A method and system for intelligent centralized monitoring of full-color LED displays

Abstract

This invention provides an intelligent centralized monitoring method for full-color LED displays, comprising: analyzing inter-node dependencies and communication delays based on a neighboring node association model to identify target areas with network segmentation risks; detecting backup communication paths for risky nodes and selecting paths with communication delays below a preset threshold as priority switching paths; deploying a distributed consistency protocol to compare data differences and generating a synchronization order list based on data update frequency and business priority; using a fragmented transmission mechanism to divide the data to be synchronized into multiple data segments and transmitting them to the target nodes in parallel via multiple paths; initiating an automatic repair mechanism for inconsistent nodes, obtaining backup data from neighboring nodes to perform differential correction; and dynamically optimizing communication path weights and synchronization cycle parameters based on the correction results to complete iterative updates of the system status.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to an intelligent centralized monitoring system for a full-color LED display screen. Background Technology

[0002] Full-color LED displays, as an important carrier of modern information display and intelligent interaction, play an indispensable role in urban management, commercial advertising, and public services. Their stability and reliability are directly related to the efficiency of information transmission. However, current management methods mostly rely on centralized control and manual maintenance, which is not only inefficient but also prone to system-wide paralysis due to single points of failure, especially in large-scale distributed scenarios.

[0003] Existing solutions are often limited to simple fault alarms and passive repairs, lacking the ability to adapt to dynamic changes in complex network environments and failing to effectively handle situations where multiple nodes fail simultaneously. This limitation often results in slow response or even data loss in display systems when faced with sudden problems, severely impacting the user experience.

[0004] Looking deeper, the core challenge in this field lies first and foremost in achieving efficient node coordination and management in a distributed environment. Due to the large number and geographically dispersed nature of nodes, traditional static network structures struggle to adapt to dynamic changes, leading to the inability to adjust in time when some nodes fail, thus causing network segmentation. Network segmentation further exacerbates the difficulty of data synchronization, as data loss from failed nodes disrupts the integrity of overall monitoring. Traditional synchronization methods often require the transmission of large amounts of redundant information, increasing the system burden. These interconnected challenges make stable system operation under high load or complex scenarios a major pain point.

[0005] Therefore, how to achieve dynamic node management, rapid fault takeover, and efficient data synchronization in a distributed full-color LED display network has become a key issue in improving system reliability and intelligence. Summary of the Invention

[0006] This invention provides an intelligent centralized monitoring method for full-color LED displays, which mainly includes the following steps:

[0007] The running status data of each node is acquired periodically through a distributed data acquisition mechanism to generate a node status distribution map that includes connection status and performance indicators.

[0008] Based on the neighbor node association model, the dependencies and communication delays between nodes are analyzed to identify target areas with network segmentation risks.

[0009] Perform backup communication path detection on risk nodes and select paths with communication latency below a preset threshold as priority switching paths;

[0010] Deploy a distributed consistency protocol to compare data differences and generate a synchronization order list based on data update frequency and business priority;

[0011] The data to be synchronized is divided into multiple data segments using a fragmented transmission mechanism and transmitted to the target node through multiple paths in parallel.

[0012] After receiving the status update information from each node, the system checks the data integrity through a status consistency verification model and generates a status consistency report that includes abnormal nodes.

[0013] An automatic repair mechanism is initiated for inconsistent nodes, and backup data is obtained from neighboring nodes to perform differential correction.

[0014] Based on the correction results, the communication path weight and synchronization period parameters are dynamically optimized to complete the iterative update of the system state.

[0015] Furthermore, when generating the node status distribution map, a weighted round-robin algorithm is used to calculate the communication quality score of each node. The score includes a comprehensive index of packet loss rate, signal strength, and historical failure rate.

[0016] Furthermore, the neighbor node association model employs an improved PageRank algorithm, which constructs a dependency matrix by analyzing the data flow intensity and topological distance between nodes.

