the collected information, calculate a priority score for each device, and sort all devices based on the calculated priority scores. The lower the score, the lower the priority of the device. The device priority score is calculated based on the three weights of key indicator device criticality, video data stream real-time requirements, and video data stream urgency. The device priority score is calculated using the following formula: B=w1*K+w2*R+w3*F; B represents the priority score of the device, w1, w2, and w3 represent the weights of device criticality, video data stream real-time requirement, and video data stream urgency, respectively, and w1+w2+w3=1.

[0013] Preferably, the S304, decoder synchronous decoding and data comparison, includes: Ensure that the decoder has sufficient processing power and storage space to support simultaneous decoding; Synchronously decoding the video data streams from the two slices at the decoder end; At the same time, the decoded video data is compared to verify whether the video data stream sent by the new slice is consistent with the original slice video data and whether the time is synchronized; The video data stream comparison process uses the NTP network time protocol to ensure time synchronization on the decoder side, so that byte-by-byte comparison can be achieved; When the video data stream is found to be inconsistent, a rollback is performed, and the video data stream returns to the original slice and resends the video data stream to the slice.

[0014] Preferably, the S302, 5G slice switching, includes: S401, set up a temporary dedicated slice separately and define a policy; S402, monitoring slice load and resource allocation; S403: For the video data stream that needs to be migrated, according to its transmission strategy and requirements in the original slice, after performing corresponding replication configuration in the temporary dedicated slice, the video data stream is migrated to the temporary dedicated slice; S404, adjusting the load of the temporary dedicated slice and the resource allocation of the video data transmission demand according to the same transmission strategy and requirement of the migrated slice; S405, for a temporary slice that has occupied a video data stream for a long time and has an excessively high load and a reduced quality of video data stream transmission, assess the feasibility of migrating the temporary slice back to the original slice or allocating the temporary slice to a slice with appropriate resources; S406, regularly evaluate the usage of temporary dedicated slices and continuously improve configuration strategies and resource allocation.

[0015] Preferably, the S404, adjusting the load of the temporary dedicated slice and adjusting its resource allocation according to the video data transmission requirement, includes: S501, setting an initial general slice to carry the transmission task of various video data streams; S502, start the slice monitoring mechanism to track the slice load, video data transmission requirements and performance; S503, monitoring the video data based on the slices, identifying video data streams with similar transmission requirements, and dividing the video data streams into different groups according to the analysis results; S504, splitting the initial slice into multiple sub-slices according to the grouping result, each sub-slice is specifically used to carry a video data stream of a specific group and inherits the transmission strategy and requirements of the video data stream of the group; S505, for the newly added video data stream, the most suitable sub-slice is selected for carrying through the intelligent matching algorithm according to its transmission requirements, and the slice monitoring mechanism is updated accordingly to continuously track the newly added sub-slice; S506: Dynamically adjust the resource configuration of each sub-slice according to the load, performance and resource utilization of each sub-slice.

[0016] S507, regularly evaluate the slice splitting and grouping management strategies, and iteratively optimize the slice splitting strategies and grouping rules according to the evaluation results.

[0017] Preferably, the S503, identifying video data streams with similar transmission requirements, and dividing the video data streams into different groups according to the analysis results, includes: Use the K-means clustering algorithm to classify the video data stream, collect the key attribute data of the video data stream, clean and standardize the data, select the appropriate number of classifications K value based on actual needs and data characteristics, and randomly select K data points as the initial centroid; Assign each data point to the class where the nearest centroid is located, and recalculate the centroid position of each class; Repeat the above steps until the centroid position no longer changes significantly or the predetermined number of iterations is reached; A label is set for each group of video data streams according to the classification result, and a similarity threshold is set to determine whether the video data streams belong to the same group.

[0018] One or more technical solutions provided in this application have at least the following technical effects or advantages: 1. The setting of load thresholds and the judgment of trigger conditions are crucial for the stable operation of slices. By reasonably setting the load threshold of each slice, its load can be monitored in real time, and the resource adjustment strategy can be triggered in time when the threshold is exceeded, so as to avoid performance degradation or service interruption, and ensure that the slice meets the video stream requirements of power plant inspections. At the same time, accurate trigger condition judgment can improve the efficiency of network resource utilization. In addition, dynamic resource reallocation is an important means to optimize slice resource utilization. When the slice load exceeds the threshold, the resource allocation is adjusted in real time through the trigger strategy to ensure that high-load slices receive sufficient support, while optimizing resource utilization, improving network adaptability and flexibility, and ensuring priority transmission of key video streams.

[0019] 2. A scientific priority sorting mechanism has been established for devices within the slice to ensure that key services and devices with high real-time requirements are given priority, providing an accurate basis for resource adjustment. When adjusting resources, low-priority devices are switched first to reduce the impact on key services while ensuring the performance and stability of the new slice. During the switching process, video data synchronization, verification and rollback mechanisms are used to ensure the integrity and reliability of data transmission. After switching, the performance of the new slice is continuously monitored, and the configuration and resource allocation are adjusted and optimized in a timely manner according to the monitoring results to improve the flexibility and efficiency of resource adjustment, thereby providing a more stable and reliable video data stream transmission environment.

[0020] 3. By setting up temporary dedicated slices and clarifying management strategies, efficient use of resources is achieved. This mechanism can dynamically adjust the resource allocation between slices. When the resources of a slice are tight, some video data streams can be quickly migrated to temporary slices to relieve pressure and avoid performance degradation. Temporary slices are designed to take over overflow data streams, which can efficiently utilize network resources and reduce waste. During the migration process, the system configures the data streams in the temporary slices according to the original slice transmission strategy to ensure that the transmission quality and performance remain unchanged after migration. This is due to the unique identifier set by the system for each data stream and the original transmission strategy maintained in the temporary slice, which ensures efficient and stable transmission of video data streams.

