An edge-computing-based intelligent park integrated management platform

Abstract

The application relates to the field of smart park management, and particularly discloses a smart park integrated management platform based on edge computing. The application detects communication quality evaluation index information between distributed edge devices and edge servers in a park, judges whether communication between the edge devices and the edge servers is abnormal, further acquires the improvement effect of communication quality between the edge devices and the edge servers after replacing a communication path, analyzes suspected cause types of communication abnormality, and can thus timely find communication faults between terminal devices and servers and make responses, which is beneficial to reliable management of the smart park. The application detects whether edge servers in the park are down, analyzes matching redundant edge servers of the down edge servers according to load conditions, network conditions and geographical positions of the redundant edge servers adjacent to the down edge servers, and takes over the device groups of the redundant edge servers, thereby improving the robustness of smart park management.

Description

Technical Field

[0001] This invention relates to the field of smart park management, and specifically to an integrated smart park management platform based on edge computing. Background Technology

[0002] With the rapid development of technologies such as the Internet of Things and artificial intelligence, smart parks, as an important carrier of urban digital transformation, are gradually evolving towards intelligence and integration. Traditional park management relies heavily on manual inspections and decentralized information systems, resulting in problems such as data silos, delayed responses, and low resource scheduling efficiency. Especially with the expansion of park scale and the surge in the number of devices, the traditional architecture based on centralized cloud computing faces challenges such as high latency, high bandwidth pressure, and insufficient real-time performance, making it difficult to meet the needs of security monitoring, energy management, and equipment collaboration scenarios for low latency, high reliability, and privacy protection.

[0003] Current smart park management platforms typically employ a vertical subsystem design, leading to difficulties in cross-system data interaction and limited dynamic resource allocation capabilities. As the core unit for industrial agglomeration and urban operation, the management efficiency of parks directly impacts resource utilization, operating costs, and user experience.

[0004] Against this backdrop, integrating edge computing technology to build an integrated management platform has become crucial for overcoming bottlenecks. Edge computing effectively reduces data transmission latency and cloud load by pushing computing power closer to the network edge, closer to terminal devices, while supporting localized data processing and privacy protection. Therefore, developing an integrated management platform for smart parks based on edge computing is of great significance for promoting the intelligent upgrading of parks, reducing operation and maintenance costs, and ensuring safe and sustainable development. It is also an urgent need for the construction of new smart cities.

[0005] However, existing edge computing-based smart park management still has some limitations and shortcomings in practical applications.

[0006] For example, Chinese Patent CN111210150A discloses a smart campus system based on edge computing, including a central server and campus servers. The campus servers are used to send data requests; the central server is used to receive data requests and obtain the immediacy requirements and bandwidth pressure of the data requests. If the bandwidth pressure obtained by the central server is greater than a first threshold and the immediacy requirement of the data request is low, the central server puts the data request into a waiting sequence. After the central server obtains that the bandwidth pressure of the target campus server is less than a second threshold, it sends the data request to the corresponding campus server according to the target campus server information in the data request. This invention, by judging the immediacy requirements of data requests and the bandwidth pressure, ensures the processing latency of data requests with high immediacy requirements and reduces the network bandwidth pressure between the central server and the campus servers during peak periods.

[0007] The above-mentioned patent has the following shortcomings: 1. It does not test the communication quality between terminal devices and servers in the park and investigate the reasons for poor communication quality, thus failing to detect communication anomalies between terminal devices and servers in a timely manner and take appropriate measures, which is detrimental to the reliability management of smart parks.

[0008] 2. Each server in the park operates independently, meaning that each server in the park only stores and processes its own data and cannot communicate with adjacent servers in the park. When a server in the park goes down, there is no backup server to take over its equipment group, thus failing to guarantee the stable operation of the smart park management system. Summary of the Invention

[0009] To address the aforementioned issues, this invention proposes an integrated smart park management platform based on edge computing, enabling the management of smart parks.

[0010] The technical solution adopted by the present invention to solve its technical problem is as follows: The present invention provides an integrated management platform for smart parks based on edge computing, including: a device and server communication quality detection module: acquiring communication performance index information between each edge device and the edge server in the park, and evaluating the communication quality between each edge device and the edge server.

