Multi-dimensional flow limiting method, system and terminal applied to stadium management system

Abstract

The application relates to the technical field of stadium management systems, and particularly discloses a multi-dimensional flow limiting method, system and terminal applied to a stadium management system. The method comprises the following steps: obtaining a multi-dimensional client request and extracting carried information in the client request; based on the combination characteristics of the carried information, matching multi-dimensional flow limiting rules corresponding to the combination characteristics, wherein the multi-dimensional flow limiting rules comprise the association constraint of a space dimension threshold value and a time dimension threshold value; and based on the multi-dimensional flow limiting rules, performing dynamic flow limiting processing on the client request, wherein the dynamic flow limiting processing comprises automatically switching flow limiting parameter weights according to different event activity types. The system comprises a request obtaining module, a rule matching module, a flow limiting processing module and the like, and the terminal executes a method program through a processor. The multi-dimensional flow limiting method, system and terminal applied to the stadium management system can accurately adapt to a stadium dynamic scene, balance system load and user experience, and improve intelligent operation level.

Description

Technical Field

[0001] This invention relates to the field of stadium management system technology, and in particular to a multi-dimensional flow restriction method, system and terminal applied to stadium management systems. Background Technology

[0002] Existing stadium management systems generally suffer from simplistic and insufficiently adaptable flow control mechanisms when facing large-scale crowds and high-frequency requests. Traditional flow control methods often rely on fixed thresholds or single-dimensional parameters, such as the number of concurrent users or network bandwidth, which are insufficient to cope with the complex dynamic scenarios of stadiums. On the one hand, stadiums have clearly defined zones and high pedestrian flow, with significant differences in the carrying capacity of different areas, such as stands, passageways, and commercial areas. Fixed thresholds cannot match the real-time capacity changes of each area. On the other hand, the types of events are diverse, such as football matches, concerts, and small-scale events, with significantly different user behavior patterns. During large-scale events, users concentrate on accessing live streaming and interactive functions, while for small-scale events, information retrieval is the primary focus. Existing systems lack the ability to dynamically adapt to different event types.

[0003] Meanwhile, existing technologies have limited perception dimensions for requests, relying heavily on user-initiated request data and ignoring physical space conditions captured by IoT devices, such as gate congestion and seat occupancy changes, as well as potential user needs, such as service requests predicted based on historical behavior. This leads to delayed rate limiting responses. Furthermore, rate limiting strategies lack differentiation, applying a uniform management approach to core services like ticketing transactions and non-core services like advertising, high-value users, and temporary visitors. This can result in either excessive rate limiting impacting user experience or insufficient control leading to system overload and critical service interruptions. These issues are particularly pronounced during peak periods of large-scale events, severely hindering the intelligent operation level of stadium management systems. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-dimensional flow restriction method, system, and terminal for use in stadium management systems, in order to solve the problems of existing flow restriction mechanisms such as singleness, poor adaptability, and delayed response, to achieve dynamic and differentiated flow restriction, to ensure system stability and user experience, and to improve the level of intelligence in stadium management.

[0005] To achieve the above objectives, this invention provides a multi-dimensional flow control method applied in a stadium management system, comprising the following steps:

[0006] S1. Obtain multi-dimensional client requests and extract the information carried in the client requests;

[0007] S2. Based on the combined features carrying information, match the multi-dimensional flow limiting rules corresponding to the combined features, including the correlation constraints of spatial dimension thresholds and time dimension thresholds;

[0008] S3. Dynamically limit client requests based on multi-dimensional rate limiting rules. Dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types.

[0009] Preferably, S1 is as follows:

[0010] The system acquires multi-dimensional client requests for accessing the stadium management system. The acquisition methods include directly receiving requests initiated by the client through the request receiving module, generating requests triggered by changes in the status of IoT devices within the stadium, and predicting potential requests based on historical user behavior data.

[0011] The information carried in the client request can be extracted from any two or more of the following: IP address, URL path, token information, user-associated physical region label, and event identifier.

[0012] Preferably, in S1, the process of triggering the generation of a request through a change in the state of IoT devices within the stadium is as follows:

[0013] Sa1. Collect status data through IoT devices, including smart gates, seat sensors, and mobile signal base stations;

[0014] Sa2: Receive status data uploaded by IoT devices in real time;

[0015] Sa3. The preset triggering conditions are divided into three levels according to priority. The preset conditions include the number of consecutive passages of the smart gate exceeding the set value, the abnormal frequency of personnel changes detected by the seat sensor, and the sudden increase in the strength of the mobile signal in a specific area.