[0021] 4. Slice splitting and group management can significantly improve video transmission efficiency and resource utilization. Through detailed analysis, video data streams with similar transmission requirements are grouped, and the initial slice is split into multiple sub-slices accordingly. Each sub-slice carries a specific grouped data stream to ensure accurate resource allocation and avoid waste. Sub-slices can be optimized for data stream types, such as bandwidth, latency, etc., to provide efficient and stable transmission. As new data streams are connected, the system can quickly allocate them to the most suitable sub-slice through intelligent algorithms without large-scale adjustments, maintaining high flexibility and responsiveness. Group management also allows the system to dynamically adjust the number and configuration of sub-slices, reserving space for business expansion, and easily responding to business growth and changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flowchart of a remote video monitoring data transmission method according to an embodiment of the present invention; Figure 2 A flowchart of adjusting slice resource allocation according to network conditions and transmission requirements in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0024] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are only for illustrative purposes and do not represent the only implementation method.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more related listed items.

[0026] Embodiment 1: Figure 1 It is a flowchart of a remote video monitoring data transmission method according to an embodiment of the present invention.

[0027] like Figure 1 As shown, a remote video monitoring data transmission method includes the following steps: S101, collecting robot visual video information data through the power inspection robot camera, and transmitting the visual video information data to the encoder through the HDMI interface.

[0028] Among them, the HDMI interface is a high-definition multimedia interface, the input interface is HDMI2.0, the resolution and frame rate settings of 1920x1080@30fps automatically adapt to the input video, and it is used to transmit high-definition video and audio data. The HDMI interface output by the camera is fully compatible with the HDMI interface input by the encoder. The HDMI interface can ensure the transmission quality and stability of video data, and support high-resolution and high-frame rate video transmission.

[0029] The camera resolution is at least 1920x1080, the frame rate is set to 30fps, and the output interface must match the camera model. HDMI2.0 is used in the implementation of the present invention to ensure that clear and smooth video images are captured.

[0030] The electric power inspection robot mainly includes a robot camera, an encoder and a 5G-CPE. The robot camera is used to collect the robot's visual video information data; the encoder is connected to the robot camera via an HDMI interface, and is used to receive and encode the robot's visual video information data; one end of the 5G-CPE is connected to the encoder via a network port, and the other end of the 5G-CPE is connected to the substation 5G base station via a 5G-SIM card, and is used to transmit the received encoded robot's visual video information data to the Internet.

[0031] Specifically, the robot camera is turned on and begins to capture real-time visual video information data of the power plant inspection. The video information data may include information data such as the appearance of the generator, operating indicators, and operation and maintenance status. The camera transmits the collected visual video information data to the encoder connected to it in real time through the HDMI interface. After receiving the data, the encoder immediately begins to compress and encode it. The compressed visual video information data is transmitted to the monitoring center of the power plant. The operator can view the robot's inspection screen in real time through the display screen of the monitoring center to understand the operating status of the site.

[0032] S102, the encoder encodes the visual video information data, transmits the encoded visual video information data to the 5G-CPE through the network port, and can actively report its own IP address in real time.

[0033] Among them, 5G-CPE is a 5G router, which is responsible for transmitting the encoded video data from the encoder to the remote monitoring center. Select the H.265 (HEVC) encoding algorithm with high compression efficiency and video quality, the resolution and frame rate are consistent with the camera settings, the initial bit rate is set to 8Mbps, the GOP size is set to the number of frames in 1 second, that is, 30 frames, and the bitstream format is selected as MP4 or TS and other common bitstream formats.

[0034] Specifically, after configuring the camera and encoder, connect the encoder's network port to the 5G-CPE's network port through a network cable, configure the encoder's network settings, ensure that the encoder's subnet mask is consistent with the 5G-CPE, and the encoder's DNS server address can be a public DNS server address (such as 8.8.8.8 and 8.8.4.4) or the power plant's internal DNS server address. Set the encoder's streaming protocols such as RTSP or RTMP to push the encoded video stream to the specified port of the 5G-CPE; for example, if the receiving port configured on the 5G-CPE is 554, the streaming server address on the encoder should be set to rtsp: / / 192.168.1.1:554 / ; Configure the network settings of the 5G-CPE to ensure that it can access the 5G network normally, set port forwarding rules on the 5G-CPE, set access control rules on the 5G-CPE, and allow the video stream from the encoder to enter the 5G-CPE through a specific port (such as 554); set port mapping rules to map the video stream from the encoder (source IP: encoder IP, source port: streaming protocol port) to the specified port on the WAN port of the 5G-CPE (destination IP: 5G-CPE WAN port IP, the destination port can be any unoccupied port, such as 10000); create forwarding rules to forward the mapped video stream to the specified address and port of the remote monitoring center or cloud server. For example, if the IP address of the remote monitoring center is 203.0.113.1 and the receiving port is 8080, then the forwarding rule should be set to forward the traffic from port 10000 on the 5G-CPE WAN port to 203.0.113.1:8080; Forward the received video stream to the designated address and port of the monitoring center. Configure the firewall and security policy of the 5G-CPE as needed to allow the video stream from the encoder to pass through the firewall and block other unauthorized access. Regularly update the security firmware and patches of the 5G-CPE to ensure the security and stability of data transmission.

[0035] S103, 5G-CPE transmits the encoded visual video information data to the decoder, and uses MEC, 5G slicing technology and SRT network transmission technology, and adjusts slice resource allocation according to real-time network conditions and transmission requirements.

[0036] Among them, network slicing is to cut a physical network into multiple virtual end-to-end networks, each of which can obtain logically independent network resources, and the slices can be insulated from each other, and network operators can choose the characteristics required for each slice, such as low latency, high throughput, connection density, spectrum efficiency, traffic capacity and network efficiency. 5G slicing technology is to cut a 5G network into multiple virtual networks, so as to support more video streams, with the advantages of high bandwidth and low latency. The slices mentioned in this article refer to 5G slices.