[0011] Device-to-server communication anomaly identification module: Based on the evaluation results of the communication quality between each edge device and the edge server, it determines whether the communication between each edge device and the edge server is abnormal, and marks and issues warnings.

[0012] Communication Anomaly Suspected Cause Analysis Module: This module detects the improvement in communication quality between edge devices and edge servers after changing the communication path, analyzes the suspected causes of communication anomalies between edge devices and edge servers, and provides feedback.

[0013] Server downtime detection module: Detects whether each edge server is down and issues an alert.

[0014] Server collaborative disaster recovery module: Based on the load, network conditions and geographical location of each failed edge server and its adjacent redundant edge servers, analyzes the load, network conditions and geographical location of each failed edge server and takes over the corresponding redundant edge servers and their equipment groups.

[0015] Database: Stores network performance evaluation metrics and application layer performance evaluation metrics for communication between edge devices and edge servers, and stores a fault characterization library for edge server downtime.

[0016] Based on the above embodiments, the specific working process of the device-server communication quality detection module includes: extracting network performance evaluation indicators of edge device-edge server communication stored in the database, and obtaining thresholds for latency, jitter, bandwidth, packet loss rate, and error rate of communication between edge device and edge server.

[0017] Based on the latency, jitter, bandwidth, packet loss rate, error rate, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a first evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating this first evaluation coefficient is as follows: in This represents the first evaluation coefficient for the communication quality between the i-th edge device and the edge server, where i = 1, 2, ..., n, and D. i J i B i L i E i Let represent the latency, jitter, bandwidth, packet loss rate, and error rate of the communication between the i-th edge device and the edge server, respectively. Let D′, J′, B′, L′, and E′ represent the thresholds for latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1, ε2, ε3, ε4, and ε5 represent the weights of latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1+ε2+ε3+ε4+ε5=1.

[0018] Based on the above embodiments, the specific working process of the device-server communication quality detection module further includes: extracting the application layer performance evaluation indicators of edge device-edge server communication stored in the database, and obtaining the thresholds for response time and throughput of communication between edge device and edge server.

[0019] Based on the response time, throughput, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a second evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating the second evaluation coefficient for communication quality is as follows: in φ represents the second evaluation coefficient for the communication quality between the i-th edge device and the edge server. i , Let φ′ and t represent the throughput and response time of the communication between the i-th edge device and the edge server, respectively. r ' represents the threshold for throughput and response time, respectively, and α1 and α2 represent the preset weights for throughput and response time, respectively, with α1 + α2 = 1.

[0020] Based on the above embodiments, the specific working process of the device-server communication quality detection module further includes: analyzing the communication quality between the distributed edge devices and edge servers within the park based on the first evaluation coefficient and the second evaluation coefficient of communication quality between the distributed edge devices and edge servers within the park, wherein the calculation formula for communication quality is as follows: Where γ i This represents the communication quality between the i-th edge device and the edge server. The preset communication quality correction factor is represented by e, which represents the natural constant. η1 and η2 represent the weights of the preset first and second evaluation coefficients of communication quality, respectively, and η1 + η2 = 1.

[0021] Based on the above embodiments, the specific working process of the communication anomaly suspected cause analysis module includes: changing the communication path between each marked edge device and the edge server, collecting the latency, jitter, bandwidth, packet loss rate, error rate, response time and throughput of the communication between each marked edge device and the edge server after changing the communication path, and evaluating the communication quality between each marked edge device and the edge server after changing the communication path.

[0022] The communication quality between each marked edge device and the edge server before and after the communication path change is compared. The change in communication quality after the change is obtained, and the improvement effect of the communication quality between each marked edge device and the edge server after the communication path change is analyzed. The formula for calculating the improvement effect of communication quality is as follows: Where λ j Δγ represents the improvement in communication quality between the j-th marked edge device and the edge server after changing the communication path, where j = 1, 2, ..., m. j μ represents the change in communication quality between the j-th marked edge device and the edge server after the communication path is changed. γ This represents the influencing factor corresponding to the preset unit change in communication quality.

[0023] Based on the above embodiments, the specific working process of the communication anomaly suspected cause analysis module further includes: comparing the improvement effect of communication quality between each marked edge device and the edge server after changing the communication path with a preset improvement effect threshold; if the improvement effect of communication quality between a marked edge device and the edge server after changing the communication path is greater than or equal to the preset improvement effect threshold, then the suspected cause type of communication anomaly between the marked edge device and the edge server is communication connection; otherwise, it is a hardware problem; the suspected cause types of communication anomalies between each marked edge device and the edge server are counted and fed back to the park management cloud center.