[0016] Sa4. When the status data meets the preset trigger conditions, automatically generate a client request associated with the status change;

[0017] The process of predicting and generating potential requests based on user historical behavior data is as follows:

[0018] Sb1. Construct a user behavior prediction model and train the model. The input parameters of the model include the user's historical access time period, access frequency, related event type and consumption record.

[0019] Sb2. Based on the user behavior prediction model, generate the types and probabilities of requests that users may initiate within a preset time window, and identify requests with probabilities exceeding a set threshold as potential requests.

[0020] Preferably, S2 includes the following steps:

[0021] S21. Pre-construct a dynamic mapping model carrying information combination characteristics and rate limiting rules, and construct a multi-parameter linkage computing framework by integrating physical area constraints, network resource status and user behavior characteristics;

[0022] S22. Based on the extracted carried information combined features, the corresponding multi-dimensional rate limiting rules are generated and obtained through real-time calculation of the dynamic mapping model, including basic rate limiting thresholds, dynamic adjustment coefficients and hierarchical rate limiting strategies.

[0023] Preferably, in S3, dynamic rate limiting is performed on client requests based on multi-dimensional rate limiting rules, specifically as follows:

[0024] Based on real-time crowd density data from the stadium, the weighted request volume of client requests with the same combination of characteristics within a preset time window is calculated, and it is determined whether the weighted request volume reaches the flow control threshold after dynamic adjustment.

[0025] If the rate limiting threshold is reached, subsequent client requests will be processed differently according to the tiered rate limiting strategy. The tiered rate limiting strategy includes two or more of the following: request priority sorting based on user membership level, request diversion and guidance based on associated physical regions, and request delay gradient settings combined with the event progress.

[0026] Preferably, in step S3, before calculating the weighted request volume of client requests with the same combination of characteristics within a preset time window, the method further includes:

[0027] The client request is marked with a spatiotemporal dimension, including the real-time crowd flow data of the stadium at the time the request was initiated and the code of the user's reserved seat area.

[0028] The sliding time window algorithm is used to incrementally count client requests with the same combination of characteristics and to establish an index linking request time and spatial location.

[0029] Preferably, the following steps are included after S3:

[0030] Step 1: Monitor the real-time operation status and physical space status of the stadium management system. The operation status includes system load, resource usage, and network transmission rate. The physical space status includes the real-time number of people in each area, the efficiency of entrance and exit passages, and the occupancy of emergency passages.

[0031] Step 2: Based on the preset multi-dimensional evaluation model, calculate the comprehensive load index according to the monitored operating status and physical space status;

[0032] Step 3: Dynamically adjust the threshold parameters and strategy weights of the multi-dimensional rate limiting rules based on the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

[0033] This invention provides a multi-dimensional flow control system applied in a stadium management system, including a request acquisition module, a rule matching module, and a flow control processing module. The request acquisition module is used to acquire client requests and extract the information carried in the client requests. The client requests are used to request access to the stadium management system.

[0034] The rule matching module is used to match multi-dimensional rate limiting rules corresponding to combined features based on the information carried. The multi-dimensional rate limiting rules include the association constraints of spatial dimension thresholds and temporal dimension thresholds.

[0035] The rate limiting module is used to dynamically limit client requests based on multi-dimensional rate limiting rules. Dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types.

[0036] Preferably, it also includes a spatiotemporal marking module, an index counting module, a polymorphic monitoring module, a load assessment module, and an intelligent adjustment module. The spatiotemporal marking module marks client requests in a spatiotemporal dimension, including real-time stadium crowd data at the time the request is initiated and the user's reserved seat area code. The index counting module is used to incrementally count client requests with the same combination of features based on a sliding time window algorithm and to establish an association index between request time and spatial location.

[0037] The multi-state monitoring module is used to monitor the operational status and physical space status of the stadium management system in real time. The operational status includes system load, resource usage, and network transmission rate, while the physical space status includes the real-time number of people in each area, entrance and exit passage efficiency, and emergency passage occupancy. The load assessment module is used to calculate the comprehensive load index based on the monitored operational status and physical space status using a preset multi-dimensional assessment model. The intelligent adjustment module is used to dynamically adjust the threshold parameters and strategy weights of the multi-dimensional flow restriction rules based on the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

[0038] The present invention also provides a terminal, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement a multi-dimensional flow restriction method applied in a stadium management system.