[0037] SRT (Secure Reliable Transport) is an open source, free and flexible specification. It is a network transmission technology that can transmit video data streams in real time and accurately in complex network environments. SRT transmission technology uses the UDP protocol at the transport layer. In the process of transmitting audio and video streams from the SRT source device to the SRT to the decoder, SRT will detect and adapt to the changing network status between the two devices in real time, resist bandwidth jitter caused by network congestion, and rely on its powerful error recovery mechanism to minimize the possibility of network packet loss. However, the end-to-end transmission function of SRT requires at least one of the encoder and decoder to have a fixed IP address.

[0038] S104, the decoder decodes the received encoded visual video information data, and transmits the decoded visual video information data to a remote monitoring center of the power plant, and finally presents it on a display screen of the monitoring center.

[0039] Specifically, due to the complexity of the power plant environment and the high requirements for transmission stability, optical fiber is used to connect the decoder and the remote monitoring center for wired transmission, and the RTSP protocol ensures the stable transmission of the video stream. A streaming media server is deployed in the remote monitoring center, and the decoder pushes the decoded video stream to the streaming media server through the RTSP protocol. The streaming media server is responsible for receiving and processing (such as transcoding and distribution) the video stream and providing it to the monitoring center as needed.

[0040] In some embodiments, slice resource allocation is adjusted according to real-time network conditions and transmission requirements, such as Figure 2 As shown, specifically including: S201, collect and analyze specific business requirements in power plant inspection scenarios, define and configure multiple slices according to the real-time, bandwidth, and reliability requirements of different application scenarios of power plant inspection, and assign priorities.

[0041] Among them, the specific business needs of power plant inspections include: inspection area, equipment type, operating status, video quality requirements and other specific demand scenarios are set according to actual conditions.

[0042] Specifically, according to the different scene requirements of power plant inspection, the service type and performance indicators of each slice are clarified, and the network architecture, resource allocation strategy, isolation mechanism, etc. of the slice are designed. A new slice instance is created in the network management system, a unique identifier is assigned to each slice, and network resources are allocated according to the slice requirements, including wireless resources, core network resources, transmission resources, etc.

[0043] For example, to monitor the key equipment in the power plant in real time, adopt high-definition video monitoring and real-time video data transmission solutions, and select slice 1 (key equipment real-time monitoring slice) for deployment. Slice 1 must meet the requirements of real-time ≤ 100ms, bandwidth ≥ 100Mbps, and reliability 99.999%, and use logical isolation to ensure video data security. Create an instance for slice 1, assign Slice ID 001, select the high-priority QCI value QCI 1, corresponding to GBR service. According to the inspection task requirements, set the MBR value to no less than 100Mbps, and ensure that the maximum end-to-end delay does not exceed 100ms and the packet loss rate does not exceed 1e-5. In terms of wireless resources, dynamically allocate PRBs to slice 1 to ensure that sufficient resources can be obtained even during network peak hours. The core network resources are allocated to dedicated SMF network elements to achieve logical isolation and resource reservation between slices. Transmission resources establish a dedicated VPN channel to ensure the isolation and security of video data transmission.

[0044] Specifically, assigning priority is to assign a priority level (QCI-P) to each video stream. The priority is set based on factors such as the content and importance of the video stream. ), this type of video stream is mainly used to record the daily operation status of the equipment, and has relatively low requirements for real-time and bandwidth; the emergency inspection video stream is set as the highest priority ( ), this type of video stream is usually generated after a device anomaly is discovered or an alarm is received, and video information data needs to be transmitted immediately, with high requirements for real-time, bandwidth, and reliability. Each element in the queue is a video stream request, including information such as the video stream ID, priority, bandwidth requirements, and real-time requirements. When a new video stream request arrives, it is added to the priority queue and sorted according to its priority. For each video stream request, slice resources are dynamically allocated based on its bandwidth requirements and real-time requirements, and appropriate resources are allocated to the video stream from the available slice resources based on the priority and bandwidth requirements of the video stream request.

[0045] S202, deploy network monitoring components to monitor slice load in real time and collect KPIs of each slice.

[0046] Among them, KPIs include bandwidth utilization, CPU / memory utilization, packet loss rate, etc.

[0047] S203, set a load threshold for each slice, and when the KPIs of a slice are monitored to exceed the set threshold, trigger the resource adjustment policy mechanism.

[0048] For example, taking slice 1 (real-time monitoring slice for key equipment) as an example, its key performance indicators (KPIs) include bandwidth utilization threshold of 90% (slice 1 needs to ensure at least 100Mbps bandwidth to support high-definition video monitoring and real-time video data transmission. Setting the bandwidth utilization threshold to 90% can ensure that there is still enough buffer space when it is close to full load. Once the bandwidth utilization exceeds 90%, it may mean that slice 1 is about to face a bandwidth bottleneck), CPU / memory utilization threshold CPU utilization of 70%, memory utilization of 80% (setting the CPU utilization threshold to 70% and the memory utilization threshold to 80% can ensure that there are still enough remaining resources to deal with emergencies when resource usage is relatively tight. When the indicator exceeds the set threshold, it may mean that the processing capacity of slice 1 is close to saturation), packet loss rate threshold (1e-4 (0.0001) Setting the video packet loss rate threshold to 1e-4 can be used as a sensitive indicator. When the packet loss rate exceeds this threshold, it may mean that the network transmission quality is degraded or there is a potential congestion problem).

[0049] S204, analyze and evaluate whether other slice resources have available remaining resources for dynamic resource reallocation, increase resources for high-load slices according to the evaluation results, and migrate low-priority video streams from high-load slices to idle slices.