[0024] Based on the above embodiments, the specific working process of the server collaborative disaster recovery module is as follows: obtain the redundant edge servers within a set range around each edge server that has crashed in the park, and record them as the redundant edge servers adjacent to each edge server that has crashed.

[0025] Monitor the load, network conditions, and geographical location of each downed edge server and its neighboring redundant edge servers. Obtain the number of edge devices connected to each downed edge server, network bandwidth, and relative distance to each neighboring redundant edge server. Analyze the matching degree between each downed edge server and its neighboring redundant edge servers. The matching degree is calculated using the following formula: in This represents the index of the x-th redundant edge server adjacent to the p-th edge server that went down, where p = 1, 2, ..., q, x = 1, 2, ..., y, N. px B px s px Let represent the number of connected edge devices, network bandwidth, and relative distance to the x-th redundant edge server adjacent to the p-th edge server that went down, respectively. Let y represent the number of adjacent redundant edge servers. Let σ1, σ2, and σ3 represent the preset weights of the number of connected edge devices, network bandwidth, and relative distance, respectively. Let σ1 + σ2 + σ3 = 1.

[0026] The nearest redundant edge server corresponding to the maximum matching degree is recorded as the matching redundant edge server of the downed edge server. The matching redundant edge servers of each downed edge server are counted and their equipment groups are taken over.

[0027] Compared with existing technologies, the integrated management platform for smart parks based on edge computing described in this invention has the following beneficial effects: 1. By detecting communication quality assessment index information between distributed edge devices and edge servers within the park, this invention can determine whether the communication between edge devices and edge servers is abnormal, further obtain the improvement effect of communication quality between edge devices and edge servers with abnormal communication after changing the communication path, analyze the suspected causes of communication abnormalities, and thus be able to promptly detect communication failures between terminal devices and servers and take countermeasures, which is conducive to the reliability management of smart parks.

[0028] 2. This invention improves the robustness of smart park management by detecting whether the edge servers in the park are down, and by analyzing the load, network conditions and geographical location of the redundant edge servers near the downed edge server, matching redundant edge servers for the downed edge server and taking over its equipment group. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a system module connection diagram of the present invention.

[0031] Figure 2 This is a diagram of the hierarchical edge computing architecture of the present invention.

[0032] Figure 3 This is a flowchart for troubleshooting suspected causes of communication anomalies according to the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Please see Figure 1 and Figure 2 As shown, the present invention provides an integrated smart park management platform based on edge computing, including a device-server communication quality detection module, a device-server communication anomaly identification module, a communication anomaly suspected cause analysis module, a server downtime detection module, a server collaborative disaster recovery module, and a database.

[0035] The device-server communication anomaly identification module is connected to the device-server communication quality detection module and the communication anomaly suspected cause analysis module, respectively. The server downtime detection module is connected to the communication anomaly suspected cause analysis module and the server collaborative disaster recovery module, respectively. The database is connected to the device-server communication quality detection module and the server downtime detection module, respectively.

[0036] The device-server communication quality detection module acquires communication performance index information between each edge device and the edge server within the park, and evaluates the communication quality between each edge device and the edge server.

[0037] Furthermore, the specific working process of the device-server communication quality detection module includes: setting the duration of the monitoring period, obtaining the one-way latency, jitter, bandwidth, packet loss rate, error rate, response time, and throughput of each communication between the distributed edge devices in the park and their connected edge servers within the monitoring period, and calculating the average values ​​for each, to obtain the latency, jitter, bandwidth, packet loss rate, error rate, response time, and throughput of the communication between the distributed edge devices in the park and the edge servers.

[0038] It should be noted that distributed edge devices within the park refer to terminal devices or data acquisition devices. Edge devices include, but are not limited to, sensors, cameras, and environmental monitoring equipment.

[0039] It should be noted that edge servers within the park are edge nodes. These edge servers not only have the function of storing and processing data, but can also communicate with neighboring edge servers.

[0040] As a preferred approach, one-way latency refers to the one-way time it takes for data to travel from the edge device to the edge server.