[0017] Furthermore, during the detection of the backup communication path, path probe technology is used to measure bandwidth utilization and link stability in real time, and a path quality assessment model including latency jitter parameters is established.

[0018] Furthermore, the distributed consensus protocol adopts an improved version of the Raft algorithm, which adapts to network topology changes by dynamically adjusting the election timeout mechanism.

[0019] Furthermore, the fragmented transmission mechanism includes the following steps:

[0020] Determine the granularity baseline value of the data to be synchronized based on the business type label;

[0021] The fragment size is dynamically adjusted based on the current transmission path quality, and the fragment size is automatically reduced when the network latency exceeds a preset threshold.

[0022] Assign high-priority paths to urgent business data and prohibit fragmentation to ensure atomicity of transmission.

[0023] Furthermore, the state consistency verification model uses a Merkle tree structure for data digest comparison and sets a difference tolerance threshold to distinguish between repairable anomalies and serious faults.

[0024] Furthermore, the automatic repair mechanism includes a three-level repair strategy: the first level repair uses data cached by adjacent nodes, the second level repair calls regional backup nodes, and the third level repair initiates cloud image recovery.

[0025] Furthermore, the dynamic optimization process employs a reinforcement learning algorithm, using a Q-learning model to train and obtain the optimal path selection strategy.

[0026] On the other hand, the present invention also provides an intelligent centralized monitoring system for a full-color LED display screen, comprising:

[0027] The status acquisition agent module is deployed on each LED display node to execute the distributed data acquisition mechanism.

[0028] A topology analysis server is configured to run the neighbor node association model and the network segmentation risk identification algorithm.

[0029] The path manager, integrated into the central control unit, is responsible for detecting and switching backup communication paths;

[0030] A distributed synchronization engine, comprising the aforementioned fragmented transmission mechanism and consistency protocol execution unit;

[0031] Repair the executor module and trigger a multi-level repair strategy in response to the state consistency report;

[0032] The modules interact with each other through message queues, forming a closed-loop control link.

[0033] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0034] This invention discloses an intelligent centralized monitoring method for full-color LED displays. It involves periodically collecting operational status data from each LED display node, generating a node status distribution map, analyzing node dependencies to identify network segmentation risk areas, detecting backup communication paths, deploying a distributed consistency protocol for data synchronization, employing a fragmented transmission mechanism for parallel data transmission, verifying data integrity and automatically repairing inconsistent nodes, and dynamically optimizing communication parameters. This invention uses a weighted round-robin algorithm to evaluate node communication quality, improves the PageRank algorithm to construct a dependency matrix, uses path probe technology to evaluate backup path quality, improves the Raft algorithm to adapt to network topology changes, formulates a fragmentation strategy based on data features, uses a Merkle tree structure to verify consistency, sets a three-level repair strategy, and applies reinforcement learning to optimize path selection. This invention can achieve efficient status synchronization and fault self-healing in large-scale distributed full-color LED display systems, improving the reliability and performance of such systems. Attached Figure Description

[0035] Figure 1 This is a flowchart of a method for intelligent centralized monitoring of a full-color LED display screen according to the present invention.

[0036] Figure 2 This is a structural diagram of a full-color LED display intelligent centralized monitoring system according to the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] like Figure 1 This embodiment of a method for intelligent centralized monitoring of a full-color LED display screen may specifically include:

[0039] S101. When the node status distribution map is generated, a weighted round-robin algorithm is used to calculate the communication quality score of each node. The score includes a comprehensive index of packet loss rate, signal strength and historical failure rate.

[0040] The core of the weighted round-robin algorithm lies in dynamically allocating node selection weights. In one embodiment, within a large-scale distributed full-color LED display cluster, node A has a packet loss rate of 2%, a signal strength of -70dBm, and a historical failure rate of 0.5%, while node B has corresponding parameters of 5%, -85dBm, and 1.2%. The algorithm assigns a 40% weight to the packet loss rate, 30% to the signal strength, and 30% to the historical failure rate. Therefore, node A's score is 2×0.4 + 70×0.3 + 0.5×0.3 = 24.35, and node B's score is 5×0.4 + 85×0.3 + 1.2×0.3 = 30.86. In this case, the system prioritizes node A, which has the better score.