[0050] Specifically, the resource usage of each slice is monitored in real time, including bandwidth, computing resources, storage resources, etc. Slices with high resource usage and heavy load are identified, and resources are increased for high-load slices, including bandwidth expansion and computing resource enhancement, to meet their video streaming needs and ensure service quality and performance stability. Low-priority video streams (with relatively low requirements for latency and bandwidth) are migrated from high-load slices to idle slices.

[0051] The evaluation includes bandwidth occupancy, CPU usage, memory occupancy, throughput, latency, packet loss rate, real-time communication, video streaming, and non-real-time video data transmission. For idle slices, the remaining resources are calculated, including available bandwidth, number of idle CPU cores, remaining storage capacity, etc.

[0052] S205, after the resources are adjusted, the transmission performance of the slices is continuously monitored, and the monitoring results are fed back to the resource adjustment strategy to make necessary fine-tuning.

[0053] Among them, transmission performance includes delay, packet loss rate, throughput, monitoring jitter, bit error rate, etc. The fine-tuning range should be determined according to the actual application situation.

[0054] It should be noted that during the dynamic slice resource adjustment process, security measures should be strengthened, and data encryption and access control can be set up to prevent data leakage and illegal access. If necessary, a fault recovery mechanism can be established to ensure that services can be quickly restored when a network failure occurs during the resource adjustment process.

[0055] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: 1. Load threshold setting and trigger condition judgment are key mechanisms to ensure stable operation of slices. By setting a reasonable load threshold for each slice, the load of the slice can be monitored in real time, and the resource adjustment strategy can be triggered in time when the load exceeds the threshold. This mechanism helps to avoid performance degradation or service interruption of the slice due to excessive load, and ensures that the slice can always meet the video streaming requirements in the power plant inspection scenario. At the same time, through accurate trigger condition judgment, network resources can be used more efficiently to avoid unnecessary resource waste.

[0056] 2. Dynamic resource reallocation and trigger condition judgment are important means to optimize slice resource utilization. When the load of a slice exceeds the set threshold, resources can be reallocated in real time by triggering the resource adjustment policy. This mechanism ensures that high-load slices obtain sufficient resource support to meet their video streaming needs while avoiding idle or wasted resources. By dynamically adjusting resource allocation, it is possible to respond more flexibly to the ever-changing network environment and video streaming needs, and improve the adaptability and flexibility of the network. In addition, migrating low-priority video streams from high-load slices to idle slices can further optimize resource utilization and ensure priority transmission of critical video streams.

[0057] Embodiment 2: In embodiment 1, switching of video data to a new slice may cause a problem of short delay of video data, which may cause a short interruption of video data transmission or an increase in delay.

[0058] In some embodiments, video data slice switching is optimized to reduce short data delays, device priorities are sorted, low-priority devices are selected for switching, dual-slice transmission of video data is set, decoders are synchronously decoded and compared with video data, and continuous monitoring and feedback are provided to ensure the continuity and integrity of video data during slice switching, and improve the real-time, reliability and low-latency performance of video data transmission.

[0059] Furthermore, in the power plant inspection scenario, real-time performance is a very important performance indicator, and any delay may affect the monitoring center's inspection engineer's accurate judgment and timely response to the equipment status. For high-priority video streams, such as emergency inspection video streams, increased delays may lead to the loss or retransmission of key data packets, thereby affecting the continuity and accuracy of the inspection work. Frequent slice switching and unreasonable resource allocation may cause network congestion, especially when multiple high-load slices request resources at the same time, which may cause global performance degradation. Through dual-slice transmission and synchronous decoding comparison mechanism, the short delay of video data caused by slice switching is significantly reduced, and then the contents of steps S201 to S205 are performed.

[0060] In some embodiments, a remote video surveillance data transmission method further includes the following: S301, prioritize all devices in the slice according to specific business requirements and network resource usage.

[0061] The sorting criteria include factors such as the criticality of the device, the real-time requirements of the video data stream, and the historical video data flow.

[0062] Specifically, key indicators of all devices are collected, such as the severity of the consequences of failures (for example, 1-3 points for minor impact, 4-6 points for medium impact, 7-9 points for severe impact, and 10 points for catastrophic impact), the position and role in the business process (core processing nodes and data convergence points have the highest weights), the delay tolerance required by the real-time requirements of video data streams (for example, the delay threshold for services with high real-time requirements is set to 10 points for 1-10 milliseconds, 5 points for medium 10-100 milliseconds, and 1 point for low 100 milliseconds), the urgency of video data streams, etc.; based on the collected information, the criticality, real-time requirements, urgency and other factors of the device are taken into consideration, a priority score is calculated for each device, and all devices are sorted according to the calculated priority scores, with the lower the score, the lower the priority of the device. Taking the three weights of the key indicators of device criticality, video data stream real-time requirements, and video data stream urgency as examples, the device priority score calculation formula is: B=w1*K+w2*R+w3*F (1) Among them, B represents the priority score of the device, w1, w2, and w3 represent the weights of device criticality, video data stream real-time requirements, and video data stream urgency, respectively, and w1+w2+w3=1. The specific value of the weight should be determined according to the actual business needs and network environment, and can usually be obtained through expert scoring or AHP hierarchical analysis method (decomposing the problem into different components, such as device criticality, video data stream real-time requirements, video data stream urgency, etc. Then, by comparing the importance of these factors, a judgment matrix is ​​constructed. The judgment matrix is ​​solved using the eigenvalue method to obtain the weight of each factor. Finally, a consistency test is performed to ensure the rationality of the weight. The specific usage method refers to related technologies and is not described in detail in this article). K represents the device criticality index, which can be quantified according to the severity of the consequences of the failure. The higher the score, the more critical the device and the more serious the consequences of the failure, such as using a score of 1 (lowest criticality) to 10 (highest criticality); R represents the real-time requirements of the video data stream, which can be quantified according to the delay tolerance. The higher the score, the higher the real-time requirements and the lower the delay, such as using the inverse of the delay threshold or grading the score according to the delay requirements. F represents the urgency of historical video data. The specific urgency needs to be set according to the actual situation. The lower the score, the higher the urgency.