[0041] As a preferred approach, jitter refers to the degree of fluctuation in communication latency between edge devices and edge servers.

[0042] As a preferred option, bandwidth refers to the maximum data transmission rate of the network link between the edge device and the edge server.

[0043] As a preferred approach, packet loss rate refers to the percentage of data packets lost during data transmission between edge devices and edge servers.

[0044] As a preferred approach, the error rate refers to the proportion of data packet errors caused by noise and interference during data transmission between edge devices and edge servers.

[0045] As a preferred approach, response time refers to the time elapsed from the time a request is sent to the time a response is received during communication between an edge device and an edge server.

[0046] As a preferred approach, throughput refers to the actual effective data transmission rate during data transmission between edge devices and edge servers.

[0047] It should be noted that the one-way latency, jitter, bandwidth, packet loss rate, error rate, response time, and throughput of each communication between the distributed edge devices in the park and their connected edge servers can all be obtained through corresponding detection tools, which are existing and relatively mature technologies, and will not be elaborated here.

[0048] Furthermore, the specific working process of the device-server communication quality detection module also includes: extracting network performance evaluation indicators of edge device-edge server communication stored in the database, and obtaining thresholds for latency, jitter, bandwidth, packet loss rate, and error rate of communication between edge device and edge server.

[0049] As a preferred approach, the network performance evaluation metrics for communication between edge devices and edge servers stored in the database are set according to actual application requirements.

[0050] Based on the latency, jitter, bandwidth, packet loss rate, error rate, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a first evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating this first evaluation coefficient is as follows: in This represents the first evaluation coefficient for the communication quality between the i-th edge device and the edge server, where i = 1, 2, ..., n, and D. i J i B i L i E i Let represent the latency, jitter, bandwidth, packet loss rate, and error rate of the communication between the i-th edge device and the edge server, respectively. Let D′, J′, B′, L′, and E′ represent the thresholds for latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1, ε2, ε3, ε4, and ε5 represent the weights of latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1+ε2+ε3+ε4+ε5=1.

[0051] It should be noted that in the first evaluation coefficient of communication quality between edge devices and edge servers, communication latency, jitter, bandwidth, packet loss rate, and error rate are the main factors. First, the communication latency, jitter, bandwidth, packet loss rate, and error rate of different units are normalized, and the normalized results are used for calculation. The pre-set thresholds for latency, jitter, bandwidth, packet loss rate, and error rate are all manually set at the initial stage of system operation. In this embodiment, D′, J′, B′, L′, and E′ are set to 200ms, 50ms, 10Mbps, 5%, and 1%, respectively.

[0052] By collecting multiple sets of specific numerical data on latency, jitter, bandwidth, packet loss rate, and error rate between distributed edge devices and edge servers within the park, and performing simulation calculations using the first evaluation coefficient formula for communication quality, a feasible simulation process is presented. This simulation assumes communication between a single edge device and an edge server, with weights for latency, jitter, bandwidth, packet loss rate, and error rate of 0.2, 0.2, 0.2, 0.2, and 0.2, respectively. The simulation results are shown in Table 1, which lists some representative data.

[0053] Table 1. Delay, jitter, bandwidth, packet loss rate, error rate, and calculation results of partial data collection.

[0054]

[0055] Based on the above simulation calculations, the smaller the latency, the smaller the jitter, the larger the bandwidth, the smaller the packet loss rate, and the smaller the error rate of communication between the edge device and the edge server, the larger the first evaluation coefficient of communication quality between the edge device and the edge server.

[0056] Furthermore, the specific working process of the device-server communication quality detection module also includes: extracting the application layer performance evaluation indicators of edge device-edge server communication stored in the database, and obtaining the thresholds for response time and throughput of communication between edge device and edge server.

[0057] As a preferred approach, the application layer performance evaluation metrics for communication between edge devices and edge servers stored in the database are set according to actual application requirements.

[0058] Based on the response time, throughput, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a second evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating the second evaluation coefficient for communication quality is as follows: in φ represents the second evaluation coefficient for the communication quality between the i-th edge device and the edge server. i , Let φ′ and t represent the throughput and response time of the communication between the i-th edge device and the edge server, respectively. r ' represents the threshold for throughput and response time, respectively, and α1 and α2 represent the preset weights for throughput and response time, respectively, with α1 + α2 = 1.