[0041] The packet loss rate metric is implemented using a sliding window. In one embodiment, the gateway within the LED display cluster samples 100 data packets every 10 seconds. If the packet loss counts for three consecutive windows are 3 / 5 / 4 respectively, the dynamic packet loss rate is calculated as (3+5+4) / 300=4%. This metric reflects real-time network congestion; a high packet loss rate triggers a weight reduction to avoid selecting nodes with unreliable transmission.

[0042] Signal strength employs a graded quantization strategy. RSSI values ​​are divided into six levels, ranging from -60dBm to -120dBm, with each level representing 10dBm. In one embodiment, the signal strength within the LED display cluster is divided as follows: -65dBm corresponds to level 1 (out of 100 points), and -93dBm corresponds to level 4 (out of 60 points). Discretization avoids score oscillations caused by signal fluctuations.

[0043] The historical failure rate incorporates a time decay factor. Assuming an edge node experienced 3 failures in the past 24 hours but has been fault-free in the last 8 hours, the effective number of failures, calculated using the exponential decay formula, is 3 × e^(-8 / 24) = 2.21, resulting in a final failure rate of 2.21 / 24 ≈ 9.2%. This mechanism makes the system focus more on recent stability, preventing historical data from interfering with current decisions.

[0044] The comprehensive indicators eliminate dimensional differences through normalization. In one embodiment, the packet loss rate (0-10%) is mapped to 0-100 points, the signal strength (-60 to -120 dBm) to 100-0 points, and the failure rate (0-5%) to 100-0 points, and then these are summed according to their weights. This standardization method ensures the comparability of indicators across different dimensions, making the scoring results practically instructive.

[0045] A dynamic weight adjustment mechanism is used to handle unexpected situations. When a node's score drops by more than 20% twice consecutively, the system automatically increases the packet loss rate weight to 50%. For example, if a node in an LED display cluster experiences sudden electromagnetic interference, causing the packet loss rate to spike from 3% to 15%, the algorithm quickly reduces the node's priority through weight shifting to ensure the reliability of control command transmission.

[0046] S102. The neighboring node association model adopts an improved PageRank algorithm, which constructs a dependency matrix by analyzing the data flow intensity and topological distance between nodes.

[0047] The traditional PageRank algorithm calculates weights based on webpage link relationships, while the improved version uses data flow intensity (such as bytes per second) and topological distance (such as hop count) as input parameters.

[0048] In one embodiment, node A transmits 1MB of data per second to node B at a distance of 2 hops, while node C transmits 500KB to node B at a distance of 1 hop. The algorithm combines the traffic volume with the distance attenuation factor (such as 20% attenuation per hop). Ultimately, the dependency weight of B on A may be 0.64 (1MB × 0.8²) and on C may be 0.5 (500KB × 0.8), thus constructing an asymmetric dependency matrix.

[0049] The specific method for measuring data flow intensity is as follows: The system calculates the average flow between nodes over a continuous 5-minute period using historical transmission logs, and removes burst noise. In one embodiment, nodes X and Y record flow rates of 800KB / s, 1.2MB / s, and 900KB / s respectively within three time windows. After moving average processing, 950KB / s is taken as the stable flow intensity to ensure that the matrix parameters reflect the actual business load.

[0050] Dynamic adjustment mechanism for topology distance: When the addition of node D shortens the original AB path from 2 hops to 1 hop, the system triggers a topology update event and recalculates the dependency weights. In one embodiment, the new weight from A to B increases from 0.64 to 0.8 (1MB × 0.8¹), while if D has no direct traffic with A, its weight is 0. The corresponding elements in the matrix are updated in real time to adapt to network changes.