[0063] S302: When adjusting slice resources, select a device with a low priority for slice switching.

[0064] Specifically, according to the device priority sorting results, devices with priority P<4 are selected as switching candidates. Devices with a priority less than 4 are considered low-priority devices. The video data streams of low-priority devices are relatively less critical or have lower real-time requirements. By giving priority to switching low-priority devices, the impact on key services is reduced, and the performance and stability of the new slice are tested at the same time.

[0065] Before switching, confirm that the candidate device is operating normally and there is no ongoing critical business, and back up important video data on the candidate device to prevent video data loss during the switching process. At the same time, ensure that the configuration and resource allocation of the new slice are ready, including network parameter settings (the slice's identification, frequency band, bandwidth, transmission mode, etc. have been configured according to business requirements and network planning), resource allocation (sufficient resources have been allocated to the new slice according to the business requirements of the slice and the expected network load, and ensure that these resources will not be occupied by other services during the operation of the slice), security policy (the corresponding security policy has been configured for the new slice, such as access control, data encryption), etc., and these configurations have passed preliminary verification and testing to ensure that the slice can work normally, and perform necessary pre-tests (functional testing verifies whether the new slice supports all the expected functions of video data transmission, session management, and mobility management; performance testing verifies the new slice's latency, throughput, and packet loss rate performance indicators to ensure that business requirements are met; stability testing tests the stability of the new slice under long-term operation and high-load conditions to ensure that services can be provided continuously and stably; compatibility testing verifies the compatibility of the new slice with existing devices and systems to ensure that it can work normally with other devices and systems after switching).

[0066] S303, after the communication status of the switched device is verified, the device video data stream is set to send video data to the original slice and the new slice simultaneously within 0.5ms.

[0067] Among them, a serial number timestamp is added to the video data packet to ensure the order and synchronization of the video data; a CRC check is used to ensure that no error occurs during the transmission of the video data, and when it reaches 0.5ms, the video data is automatically stopped from being sent to the original slice. If the new slice fails to switch due to timeout, a rollback is performed, and video data is stopped from being sent to the new slice, the connection with the original slice is restored, and the video data stream is resent to the original slice. After confirming that the video data stream is received normally, the new slice is selected to continue the slice switching operation.

[0068] S304, the decoder synchronously decodes and compares with the video data.

[0069] Specifically, first ensure that the decoder has sufficient processing power and storage space to support synchronous decoding. The video data streams from the two slices are synchronously decoded at the decoder end. At the same time, the decoded video data is compared to verify whether the video data stream sent by the new slice is consistent with the video data of the original slice and whether the time is synchronized. The video data stream comparison process uses the Network Time Protocol (NTP) to ensure time synchronization at the decoder end and achieve byte-by-byte comparison. When the video data stream is found to be inconsistent, a rollback is performed, and the video data stream returns to the original slice and resends the video data stream to the slice.

[0070] S305, after the video data comparison and verification is passed and it is confirmed that the video data transmitted by the new slice is consistent with the original slice video data and time synchronized, a switching instruction is issued to stop sending video data to the original slice and completely switch to the new slice for video data transmission.

[0071] Among them, the switching instruction content includes the target slice information of the new slice's identification, IP address, and port number, and the subsequent operation instructions for disconnecting the connection with the original slice and establishing a connection with the new slice. The TCP communication protocol is used to ensure the accurate communication of the instruction acceptance event and the execution result. After receiving the instruction, the device should return a confirmation message to ensure that the instruction has been executed correctly. If the confirmation message is not received, it can be considered that the instruction has failed to be sent. At this time, the instruction can be resent or other processing logic can be executed. Verify whether the switch is successful by monitoring the operating status of the new slice and the transmission of video data.

[0072] S306: After the switch is completed, the performance and stability of the new slice are continuously monitored, and the slice configuration and resource allocation are adjusted in a timely manner according to the monitoring results and feedback.

[0073] Among them, monitoring indicators include throughput, latency, packet loss rate, CPU load, memory usage, etc.; the monitoring frequency is set to every 5 minutes according to business needs and network status.

[0074] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: 1. By considering the priority of the device, a scientific priority sorting mechanism is established for all devices in the slice. This mechanism not only ensures the high priority processing of key business and high real-time demand devices, but also provides an accurate basis for subsequent resource adjustments.

[0075] 2. By switching low-priority devices first, the impact on key services is reduced while ensuring the performance and stability of the new slice. During the switching process, the integrity and reliability of video data transmission are guaranteed through video data synchronization, verification and rollback mechanisms. After the switch is completed, the performance and stability of the new slice are continuously monitored, and the configuration and resource allocation are adjusted and optimized in a timely manner according to the monitoring results, which further improves the flexibility and efficiency of slice resource adjustment and provides a more stable and reliable video data stream transmission environment.

[0076] Embodiment 3: Embodiment 2 above ensures that critical video streams always have sufficient bandwidth and priority by adjusting slice resource allocation. The dual-slice transmission and synchronous decoding comparison mechanism significantly reduces the short delay of video data caused by slice switching. To ensure the real-time and stability of video data, temporary slices are set separately to handle the load overflow of each slice, which is used to take over the overflow video data stream of each slice, thereby reducing the video data transmission deviation caused by switching slices. Further improvements are made on the basis of embodiment 2.

[0077] In some embodiments, selecting a device with a low priority for slice switching specifically includes: S401, set up a temporary dedicated slice separately and define the strategy.