[0059] Furthermore, the specific working process of the device-server communication quality detection module also includes: analyzing the communication quality between the distributed edge devices and edge servers within the park based on the first evaluation coefficient and the second evaluation coefficient of communication quality between the distributed edge devices and edge servers within the park, wherein the calculation formula for communication quality is as follows: Where γ i This represents the communication quality between the i-th edge device and the edge server. The preset communication quality correction factor is represented by e, which represents the natural constant. η1 and η2 represent the weights of the preset first and second evaluation coefficients of communication quality, respectively, and η1 + η2 = 1.

[0060] It should be noted that the components of the above communication quality calculation formula are explained as follows: The preset communication quality correction factor is used to adjust the overall proportion of communication quality, such as globally amplifying or reducing the calculation results. η1 and η2 represent the weights of the preset first and second evaluation coefficients of communication quality, respectively. η1 + η2 = 1, used to quantify the relative importance of the two evaluation indicators to communication quality and to constrain the weights to avoid model imbalance caused by excessively large or small weights. An exponential function with base e is chosen to convert the linear combination of evaluation coefficients into a non-linear output, which better reflects the actual changes in communication quality with the indicators in real-world scenarios. Furthermore, the exponential function output is always positive, consistent with the physical meaning of communication quality as a non-negative value. The exponential function is also sensitive to input changes, amplifying the impact of key indicators on communication quality. Naturally, the exponential function is mathematically easy to differentiate and optimize, making it suitable for subsequent parameter adjustments or combination with other models.

[0061] The device-server communication anomaly identification module: Based on the evaluation results of the communication quality between each edge device and the edge server, it determines whether the communication between each edge device and the edge server is abnormal, and marks and issues warnings accordingly.

[0062] Furthermore, the specific working process of the device-server communication anomaly identification module is as follows: the communication quality between each distributed edge device and the edge server in the park is compared with a preset communication quality threshold. If the communication quality between an edge device and the edge server is less than the preset communication quality threshold, then the communication between the edge device and the edge server is abnormal, and the edge device is marked as a marked edge device. The marked edge devices are counted and an early warning is issued.

[0063] The communication anomaly suspected cause analysis module: detects the improvement effect of communication quality between edge devices and edge servers after changing the communication path, analyzes the suspected cause types of communication anomalies between edge devices and edge servers and provides feedback.

[0064] Further, see Figure 3 As shown, the specific working process of the communication anomaly suspected cause analysis module includes: changing the communication path between each marked edge device and the edge server, collecting the latency, jitter, bandwidth, packet loss rate, error rate, response time and throughput of the communication between each marked edge device and the edge server after changing the communication path, and evaluating the communication quality between each marked edge device and the edge server after changing the communication path.

[0065] As a preferred approach, the method for evaluating the communication quality between each marked edge device and the edge server after changing the communication path is based on the same principle as the method for evaluating the communication quality between each edge device and the edge server.

[0066] The communication quality between each marked edge device and the edge server before and after the communication path change is compared. The change in communication quality after the change is obtained, and the improvement effect of the communication quality between each marked edge device and the edge server after the communication path change is analyzed. The formula for calculating the improvement effect of communication quality is as follows: Where λ j Δγ represents the improvement in communication quality between the j-th marked edge device and the edge server after changing the communication path, where j = 1, 2, ..., m. j μ represents the change in communication quality between the j-th marked edge device and the edge server after the communication path is changed. γ This represents the influencing factor corresponding to the preset unit change in communication quality.

[0067] It should be noted that the components of the above formula for calculating the improvement in communication quality are explained as follows: μ γ This represents the influence factor corresponding to the preset unit change in communication quality, used to adjust Δγ. j The degree of contribution to the improvement effect, and through μ γ Δγ of different dimensions can be j Unifying the scale to a single level facilitates horizontal comparisons; using e⁻¹ as the base of the exponential function balances the numerical range, preventing results from being too large or too small, and controlling the rate of increase in the improvement effect. The use of an exponential function reflects the diminishing marginal returns of communication quality improvement, i.e., as Δγ increases... j or μ γ As the value increases, the rate of improvement accelerates, consistent with the characteristic that the benefits of improved communication quality may exhibit non-linear growth in real-world scenarios. Using a piecewise function clearly distinguishes between effective and ineffective scenarios, simplifying the calculation process.