[0051] The technical effects of the dependency matrix: This matrix can identify key hub nodes (such as nodes whose total weight exceeds a threshold) and prioritize resource allocation during video stream segmentation. In one embodiment, if a matrix shows that node E is depended on by 8 nodes with a total weight of 5.6, the system will mark it as a high-priority segment receiver, reducing the risk of cascading failures caused by congestion at E. Synergistic effect with path quality assessment: High-weight links in the dependency matrix will be monitored by path probes. For example, due to its 0.8 weight, the AB link's probe sampling frequency is increased from 10 seconds / time to 2 seconds / time, and latency jitter data is synchronously fed back to the matrix to correct weights, forming a closed-loop optimization.

[0052] S103. When detecting the backup communication path, path probe technology is used to measure bandwidth utilization and link stability in real time, and a path quality assessment model including latency jitter parameters is established.

[0053] The implementation method of path probe technology is as follows: In backup communication path detection, the path probe periodically sends probe data packets (e.g., 100-byte UDP packets every 5 seconds) to the target node, recording the round-trip time (RTT) and packet loss rate. In one embodiment, when the primary path RTT exceeds 50ms, the probe triggers backup path evaluation and simultaneously collects the response time of each hop in the path, forming a link state snapshot. This technology identifies instantaneous jitter by comparing historical data (e.g., the average over the past 10 minutes), avoiding misjudgment of path quality due to single measurement errors.

[0054] The specific operation of bandwidth utilization measurement is as follows: A sliding window statistical method is used to calculate the link throughput percentage every 30 seconds. In one embodiment, if the bandwidth utilization of a certain path exceeds 80% for three consecutive times (the threshold is configurable), it is marked as a congested path. Service priorities are distinguished during measurement; for example, video stream probe packets are prioritized over ordinary data packets to ensure the accuracy of measurements for critical services.

[0055] Link stability assessment model construction: Link outage frequency and recovery time are calculated based on probe data. In one embodiment, if a path experiences more than three latency fluctuations exceeding 200ms within one hour, the stability score decreases by 50%. The model incorporates an exponentially weighted moving average algorithm, giving higher weight to recent fault data (e.g., 70% from the past hour) to quickly reflect topology changes.

[0056] Quantification of latency jitter parameters: Jitter values ​​(e.g., standard deviation) are divided into four levels: 0-10ms is excellent, 10-30ms is good, 30-50ms is average, and above 50ms is poor. The evaluation model combines jitter level with duration (e.g., if 5 consecutive jitters are "average", the level is downgraded) and dynamically adjusts the path weights.

[0057] The decision logic of the path quality assessment model is as follows: a total score is generated by combining bandwidth (weight 40%), stability (weight 30%), and latency jitter (weight 30%). In one embodiment, when the total score of the primary path falls below 60 points, it automatically switches to a backup path (requiring a total score ≥ 75 points). The model sets a hysteresis threshold (5-point difference) to prevent frequent switching and dynamically optimizes the weight allocation through reinforcement learning. Compared to traditional heartbeat detection, the system can identify path degradation faster through real-time probes and multi-dimensional evaluation.

[0058] S104. The distributed consensus protocol adopts an improved version of the Raft algorithm, which adapts to network topology changes by dynamically adjusting the election timeout mechanism.

[0059] When a distributed consensus protocol employs an improved version of the Raft algorithm, the core purpose of dynamically adjusting the election timeout mechanism is to address the election conflicts or delays caused by fixed timeouts in traditional Raft when network topology changes frequently. The improved algorithm dynamically calculates the election timeout threshold for each node by monitoring inter-node communication latency and topology changes in real time. In one embodiment, when the average round-trip latency between a node and the Leader node increases from 100ms to 300ms, the election timeout for that node is adjusted from the original range of 150ms-300ms to 300ms-600ms, preventing invalid elections triggered by network fluctuations.