[0078] Specifically, a dedicated temporary dedicated slice is created to receive the overflow video data streams of each slice. A unique identifier is set for each video data stream, and the original transmission strategy and requirements, including bandwidth, delay, packet loss rate, etc., are maintained for each video data stream in the temporary dedicated slice to ensure that the video data stream can still maintain its original transmission quality and performance after being migrated to the temporary dedicated slice.

[0079] S402, monitor the slice load and resource allocation to avoid frequent switching and unstable video data transmission caused by slice overload.

[0080] Specifically, a load threshold and resource usage threshold are set for each slice. When the load or resource usage of a slice exceeds the threshold, the early warning mechanism is triggered. When the early warning mechanism detects that the slice load or resource usage exceeds the threshold, it will issue an early warning signal in a timely manner. The load threshold and resource usage threshold are set according to specific business needs and the load threshold and resource usage threshold of existing slices.

[0081] S403, for the video data stream that needs to be migrated, according to its transmission strategy and requirements in the original slice, after performing corresponding replication configuration in the temporary dedicated slice, the video data stream is migrated to the temporary dedicated slice.

[0082] S404, adjusting the load of the temporary dedicated slice and its resource allocation according to the transmission strategy and requirements of the migrated slice.

[0083] S405, for a temporary slice that has occupied video data streams for a long time and has an excessively high load and reduced video data stream transmission quality, assess the feasibility of migrating the temporary slice back to the original slice or allocating it to a slice with appropriate resources.

[0084] Among them, the feasibility of migrating back to the original slice is evaluated, including whether the original slice resources have been restored to accommodate the migrated video data stream, whether the service quality consistent with the original transmission strategy can be guaranteed after migrating back to the original slice, and the impact of migration on the stability of the original slice and the entire network. Resource-appropriate slices include slice bandwidth, computing resources, latency, packet loss rate, bandwidth fluctuations, etc. to meet allocation requirements.

[0085] S406, regularly evaluate the usage of temporary dedicated slices and continuously improve configuration strategies and resource allocation.

[0086] The regular time period may be one week, and the specific time depends on actual business needs.

[0087] Specifically, monitoring probes are deployed at key nodes of 5G network slices to track the operation status of slices in real time, and the network management tool SNMP is used to regularly collect video data on the use of temporary dedicated slices, including bandwidth usage, latency, packet loss rate, resource utilization, etc. The collected video data is analyzed and compared with the preset performance indicators and thresholds to evaluate the actual performance of the slices and check whether there are bottlenecks or anomalies, such as resource overload, configuration errors, or security vulnerabilities.

[0088] For example, for the usage of temporary dedicated slices for power plant inspection, assume that the period of regular evaluation is set to one week. Bandwidth usage is the proportion of the bandwidth currently used by the slice to the total bandwidth. The preset performance threshold is 70%. If it exceeds 70%, it indicates that the resource is close to overload; delay is the average time delay from the sending to the receiving of video data packets. The preset performance threshold is 20ms. If it exceeds 20ms, it may affect the efficiency of real-time inspection; packet loss rate is the proportion of video data packets lost during transmission. The preset performance threshold is 0.05%. If it exceeds 0.05%, it may affect the quality of video data transmission; resource utilization (CPU) is the proportion of CPU resources used by the slice. The preset performance threshold is 60%. If it exceeds 60%, it may indicate that computing resources are tight; resource utilization (memory) is the proportion of memory resources used by the slice. The preset performance threshold is 70%. If it exceeds 70%, it may indicate that memory resources are tight. The above data thresholds can be used as a reference for setting basic thresholds.

[0089] When the average bandwidth utilization this week is 65%, the average latency is 18ms, the average packet loss rate this week is 0.03%, the average CPU utilization is 58%, and the average memory utilization is 65%, all of which are within the preset thresholds, the performance of the temporary dedicated slice is good, and all indicators are within the preset performance thresholds, and no emergency optimization is required.

[0090] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: 1. By setting up temporary dedicated slices separately and defining clear policies to manage video data streams, efficient resource utilization is achieved. The core of this advantage is that it can dynamically adjust resource allocation between slices according to actual business needs and network load conditions. When a slice is short of resources due to business peaks or emergencies, the system can quickly migrate part of the video data stream to the temporary dedicated slice, thereby alleviating the pressure on the original slice and avoiding performance degradation caused by resource overload. At the same time, since the temporary dedicated slice is specially designed to take over overflow video data streams, it can use network resources more efficiently and reduce resource waste and idleness.

[0091] 2. During the video data stream migration process, for the video data stream that needs to be migrated, according to its transmission strategy and requirements in the original slice, the corresponding replication configuration is performed in the temporary dedicated slice to ensure that the video data stream can still maintain its original transmission quality and performance after being migrated to the temporary dedicated slice. The realization of this advantage depends on the unique identifier set by the system for each video data stream, and the original transmission strategy and requirements maintained for each data stream in the temporary dedicated slice.

[0092] Embodiment 4: Embodiment 3 above increases temporary dedicated slices, and each video data stream maintains the original transmission strategy, thereby reducing the video data transmission deviation caused by switching slices. If the slices with similar internal video data transmission requirements are grouped, a corresponding relationship is formed with the various slices initially set, and then it can be determined which split slice the overflow video data stream enters. In this way, each task has a fallback mechanism to avoid frequent slice modifications and make up for the shortcomings of the initial settings. For this reason, further improvements are made on the basis of Embodiment 3.

[0093] S501, setting an initial general slice to carry the transmission task of various video data streams.

[0094] Specifically, a general 5G network slice is initially set up, which is designed to handle various types of video data streams and configure the network functions of the slice. According to the expected number and type of video streams, combined with the actual carrying capacity of the network and future business expansion needs, the initial bandwidth is set to 100-500Mbps, and is adjusted according to the expected number and type of video streams; the delay should be less than 100ms to ensure the smoothness of real-time video; the packet loss rate is less than 0.1% to reduce video freezes.