[0068] As a preferred option, the communication path between edge devices and edge servers includes, but is not limited to, Wi-Fi, Bluetooth, Zigbee, cellular networks, and satellite communication.

[0069] It should be noted that the change in communication quality between the marked edge device and the edge server after changing the communication path is the difference between the communication quality between the marked edge device and the edge server after changing the communication path and the communication quality between the marked edge device and the edge server before changing the communication path.

[0070] Furthermore, the specific working process of the communication anomaly suspected cause analysis module also includes: comparing the improvement effect of communication quality between each marked edge device and the edge server after changing the communication path with a preset improvement effect threshold. If the improvement effect of communication quality between a marked edge device and the edge server after changing the communication path is greater than or equal to the preset improvement effect threshold, then the suspected cause type of communication anomaly between the marked edge device and the edge server is communication connection; otherwise, it is a hardware problem. The suspected cause types of communication anomalies between each marked edge device and the edge server are counted and fed back to the park management cloud center.

[0071] It should be noted that this invention determines whether the communication between the edge devices and edge servers is abnormal by detecting the communication quality assessment index information between the distributed edge devices and edge servers in the park. It further obtains the improvement effect of the communication quality between the edge devices and edge servers after changing the communication path, analyzes the suspected causes of communication abnormalities, and thus can promptly detect communication failures between terminal devices and servers and take corresponding measures, which is beneficial to the reliability management of smart parks.

[0072] The server downtime detection module detects whether each edge server is down and issues an early warning.

[0073] Furthermore, the specific working process of the server downtime detection module is as follows: real-time monitoring of the operating status of each edge server in the park, extracting the fault characterization library of edge server downtime stored in the database, if an edge server shows a fault characterization of edge server downtime, then the edge server is downtime, and the downtime edge servers are counted and an early warning is issued.

[0074] As a preferred option, the fault characterization database of edge server downtime stored in the database is obtained by summarizing and compiling historical experience.

[0075] As a preferred solution, edge server downtime may manifest in different ways at multiple levels, such as network connectivity (Traceroute interruption, network port unresponsiveness, etc.), service response (web pages inaccessible, application connection failures, API call failures, etc.), resource usage (CPU and memory anomalies, disk I / O stagnation, etc.), system and application logs (numerous error logs, interrupted logging, etc.), and physical devices (indicator lights malfunctions, fan malfunctions, etc.).

[0076] The server collaborative disaster recovery module analyzes the load, network conditions, and geographical location of each failed edge server and its adjacent redundant edge servers to identify and take over their equipment groups.

[0077] Furthermore, the specific working process of the server collaborative disaster recovery module is as follows: obtain the redundant edge servers within a set range around each edge server that has crashed in the park, and record them as the redundant edge servers adjacent to each edge server that has crashed.

[0078] Monitor the load, network conditions, and geographical location of each downed edge server and its neighboring redundant edge servers. Obtain the number of edge devices connected to each downed edge server, network bandwidth, and relative distance to each neighboring redundant edge server. Analyze the matching degree between each downed edge server and its neighboring redundant edge servers. The matching degree is calculated using the following formula: in This represents the index of the x-th redundant edge server adjacent to the p-th edge server that went down, where p = 1, 2, ..., q, x = 1, 2, ..., y, N. px B px s px Let represent the number of connected edge devices, network bandwidth, and relative distance to the x-th redundant edge server adjacent to the p-th edge server that went down, respectively. Let y represent the number of adjacent redundant edge servers. Let σ1, σ2, and σ3 represent the preset weights of the number of connected edge devices, network bandwidth, and relative distance, respectively. Let σ1 + σ2 + σ3 = 1.

[0079] It should be noted that using the above function to calculate the matching degree of neighboring redundant edge servers of the downtime edge server can standardize various indicators and eliminate differences in dimensions; the linear weighted sum has low computational complexity, is suitable for real-time load balancing scenarios, and is simple and efficient; the weights σ1, σ2, and σ3 intuitively reflect the importance of each indicator, making it easy to adjust the strategy; adding a negative sign before negative indicators and a positive sign before positive indicators can intuitively reflect the relationship between the calculation results and the changes of positive and negative indicators.