[0039] Therefore, the present invention employs the above-mentioned multi-dimensional flow restriction method, system, and terminal applied to stadium management systems, and the beneficial effects are as follows:

[0040] (1) This invention comprehensively covers various requests through active reception, IoT triggering and behavior prediction. It combines multi-dimensional features to construct a dynamic mapping model and generate rules with associated spatiotemporal constraints. It can accurately adapt to different events and regional load changes, break the limitations of fixed thresholds, and solve the problems of narrow perception and delayed response in traditional flow limiting.

[0041] (2) This invention adopts a hierarchical strategy to distinguish between business and user types, combines the event process and regional diversion to balance load and experience, and can achieve dynamic adjustment of rules through real-time monitoring and comprehensive evaluation. It has the ability to enhance cross-regional traffic diversion and cope with extreme scenarios, and improves the level of intelligent management of stadiums.

[0042] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0043] Figure 1 This is an overall flowchart of an embodiment of the multi-dimensional flow restriction method of the present invention applied to a stadium management system;

[0044] Figure 2 This is a structural diagram of an embodiment of the multi-dimensional flow restriction system of the present invention applied to a stadium management system;

[0045] Figure 3 This is a schematic diagram of the implementation process of the multi-dimensional flow restriction method, system and terminal embodiment 1 of the present invention applied to the stadium management system. Detailed Implementation

[0046] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0047] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0048] like Figure 1 As shown, a multi-dimensional flow control method applied in a stadium management system includes the following steps:

[0049] S1. Obtain multi-dimensional client requests and extract the information carried in the client requests, specifically:

[0050] The system acquires multi-dimensional client requests for accessing the stadium management system. The acquisition methods include directly receiving requests initiated by the client through the request receiving module, generating requests triggered by changes in the status of IoT devices within the stadium, and predicting potential requests based on historical user behavior data. The system extracts information carried in the client requests, including any two or more of the following: IP address, URL path, token information, user-associated physical area tag, and event identifier.

[0051] In this embodiment, the process of triggering the generation of a request through the state change of IoT devices within the stadium is as follows:

[0052] Sa1. Collect status data through IoT devices, including smart gates, seat sensors, and mobile signal base stations;

[0053] Throughout the stadium, three types of IoT devices are deployed in a distributed manner according to different functional areas: smart turnstiles, seat sensors, and mobile signal micro base stations. The smart turnstiles have built-in infrared beam sensors, RFID card readers, and status indicator lights. Thin-film pressure sensor arrays are embedded in the seats to detect pressure and simultaneously collect seat numbers and area codes. The mobile signal micro base stations collect data in real time on the number of MAC addresses, signal strength, and channel occupancy of active terminals within the coverage area.

[0054] Sa2: Receives status data uploaded by IoT devices in real time and performs local preprocessing through the edge computing gateway.

[0055] For smart gate data, false trigger signals are filtered out to generate access event records that include device ID, access direction, timestamp, and verification result;

[0056] For seat sensor data, a sliding mean filtering algorithm is used to process pressure value fluctuations. When the detected value jumps from less than 5kg to more than 20kg, it is marked as a sitting event; when it drops from more than 20kg to less than 5kg, it is marked as a leaving event.

[0057] For mobile signal data, invalid terminals with signal strength less than -90dBm are removed, the number of active terminals is counted according to a fixed window period, and the terminal density in the area is calculated.

[0058] Sa3. The preset triggering conditions are divided into three levels according to priority. The preset conditions include the number of consecutive passages of the smart gate exceeding the set value, the abnormal frequency of personnel changes detected by the seat sensor, and the sudden increase in the strength of the mobile signal in a specific area.

[0059] Among them, the first-level trigger belongs to the safety warning category. When the smart gate shows that a single device passes through 50 times in a row within 5 minutes and the direction is consistent, it is determined to be a one-way congestion. Or when the mobile base station detects that the terminal density in a specific area increases by 200% or more within 30 seconds and the average signal strength increases by 15dB or more, it is determined to be a sudden increase in people. The first-level condition is triggered.