[0095] S502, start the slice monitoring mechanism to track the slice load, video data transmission requirements and performance.

[0096] Specifically, the deployed real-time monitoring tool is launched to track the load of the slices, the video data transmission requirements, analyze the video data situation, and identify problems such as load peaks, video data transmission bottlenecks, and performance degradation within specified and reasonable load thresholds.

[0097] S503 , monitoring the video data based on the slices, identifying video data streams with similar transmission requirements, and dividing the video data streams into different groups according to the analysis results.

[0098] Among them, similar transmission requirements include bandwidth requirements, delay tolerance, priority, etc.

[0099] Specifically, the clustering algorithm K-means is used to classify the video data stream, collect the key attribute data of the video data stream, clean and standardize the data, select the appropriate K value (i.e. the number of classifications) based on actual needs and data characteristics, and randomly select K data points as the initial centroid. Assign each data point to the class where the nearest centroid is located, and recalculate the centroid position of each class. Repeat the above steps until the centroid position no longer changes significantly or reaches the predetermined number of iterations. Set a label for each group of video data streams based on the classification results, and set a similarity threshold to determine whether the video data streams belong to the same group.

[0100] S504, splitting the initial slice into multiple sub-slices according to the grouping result, each sub-slice is specially used to carry a video data stream of a specific group and inherits the transmission strategy and requirements of the video data stream of the group.

[0101] S505, for the newly added video data stream, the most suitable sub-slice is selected for carrying according to its transmission requirements through the intelligent matching algorithm, and the slice monitoring mechanism is updated to continuously track the newly added sub-slice. The intelligent matching algorithm refers to the existing technology and will not be described in detail in this article.

[0102] Specifically, matching rules are defined based on the classification results of the video data stream and the characteristics of each sub-slice (such as bandwidth, delay, priority, etc.). When a new video data stream is connected, the most suitable sub-slice is quickly determined for carrying based on its attributes and matching rules using an intelligent matching algorithm.

[0103] S506: Dynamically adjust the resource configuration of each sub-slice according to the load, performance and resource utilization of each sub-slice.

[0104] Specifically, for overloaded sub-slices, consider further optimizing their internal video data flow management or consider further splitting; for sub-slices with low resource utilization, consider merging or reclaiming some resources.

[0105] S507, regularly evaluate the slice splitting and grouping management strategies, and iteratively optimize the slice splitting strategies and grouping rules according to the evaluation results.

[0106] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: 1. Slice splitting and grouping management can significantly improve the transmission efficiency and resource utilization of video data streams. Through detailed analysis based on monitoring data, video data streams with similar transmission requirements are divided into different groups, and the initial slice is split into multiple sub-slices accordingly. Each sub-slice specifically carries the video data stream of a specific group. This strategy ensures accurate allocation of resources and avoids resource waste and reduced efficiency caused by mixed transmission of different types of data streams. Sub-slices can be optimized for the type of data stream they carry, such as adjusting parameters such as bandwidth, delay, and priority, to provide more efficient and stable transmission services.

[0107] 2. As new video data streams are continuously connected, the system can quickly allocate them to the most appropriate sub-slices through intelligent matching algorithms without making large-scale adjustments to the entire slice architecture. This plug-and-play capability enables the system to easily cope with business growth and changes, maintaining a high degree of flexibility and responsiveness. At the same time, the group management strategy allows the system to dynamically adjust the number and configuration of sub-slices according to actual needs, leaving ample room for future business expansion.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A remote video monitoring data transmission method, characterized in that: include: S101, collecting robot visual video information data through a robot camera, and transmitting the visual video information data to an encoder through an HDMI interface; S102, the encoder encodes the visual video information data, and transmits the encoded visual video information data to the 5G-CPE through the network port; S103, 5G-CPE transmits the encoded visual video information data to the decoder, and uses MEC, 5G slicing technology and SRT network transmission technology, and adjusts 5G slice resource allocation according to real-time network conditions and transmission requirements; S104, the decoder decodes the received encoded visual video information data, and transmits the decoded visual video information data to a remote monitoring center of the power plant, and finally presents it on a display screen of the monitoring center.

2. A remote video monitoring data transmission method as claimed in claim 1, characterized in that: The S103 adjusts the 5G slice resource allocation method according to the real-time network status and transmission requirements, including: S201, collect and analyze specific business requirements in power plant inspection scenarios, define and configure multiple 5G slices according to the real-time, bandwidth, and reliability requirements of different application scenarios of power plant inspection, and assign priorities; S202, deploy network monitoring components to monitor the load of 5G slices in real time and collect KPIs of each slice; S203, setting a load threshold for each slice, and triggering a resource adjustment policy mechanism when the KPIs of a slice are monitored to exceed the set threshold; S204: Analyze and evaluate whether other 5G slice resources have available remaining resources for dynamic resource reallocation, increase resources for high-load 5G slices according to the evaluation results, and migrate low-priority video streams from high-load 5G slices to idle slices; S205: After the resources are adjusted, the transmission performance of the slices is continuously monitored, and the resource adjustment strategy is adjusted based on the monitoring results.

3. A remote video monitoring data transmission method as claimed in claim 2, characterized in that: S201, defining and configuring multiple slices according to the real-time, bandwidth, and reliability requirements of different application scenarios of power plant inspection, includes: According to the different scene requirements of power plant inspection, the service type and performance indicators of each slice are clarified, and the network architecture, resource allocation strategy, isolation mechanism, etc. of the slice are designed; Create a new slice instance in the network management system, assign a unique identifier to each slice, and allocate network resources according to slice requirements, including wireless resources, core network resources, and transmission resources.