[0080] The nearest redundant edge server corresponding to the maximum matching degree is recorded as the matching redundant edge server of the downed edge server. The matching redundant edge servers of each downed edge server are counted and their equipment groups are taken over.

[0081] As a preferred option, the relative distance between adjacent redundant edge servers refers to the distance between adjacent redundant edge servers and the edge servers that have failed.

[0082] The database stores network performance evaluation metrics and application layer performance evaluation metrics for communication between edge devices and edge servers, and also stores a fault characterization library for edge server downtime.

[0083] It should be noted that this invention improves the robustness of smart park management by detecting whether the edge servers in the park are down, and by analyzing the load, network conditions and geographical location of the redundant edge servers near the downed edge server, and taking over its equipment group.

[0084] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications or additions should fall within the protection scope of the present invention.

Claims

1. A smart park integrated management platform based on edge computing, characterized in that, include: Device-Server Communication Quality Detection Module: Acquires communication performance index information between each edge device and edge server within the park, and evaluates the communication quality between each edge device and edge server; Device-to-server communication anomaly identification module: Based on the evaluation results of the communication quality between each edge device and the edge server, it determines whether the communication between each edge device and the edge server is abnormal, and marks and issues warnings accordingly; Communication Anomaly Suspected Cause Analysis Module: Detects the improvement effect of communication quality between edge devices and edge servers after changing the communication path, analyzes the suspected causes of communication anomalies between edge devices and edge servers, and provides feedback. Server downtime detection module: Detects whether each edge server is downtime and issues an alert; Server collaborative disaster recovery module: Based on the load, network conditions and geographical location of each failed edge server and its adjacent redundant edge servers, analyze the load, network conditions and geographical location of each failed edge server and take over its equipment group. Database: Stores network performance evaluation metrics and application layer performance evaluation metrics for communication between edge devices and edge servers, and stores a fault characterization library for edge server downtime.

2. The integrated smart park management platform based on edge computing according to claim 1, characterized in that: The specific working process of the device-server communication quality detection module includes: Set the duration of the monitoring period, obtain the one-way latency, jitter, bandwidth, packet loss rate, error rate, response time, and throughput of each communication between the distributed edge devices in the park and their connected edge servers within the monitoring period, and calculate the average values ​​for each communication to obtain the latency, jitter, bandwidth, packet loss rate, error rate, response time, and throughput of the communication between the distributed edge devices in the park and the edge servers.

3. The integrated smart park management platform based on edge computing according to claim 1, characterized in that: The specific working process of the device-server communication quality detection module also includes: Extract the network performance evaluation metrics for communication between edge devices and edge servers stored in the database to obtain the thresholds for latency, jitter, bandwidth, packet loss rate, and error rate of communication between edge devices and edge servers; Based on the latency, jitter, bandwidth, packet loss rate, error rate, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a first evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating this first evaluation coefficient is as follows: in This represents the first evaluation coefficient for the communication quality between the i-th edge device and the edge server, where i = 1, 2, ..., n, and D. i J i B i L i E i Let represent the latency, jitter, bandwidth, packet loss rate, and error rate of the communication between the i-th edge device and the edge server, respectively. Let D′, J′, B′, L′, and E′ represent the thresholds for latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1, ε2, ε3, ε4, and ε5 represent the weights of latency, jitter, bandwidth, packet loss rate, and error rate, respectively. Let ε1+ε2+ε3+ε4+ε5=1.

4. The integrated smart park management platform based on edge computing according to claim 3, characterized in that: The specific working process of the device-server communication quality detection module also includes: Extract the application layer performance evaluation metrics of communication between edge devices and edge servers stored in the database to obtain the thresholds for response time and throughput of communication between edge devices and edge servers; Based on the response time, throughput, and corresponding thresholds of communication between distributed edge devices and edge servers within the park, a second evaluation coefficient for communication quality between each edge device and edge server is analyzed. The formula for calculating the second evaluation coefficient for communication quality is as follows: in φ represents the second evaluation coefficient for the communication quality between the i-th edge device and the edge server. i , Let φ′ and t represent the throughput and response time of the communication between the i-th edge device and the edge server, respectively. r ' represents the threshold for throughput and response time, respectively, and α1 and α2 represent the preset weights for throughput and response time, respectively, with α1 + α2 = 1.