[0060] The second-level trigger is a service adjustment type. When the seat sensor detects that the frequency of personnel changes in the same zone within 10 minutes, that is, the sum of the number of times people sit down and leave their seats is not less than 50% of the total number of seats in the zone, it is determined to be an abnormal flow in the area. Or when the smart gate shows that the cumulative passage difference of 3 adjacent gates within 5 minutes is not less than 30 people, it is determined to be an uneven channel load. The second-level condition is triggered.

[0061] Level 3 triggering is a system optimization type. It is triggered when a mobile base station detects that the terminal density is stable at 80% to 90% of the area capacity within three consecutive 1-minute windows, indicating high load operation, or when a seat sensor detects that there are no less than 5 departure events in a single VIP area within 10 minutes and no corresponding re-seating events occur, indicating that service demand may increase.

[0062] Sa4. When the status data meets the preset trigger conditions of the corresponding level, the system will automatically generate a client request associated with the status change, including the trigger source identifier, trigger type, associated parameters, timestamp, and processing priority. Among them, the first-level trigger request is directly routed to the system emergency processing module and simultaneously pushed to the security command terminal; the second-level trigger request is sent to the regional management subsystem to trigger the generation of dynamic guidance instructions; and the third-level trigger request is pushed to the resource scheduling module for the pre-allocation of service resources.

[0063] Specifically, the process of predicting and generating potential requests based on users' historical behavior data is as follows:

[0064] Sb1. Construct a user behavior prediction model and train it. The model input parameters include the user's historical access time period, access frequency, associated event type, and consumption record. The specific process is as follows:

[0065] First, the model input parameters are preprocessed, including parsing the timestamps of each user's visit to extract specific hour segments, weekday / weekend attributes, and whether it is a special holiday. Through statistical analysis, user access preferences at different times are determined, such as a user's higher access frequency between 7 PM and 10 PM on weekdays.

[0066] Calculate the total number of visits per user per unit of time, such as per day or per week. Combine different time periods to analyze whether the user's access frequency increases significantly during the peak of the event, and the distribution of the time interval between two adjacent visits, in order to determine the user's access patterns.

[0067] The system categorizes and labels the events that users have visited, first classifying them into broad categories such as football, basketball, and tennis, and then further subdividing them into specific leagues such as the English Premier League and the NBA, and event levels such as the World Cup and regular season.

[0068] By constructing a correlation matrix between users and event types, we can analyze users' attention levels and conversion patterns for different event types. We collect information such as users' spending amounts, spending times, and the corresponding event and service types to calculate average spending, spending frequency, and the proportion of different spending types in total spending. This allows us to uncover potential connections between spending behavior, browsing behavior, and event attention; for example, a user is more likely to make related purchases after watching a specific event.

[0069] Then, based on the preprocessed parameter features, models such as logistic regression, decision trees, and neural networks are selected for training. During training, historical data is divided into training and validation sets, and model parameters are continuously adjusted to ensure the model accurately fits user behavior patterns. For example, increasing the number of hidden layers in the neural network improves the model's predictive ability for complex user behaviors.

[0070] Sb2. Based on the user behavior prediction model, generate the types and probabilities of requests that users may initiate within a preset time window, and identify requests with probabilities exceeding a set threshold as potential requests. The specific process is as follows:

[0071] The size of the time window needs to be determined based on the actual business scenario. For businesses with high real-time requirements, such as predicting user requests during live sports events, the time window can be set to a few minutes or tens of minutes. For businesses with strong periodicity, such as predicting user requests related to monthly events, the time window can be set to a month. By setting the time window appropriately, it is ensured that the model can capture the behavioral characteristics of users within a specific time period.

[0072] Based on the input user's historical behavior data, the model predicts the types of requests a user might make within a preset time window and calculates the probability of each request type. For example, the model can predict that there is a 70% probability that a user will request to watch a live football match within the next 30 minutes, and a 30% probability that they will request to buy merchandise related to a particular team.

[0073] Threshold filtering: A reasonable probability threshold is set, and request types with a probability exceeding this threshold are considered potential requests. The threshold setting needs to comprehensively consider the accuracy requirements and coverage scope of the business. If higher accuracy is required, the threshold can be set higher. In this embodiment, the probability threshold is set to 80%; if it is desired to cover more potential users, the threshold can be lowered to 50%. When the threshold is set to 60%, the request to view a live broadcast of a football match in the above example will be identified as a potential request. The above steps can more accurately predict and generate potential requests based on users' historical behavior data, providing strong support for subsequent business decisions and user services.