[0060] The specific implementation of dynamic adjustment relies on link quality data periodically collected by the topology-aware module. This module measures latency jitter and packet loss rate between nodes through heartbeat packets. When the packet loss rate between nodes in a certain area exceeds 5%, it is determined to be a topologically unstable state. At this time, the Raft algorithm will increase the election timeout baseline value of the affected nodes by 50%. In one embodiment, the average latency between node A and the majority of nodes in the cluster is 200ms ± 50ms. When a sudden increase in latency to 500ms ± 200ms and a packet loss rate of 8% are detected, its election timeout will be adjusted from the default 150ms-300ms to 225ms-450ms, thereby reducing the risk of multiple leaders caused by network partitioning.

[0061] The improved algorithm also incorporates historical topology change trend prediction. By analyzing the Markov chain model of node connection states over the past 10 cycles, the stability probability of the next cycle is predicted. If a node is predicted to have a 70% probability of entering a high-latency state, its election timeout is increased by 20% in advance. In one embodiment, if node B experiences latency fluctuations exceeding 200ms for three consecutive cycles, the system predicts a 75% probability of instability in the next cycle. Therefore, the node's timeout is adjusted from 200ms to 240ms to reduce election storms caused by unpredictable timeouts.

[0062] The technical benefits are reflected in the following three aspects: First, dynamic adjustment matches the election timeout with the current network conditions, greatly shortening the cluster's downtime; second, by setting differentiated timeout thresholds for each node, large-scale nodes can avoid simultaneously initiating voting, reducing bandwidth contention; and finally, the prediction mechanism enables the system to anticipate topology changes, reducing ineffective leader switching.

[0063] S105, The fragmented transmission mechanism includes the following steps:

[0064] Determine the granularity baseline value of the data to be synchronized based on the business type label;

[0065] The fragment size is dynamically adjusted based on the current transmission path quality, and the fragment size is automatically reduced when the network latency exceeds a preset threshold.

[0066] Assign high-priority paths to urgent business data and prohibit fragmentation to ensure atomicity of transmission.

[0067] Based on the service type tags of the data to be synchronized, the corresponding service priority and data characteristics are extracted to determine the baseline value for fragmentation granularity. In one embodiment, video stream data adopts a keyframe-first fragmentation rule, setting the baseline fragment size to 1MB. Control commands adopt a whole-packet transmission mode, with the baseline fragment size being the size of the complete data packet. Network parameters of the current transmission path are collected in real time to analyze network latency and bandwidth utilization. In one embodiment, when the packet loss rate exceeds 5% or the signal strength is below -80dBm, a fragmentation adjustment mechanism is triggered to obtain the basis for dynamic adjustment of the fragment size. If the network latency is higher than a preset threshold (e.g., 100ms), a fragment size reduction strategy is triggered, adjusting the fragment size to 50% of the original baseline value. The adjusted fragment size is calculated to determine the new fragmentation scheme. Based on the service type tags and priority identifiers, urgent service data, such as data marked as real-time control commands, is filtered out to obtain its transmission path allocation requirements. The availability of high-priority paths is determined. If a high-priority path meets the stability requirements, the urgent service data is allocated to that path, for example, selecting the node path with the highest communication quality score to determine the transmission channel. Based on the atomicity requirements of emergency business data, rules prohibiting fragmentation are set. For example, control command data is transmitted directly in whole packet mode, and transmission mode for obtaining complete data packets is adopted.

[0068] S106. The state consistency verification model uses a Merkle tree structure for data digest comparison and sets a difference tolerance threshold to distinguish between repairable anomalies and serious faults.