4. A remote video monitoring data transmission method as claimed in claim 2, characterized in that: The step S201, allocating priorities, includes: Assign a QCI-P priority level to each video stream. The priority is set based on factors such as the content and importance of the video stream. Daily inspection video streams have a lower priority. This type of video stream is mainly used to record the daily operation status of the equipment, and has relatively low requirements for real-time and bandwidth; emergency inspection video streams are set as the highest priority ,This type of video stream is usually generated after a device anomaly is discovered or an alarm is received, and video information data needs to be transmitted immediately, which has high requirements for real-time performance, bandwidth, and reliability; Each element in the queue is a video stream request, including information such as the video stream identifier, priority, bandwidth requirement, and real-time requirement. When a new video stream request arrives, it is added to the priority queue and sorted according to its priority. For each video stream request, slice resources are dynamically allocated according to its bandwidth requirements and real-time requirements. According to the priority and bandwidth requirements of the video stream request, appropriate resources are allocated to the video stream from the available slice resources.

5. A remote video monitoring data transmission method as claimed in claim 2, characterized in that: The step S201, allocating priorities, further includes: S301, prioritize all devices in the 5G slice according to specific business requirements and network resource usage; S302, when deciding to adjust 5G slice resources, give priority to devices with lower priority for 5G slice switching; S303, after the communication status of the switched device is verified, set the device to send data to the original slice and the new slice simultaneously within x seconds; S304, decoder synchronous decoding and data comparison; S305, after the video data comparison and verification pass, confirming that the video data transmitted by the new slice is consistent with the video data of the original slice and time synchronized, issue a switching instruction to stop sending video data to the original slice and completely switch to the new slice for video data transmission; S306: After the switch is completed, the performance and stability of the new slice are continuously monitored, and the slice configuration and resource allocation are adjusted in a timely manner according to the monitoring results and feedback.

6. A remote video monitoring data transmission method as claimed in claim 5, characterized in that: S301, prioritizing all devices in the 5G slice, includes: Collect key indicators of all devices, take into account factors such as device criticality, real-time requirements, and urgency based on the collected information, calculate a priority score for each device, and sort all devices based on the calculated priority scores. The lower the score, the lower the priority of the device. The device priority score is calculated based on the three weights of key indicator device criticality, video data stream real-time requirements, and video data stream urgency. The device priority score is calculated using the following formula: B=w1*K+w2*R+w3*F; B represents the priority score of the device, w1, w2, and w3 represent the weights of device criticality, video data stream real-time requirement, and video data stream urgency, respectively, and w1+w2+w3=1.

7. A remote video monitoring data transmission method as claimed in claim 5, characterized in that: The S304, decoder synchronous decoding and data comparison, includes: Ensure that the decoder has sufficient processing power and storage space to support simultaneous decoding; Synchronously decoding the video data streams from the two slices at the decoder end; At the same time, the decoded video data is compared to verify whether the video data stream sent by the new slice is consistent with the original slice video data and whether the time is synchronized; The video data stream comparison process uses the NTP network time protocol to ensure time synchronization on the decoder side, so that byte-by-byte comparison can be achieved; When the video data stream is found to be inconsistent, a rollback is performed, and the video data stream returns to the original slice and resends the video data stream to the slice.

8. A remote video monitoring data transmission method as claimed in claim 5, characterized in that: The S302, 5G slice switching, includes: S401, set up a temporary dedicated slice separately and define a policy; S402, monitoring slice load and resource allocation; S403: For the video data stream that needs to be migrated, according to its transmission strategy and requirements in the original slice, after performing corresponding replication configuration in the temporary dedicated slice, the video data stream is migrated to the temporary dedicated slice; S404, adjusting the load of the temporary dedicated slice and the resource allocation of the video data transmission demand according to the same transmission strategy and requirement of the migrated slice; S405, for a temporary slice that has occupied a video data stream for a long time and has an excessively high load and a reduced quality of video data stream transmission, assess the feasibility of migrating the temporary slice back to the original slice or allocating the temporary slice to a slice with appropriate resources; S406, regularly evaluate the usage of temporary dedicated slices and continuously improve configuration strategies and resource allocation.

9. A remote video monitoring data transmission method as claimed in claim 8, characterized in that: The S404, adjusting the load of the temporary dedicated slice and the resource allocation according to the video data transmission requirement, includes: S501, setting an initial general slice to carry the transmission task of various video data streams; S502, start the slice monitoring mechanism to track the slice load, video data transmission requirements and performance; S503, monitoring the video data based on the slices, identifying video data streams with similar transmission requirements, and dividing the video data streams into different groups according to the analysis results; S504, splitting the initial slice into multiple sub-slices according to the grouping result, each sub-slice is specifically used to carry a video data stream of a specific group and inherits the transmission strategy and requirements of the video data stream of the group; S505, for the newly added video data stream, the most suitable sub-slice is selected for carrying through the intelligent matching algorithm according to its transmission requirements, and the slice monitoring mechanism is updated accordingly to continuously track the newly added sub-slice; S506, dynamically adjusting resource configuration of each sub-slice according to the load, performance, and resource utilization of each sub-slice; S507, regularly evaluate the slice splitting and grouping management strategies, and iteratively optimize the slice splitting strategies and grouping rules according to the evaluation results.

10. A remote video monitoring data transmission method as claimed in claim 9, characterized in that: The step S503, identifying video data streams with similar transmission requirements, and dividing the video data streams into different groups according to the analysis results, includes: Use the K-means clustering algorithm to classify the video data stream, collect the key attribute data of the video data stream, clean and standardize the data, select the appropriate number of classifications K value based on actual needs and data characteristics, and randomly select K data points as the initial centroid; Assign each data point to the class where the nearest centroid is located, and recalculate the centroid position of each class; Repeat the above steps until the centroid position no longer changes significantly or the predetermined number of iterations is reached; A label is set for each group of video data streams according to the classification result, and a similarity threshold is set to determine whether the video data streams belong to the same group.

Images