[0074] S2. Based on the combined features carrying information, match the multi-dimensional rate limiting rules corresponding to the combined features, including the correlation constraints of spatial dimension thresholds and time dimension thresholds, including the following steps:

[0075] S21. A dynamic mapping model carrying information combination characteristics and rate limiting rules is pre-constructed, and a multi-parameter linkage computing framework is constructed by integrating physical area constraints, network resource status and user behavior characteristics.

[0076] The combined characteristics of the information carried include spatial characteristics, network characteristics, user characteristics, and temporal characteristics. Spatial characteristics include physical area capacity C and user density within the area. U 总 S represents the total number of users, and S represents the area of ​​the region; network characteristics include available network bandwidth B and bandwidth utilization. Network latency L, where B' is the actual bandwidth used; user characteristics include concurrent user count U, user activity A, and request frequency F; time characteristics include time period factors and periodic coefficients such as daily, weekly, and monthly.

[0077] The process of constructing a multi-parameter linkage computational framework is as follows:

[0078] First, establish the basic relationship between physical constraints and network resources, including:

[0079] Network support user limit Where bavg represents the average bandwidth consumption per user; and the maximum number of users supported by the space is U. Cmax =C×(1-γ), where γ is the spatial redundancy coefficient, ranging from 0.1 to 0.15; the upper limit of the foundation bearing capacity U base =min(U Bmax U Cmax ).

[0080] Then, real-time fluctuation factors are introduced to correct the basic correlation, including:

[0081] Bandwidth fluctuation correction B' = B × (1 - η × 3) is used to reflect the impact of bandwidth congestion on actual available resources; user behavior correction U' = U × (1 + A × 2) is used to reserve more redundancy for highly active user groups.

[0082] Time period correction In this embodiment, the peak period T = 0.8 and the trough period T = 1.2.

[0083] Finally, the weights of each feature are dynamically adjusted through training with real-time data.

[0084]

[0085] Among them, f i (t) is the normalized value of feature i at time t, δ i (t) represents the prediction error of feature i at time t, and λ is the learning rate, which ranges from 0.01 to 0.05.

[0086] S22. Based on the extracted carried information and combined features, the corresponding multi-dimensional rate limiting rules are generated and obtained through real-time calculation of the dynamic mapping model, including basic rate limiting thresholds, dynamic adjustment coefficients, and tiered rate limiting strategies. The specific process is as follows:

[0087] The basic threshold T is calculated by integrating historical data and real-time status. base :

[0088] T base =α·T hist +(1-α)·U dyn ×θ;

[0089] Among them, T hist The threshold is the best value for the same period in history; α is the historical weighting coefficient, with a value of 0.3 to 0.5; θ is the system health coefficient, with a value of 0.8 to 1.0, which is dynamically adjusted based on CPU or memory load.

[0090] The dynamic scaling adjustment coefficient K is based on the real-time load status, as shown in the following formula:

[0091]

[0092] The tiered rate limiting strategy is implemented from the dimensions of user level, business type, and time and space. For the user level dimension, differentiated thresholds are set according to user value.

[0093] T m =T base ×K×r m ;

[0094] Where, r mThe coefficients are: r1 = 1.5 for core users, r2 = 1.0 for ordinary users, and r3 = 0.6 for temporary users.

[0095] Traffic quotas are allocated based on business type importance. For example, core businesses such as transactions account for 40%, important businesses such as messaging account for 30%, browsing accounts for 20%, and low-priority businesses such as advertising account for 10%.

[0096] The system dynamically adjusts based on time and space dimensions, with the flow restriction coefficient increasing by 20% in hotspot areas and by 30% during peak hours. Temporary flow restriction plans are triggered for special scenarios.

[0097] S3. Dynamic rate limiting is applied to client requests based on multi-dimensional rate limiting rules. This dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types.

[0098] Based on real-time crowd density data from the stadium, the weighted request volume of client requests with the same combination of characteristics within a preset time window is calculated, and it is determined whether the weighted request volume reaches the flow control threshold after dynamic adjustment.