[0069] Merkle trees aggregate data blocks into a tree structure layer by layer using a hash algorithm. Each non-leaf node stores a combined hash of the hash values ​​of its child nodes. In one embodiment, after dividing 1GB of data into 1024 1MB data blocks, a SHA-256 hash value is calculated for each block. The hash values ​​of adjacent two blocks are concatenated and hashed again, recursively forming the root hash. When node A and node B perform data calibration, differences can be quickly found by comparing the root hash value. If they are inconsistent, the specific difference block is located layer by layer downwards, reducing the complexity of traditional full comparison from O(n) to O(log n).

[0070] The system employs a dynamic method for setting the difference tolerance threshold: Two threshold levels are defined based on the service type. For sensor data from LED display nodes, a tolerance threshold of 0 (zero difference) is set, while for streaming media data, a 3% block hash mismatch is allowed. In one embodiment, when a Merkle tree comparison between nodes reveals 2% of leaf node hash inconsistencies, the system determines it as a repairable anomaly and automatically triggers data synchronization between adjacent nodes. If the difference exceeds 5%, it is determined as a serious fault, initiating a three-level repair mechanism. The threshold parameters are dynamically adjusted based on historical network packet loss rates, node stability, and other indicators to avoid misjudgments.

[0071] The logic for resolving anomalies is as follows: When the Merkle tree detects an inconsistency in the intermediate layer hash of a video segment, but the hash of the branch containing the keyframe matches, the system only retransmits the differing segment. In one embodiment, the hash value of the left branch of the third level of the Merkle tree of node A is X1, and the corresponding branch of node B is X2, but the hash of the keyframe branch of both is Y. At this time, the system marks the segment as "partially corrupted" and prioritizes the synchronization of non-keyframe data to maintain playback continuity.

[0072] Severe Fault Handling Procedure: If the Merkle root hash of a node does not match that of 90% of the nodes in the cluster, and the differing blocks are distributed across multiple subtrees, the system determines it to be a severe fault. In one embodiment, if storage shards 1, 3, and 7 of node C are detected to have hash breaks simultaneously, the node is immediately isolated, a complete data image is pulled from the regional backup node, and the topology routing table is updated to avoid the node.

[0073] By adopting a Merkle tree structure, the system can quickly locate differences from a large amount of data in a very short time, significantly improving efficiency compared to traditional CRC checksums. The difference tolerance threshold mechanism ensures strong consistency of node sensor data while guaranteeing eventual consistency of other non-critical data, thereby reducing resource consumption.

[0074] S107. The automatic repair mechanism includes a three-level repair strategy: Level 1 repair uses data cached by adjacent nodes, Level 2 repair calls regional backup nodes, and Level 3 repair initiates cloud image recovery.

[0075] The first-level repair utilizes data cached by adjacent nodes: This strategy is based on the principle of local data redundancy. When an anomaly is detected in node A's data, it preferentially retrieves the most recently synchronized data copy from the caches of its directly connected nodes B and C. In one embodiment, when node A loses a video stream fragment, the system queries the cache of node B for keyframe data (such as H.264 encoded I-frames) stored within the last 3 seconds, and selects the latest version for overwriting repair by comparing timestamps. This stage is designed for millisecond-level response (typically <50ms). Because it only involves single-hop communication and the data volume is small (e.g., a single keyframe is about 5KB), it avoids cross-regional transmission overhead. It can quickly repair high-frequency, small-scale anomalies (such as packet loss caused by instantaneous network jitter) while reducing dependence on upper-level nodes.

[0076] Secondary repair calls the regional backup node: When adjacent nodes have incomplete caches (e.g., nodes B and C are both missing keyframes), the system will send a request to backup node D in the same region (such as the same control room or subnet). For example, in a video stream fragment repair scenario, node D stores complete time window data (e.g., the GOP group of the most recent 10 seconds). After locating the difference fragments by comparing the Merkle tree root hash value, it is transmitted. The latency at this stage is controlled within 200ms, suitable for medium-scale data corruption (e.g., node A loses 20% of its fragments). The regional backup node uses a periodic snapshot mechanism (e.g., generating incremental backups every 5 seconds) to balance storage overhead and recovery efficiency. This strategy trades space for time to solve the problem of localized multi-node cascading failures.