5. The integrated smart park management platform based on edge computing according to claim 4, characterized in that: The specific working process of the device-server communication quality detection module also includes: Based on the first and second evaluation coefficients of communication quality between distributed edge devices and edge servers within the park, the communication quality between these devices and servers is analyzed. The formula for calculating communication quality is as follows: Where γ i This represents the communication quality between the i-th edge device and the edge server. The preset communication quality correction factor is represented by e, which represents the natural constant. η1 and η2 represent the weights of the preset first and second evaluation coefficients of communication quality, respectively, and η1 + η2 = 1.

6. The integrated smart park management platform based on edge computing according to claim 1, characterized in that: The specific working process of the device-server communication anomaly identification module is as follows: The communication quality between each distributed edge device and the edge server in the park is compared with a preset communication quality threshold. If the communication quality between an edge device and the edge server is less than the preset communication quality threshold, the communication between the edge device and the edge server is abnormal, and the edge device is marked as a marked edge device. The marked edge devices are counted and an early warning is issued.

7. The integrated smart park management platform based on edge computing according to claim 6, characterized in that: The specific working process of the suspected cause analysis module for communication anomalies includes: Change the communication path between each marked edge device and the edge server, and collect the latency, jitter, bandwidth, packet loss rate, error rate, response time and throughput of each marked edge device and the edge server after changing the communication path, and evaluate the communication quality between each marked edge device and the edge server after changing the communication path. The communication quality between each marked edge device and the edge server before and after the communication path change is compared. The change in communication quality after the change is obtained, and the improvement effect of the communication quality between each marked edge device and the edge server after the communication path change is analyzed. The formula for calculating the improvement effect of communication quality is as follows: Where λ j Δγ represents the improvement in communication quality between the j-th marked edge device and the edge server after changing the communication path, where j = 1, 2, ..., m. j μ represents the change in communication quality between the j-th marked edge device and the edge server after the communication path is changed. γ This represents the influencing factor corresponding to the preset unit change in communication quality.

8. The integrated smart park management platform based on edge computing according to claim 7, characterized in that: The specific working process of the suspected cause analysis module for communication anomalies also includes: The improvement effect of communication quality between each marked edge device and the edge server after changing the communication path is compared with the preset improvement effect threshold. If the improvement effect of communication quality between a marked edge device and the edge server after changing the communication path is greater than or equal to the preset improvement effect threshold, the suspected cause of communication abnormality between the marked edge device and the edge server is communication connection. Otherwise, it is a hardware problem. The suspected cause types of communication abnormality between each marked edge device and the edge server are counted and fed back to the park management cloud center.

9. The integrated smart park management platform based on edge computing according to claim 1, characterized in that: The specific working process of the server downtime detection module is as follows: The system monitors the operational status of each edge server within the park in real time, extracts fault characteristics of edge server downtime from the database, and determines whether an edge server is down if it exhibits a fault characteristic. The system then compiles statistics on all downed edge servers and issues an early warning.

10. The integrated smart park management platform based on edge computing according to claim 1, characterized in that: The specific working process of the server collaborative disaster recovery module is as follows: Obtain the redundant edge servers within a set range around each downed edge server in the park, and record them as the redundant edge servers adjacent to each downed edge server. Monitor the load, network conditions, and geographical location of each downed edge server and its neighboring redundant edge servers. Obtain the number of edge devices connected to each downed edge server, network bandwidth, and relative distance to each neighboring redundant edge server. Analyze the matching degree between each downed edge server and its neighboring redundant edge servers. The matching degree is calculated using the following formula: in This represents the index of the x-th redundant edge server adjacent to the p-th edge server that went down, where p = 1, 2, ..., q, x = 1, 2, ..., y, N. px B px s px Let represent the number of connected edge devices, network bandwidth, and relative distance to the x-th redundant edge server adjacent to the p-th edge server that went down, respectively; let y represent the number of adjacent redundant edge servers; and let σ1, σ2, and σ3 represent the preset weights of the number of connected edge devices, network bandwidth, and relative distance, respectively, with σ1 + σ2 + σ3 = 1. The nearest redundant edge server corresponding to the maximum matching degree is recorded as the matching redundant edge server of the downed edge server. The matching redundant edge servers of each downed edge server are counted and their equipment groups are taken over.

Images