[0077] Level 3 repair initiates cloud image recovery: For global failures (such as regional backup node downtime), the system pulls a complete image from cloud storage. For example, if node A loses all video stream data due to hard drive failure, the cloud image provides point-in-time recovery capabilities (such as timestamp indexes accurate to the second), with accelerated transmission via CDN edge nodes. This stage takes a relatively long time (approximately 2-5 seconds), but ensures ultimate data consistency. The cloud image employs a tiered storage design, storing hot data (the most recent hour) on SSDs and archiving cold data to object storage. As a fallback solution, the technical value of Level 3 repair lies in maintaining system robustness in the face of catastrophic failures.

[0078] The collaborative logic of the three-tier strategy: The system triggers strategy upgrades based on difference thresholds. For example, when the detected data difference rate is <5%, only Level 1 repair is initiated; when the difference rate is 5%-30%, Level 2 repair is activated; and when the difference rate is >30%, it directly jumps to Level 3 repair. This hierarchical mechanism achieves optimal resource allocation by dynamically assessing the scope of fault impact (e.g., single node / multi-node / global) and repair costs (bandwidth, latency, computing resources). For example, in a combined LED display system playing video, Level 1 repair ensures lip-sync, Level 2 repair handles local stuttering, and Level 3 repair addresses service interruptions, forming a fault-tolerant system covering the entire scenario.

[0079] S108. The dynamic optimization process adopts a reinforcement learning algorithm and obtains the optimal path selection strategy through Q-learning model training.

[0080] Q-learning is a model-free reinforcement learning algorithm that learns the optimal policy by iteratively updating the Q-value table. In path selection scenarios, the system treats network nodes as states, available paths as actions, and path latency, bandwidth, and other metrics as rewards. In one embodiment, when node A needs to transmit data to node B, the Q-learning model evaluates all possible paths (such as A→C→B or A→D→E→B) and calculates the Q-value for each path based on historical data.

[0081] In the initial stage, the Q-value is randomly initialized. The model gradually updates the Q-value by balancing exploration (randomly selecting paths) and exploitation (selecting known high Q-value paths). In one embodiment, assuming the initial Q-value of path A→C→B is 0, the measured latency of the first transmission is 50ms, the reward is set to -50, the learning rate α=0.1, and the discount factor γ=0.9, then the updated Q-value is: 0 + 0.1×(-50+ 0.9×0 - 0) = -5. After multiple iterations, the model converges to the optimal path strategy.

[0082] To adapt to dynamic network environments, Q-learning models need to perceive network state changes in real time. In one embodiment, when node C becomes congested, the latency of path A→C→B increases from 50ms to 200ms, and the reward is adjusted to -200. The model will quickly reduce the Q-value of this path and instead choose the alternative path A→D→E→B. This process is achieved through periodic probing, for example, updating link state data every 5 seconds. In one embodiment, the model introduces an ε-greedy strategy (e.g., ε=0.1), selecting the current optimal path in 90% of cases and randomly exploring new paths with a 10% probability, thus avoiding getting trapped in local optima.

[0083] Q-learning's dynamic optimization capabilities are significantly superior to static routing algorithms. For example, in scenarios with sudden traffic surges, the traditional OSPF protocol requires synchronizing the entire network's link-state database, with convergence times potentially reaching the millisecond level. In contrast, the Q-learning model only needs to update local Q-values, completing path switching within milliseconds. Furthermore, the model can flexibly optimize different metrics through reward function design; for instance, setting the reward to "-latency + 0.5 × bandwidth" can balance transmission speed and throughput requirements.

[0084] like Figure 2 As shown, this invention also discloses an intelligent centralized monitoring system for a full-color LED display screen based on any of the above embodiments, comprising:

[0085] The status acquisition agent module is deployed on each LED display node to execute the distributed data acquisition mechanism.

[0086] A topology analysis server is configured to run the neighbor node association model and the network segmentation risk identification algorithm.

[0087] The path manager, integrated into the central control unit, is responsible for detecting and switching backup communication paths;

[0088] A distributed synchronization engine, comprising the aforementioned fragmented transmission mechanism and consistency protocol execution unit;

[0089] Repair the executor module and trigger a multi-level repair strategy in response to the state consistency report;

[0090] The modules interact with each other through message queues, forming a closed-loop control link.

[0091] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for intelligent centralized monitoring of a full-color LED display screen, characterized in that, Includes the following steps: The running status data of each node is acquired periodically through a distributed data acquisition mechanism to generate a node status distribution map that includes connection status and performance indicators. Based on the neighbor node association model, the dependencies and communication delays between nodes are analyzed to identify target areas with network segmentation risks. Perform backup communication path detection on risk nodes and select paths with communication latency below a preset threshold as priority switching paths; Deploy a distributed consistency protocol to compare data differences and generate a synchronization order list based on data update frequency and business priority; The data to be synchronized is divided into multiple data segments using a fragmented transmission mechanism and transmitted to the target node through multiple paths in parallel. After receiving the status update information from each node, the system checks the data integrity through a status consistency verification model and generates a status consistency report that includes abnormal nodes. An automatic repair mechanism is initiated for inconsistent nodes, and backup data is obtained from neighboring nodes to perform differential correction. Based on the correction results, the communication path weight and synchronization period parameters are dynamically optimized to complete the iterative update of the system state.

2. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, When generating the node status distribution map, a weighted round-robin algorithm is used to calculate the communication quality score of each node. The score includes a comprehensive index of packet loss rate, signal strength and historical failure rate.

3. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, The neighbor node association model uses an improved PageRank algorithm to construct a dependency matrix by analyzing the data flow intensity and topological distance between nodes.

4. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, During the detection of the backup communication path, path probe technology is used to measure bandwidth utilization and link stability in real time, and a path quality assessment model including latency jitter parameters is established.

5. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, The distributed consensus protocol adopts an improved version of the Raft algorithm, which adapts to network topology changes by dynamically adjusting the election timeout mechanism.

6. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, The fragmented transmission mechanism includes the following steps: Determine the granularity baseline value of the data to be synchronized based on the business type label; The fragment size is dynamically adjusted based on the current transmission path quality, and the fragment size is automatically reduced when the network latency exceeds a preset threshold. Assign high-priority paths to urgent business data and prohibit fragmentation to ensure atomicity of transmission.

7. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, The state consistency verification model uses a Merkle tree structure for data digest comparison and sets a difference tolerance threshold to distinguish between repairable anomalies and serious faults.

8. The intelligent centralized monitoring method for full-color LED displays as described in claim 7, characterized in that, The automatic repair mechanism includes a three-level repair strategy: Level 1 repair uses cached data from adjacent nodes; Level 2 repair calls regional backup nodes; Level 3 repair initiates cloud image recovery.

9. The intelligent centralized monitoring method for full-color LED displays as described in claim 1, characterized in that, The dynamic optimization process employs a reinforcement learning algorithm, using a Q-learning model to train the optimal path selection strategy.

10. A full-color LED display intelligent centralized monitoring system based on any one of the methods in claims 1-9, characterized in that, Including: The status acquisition agent module is deployed on each LED display node to execute the distributed data acquisition mechanism. A topology analysis server is configured to run the neighbor node association model and the network segmentation risk identification algorithm. The path manager, integrated into the central control unit, is responsible for detecting and switching backup communication paths; A distributed synchronization engine, comprising the aforementioned fragmented transmission mechanism and consistency protocol execution unit; Repair the executor module and trigger a multi-level repair strategy in response to the state consistency report; The modules interact with each other through message queues, forming a closed-loop control link.

Images