[0099] If the rate limiting threshold is reached, subsequent client requests will be processed differently according to the tiered rate limiting strategy. The tiered rate limiting strategy includes two or more of the following: request priority sorting based on user membership level, request diversion and guidance based on associated physical regions, and request delay gradient settings combined with the event progress.

[0100] Before calculating the weighted request volume of client requests with the same combination of characteristics within a preset time window, this embodiment further includes:

[0101] The client request is marked with a spatiotemporal dimension, including the real-time crowd flow data of the stadium at the time the request was initiated and the code of the user's reserved seat area.

[0102] The sliding time window algorithm is used to incrementally count client requests with the same combination of characteristics and to establish an index linking request time and spatial location.

[0103] S3 is followed by the following steps:

[0104] Step 1: Monitor the real-time operation status and physical space status of the stadium management system. The operation status includes system load, resource usage, and network transmission rate. The physical space status includes the real-time number of people in each area, the efficiency of entrance and exit passages, and the occupancy of emergency passages.

[0105] Step 2: Based on the preset multi-dimensional evaluation model, calculate the comprehensive load index according to the monitored operating status and physical space status.

[0106] Step 3: Dynamically adjust the threshold parameters and strategy weights of the multi-dimensional rate limiting rules based on the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

[0107] like Figure 2 As shown, a multi-dimensional flow control system applied in a stadium management system is characterized by comprising a request acquisition module, a rule matching module, and a flow control processing module. The request acquisition module is used to acquire client requests and extract the information carried in the client requests. The client requests are used to request access to the stadium management system.

[0108] The rule matching module is used to match multi-dimensional rate limiting rules corresponding to combined features based on the information carried. The multi-dimensional rate limiting rules include the association constraints of spatial dimension thresholds and time dimension thresholds.

[0109] The rate limiting module is used to dynamically limit client requests based on multi-dimensional rate limiting rules. Dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types.

[0110] The system also includes a spatiotemporal marking module, an index counting module, a polymorphic monitoring module, a load assessment module, and an intelligent adjustment module. The spatiotemporal marking module marks client requests with spatiotemporal dimensions, including real-time stadium traffic data at the time the request is initiated and the user's reserved seat area code. The index counting module is used to incrementally count client requests with the same combination of characteristics based on a sliding time window algorithm and establish an association index between request time and spatial location. The polymorphic monitoring module is used to monitor the real-time operating status and physical space status of the stadium management system. The operating status includes system load, resource usage, and network transmission rate, while the physical space status includes real-time number of people in each area, entrance and exit passage efficiency, and emergency passage occupancy.

[0111] The load assessment module is used to calculate the comprehensive load index based on the monitored operating status and physical space status according to the preset multi-dimensional assessment model; the intelligent adjustment module is used to dynamically adjust the threshold parameters and strategy weights of the multi-dimensional rate limiting rules according to the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

[0112] A terminal includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement a multi-dimensional flow limiting method applied to a stadium management system.

[0113] Example 1

[0114] like Figure 3As shown, a stadium is hosting an international friendly football match, with an estimated attendance of 35,000 spectators. The system needs to handle high-frequency requests such as ticket verification, food and beverage reservations, and live streaming interaction. The specific implementation process of the method and system of this invention is as follows:

[0115] First, multi-dimensional request acquisition is performed. The request acquisition module captures demand through three methods: directly receiving proactive requests from viewers' mobile devices, such as seat inquiries and product purchases; the smart gate detecting that more than 80 people pass through the east entrance within 10 minutes triggers a level-two congestion warning request; and based on user historical data, such as a user who purchased drinks during halftime in their last three matches, 1500 potential food and beverage reservation requests are generated. The extracted information includes the user's seating area, such as section A of the west stand, the event identifier (e.g., a friendly football match), and their membership level.

[0116] Then, dynamic rule matching is performed: the rule matching module calls the dynamic mapping model based on the combination characteristics of West Stand A, football match, and gold members: the spatial dimension threshold is set to ≤20 simultaneous online requests per 10 square meters in the area, the time dimension threshold is set to ≥30 seconds per user request interval within 20 minutes during halftime, and the association constraint is to automatically reduce the time threshold to 20 seconds when the area's population density exceeds 80%.

[0117] Dynamic traffic limiting is implemented. The traffic limiting module combines real-time crowd flow data. For example, if the current density of Zone A in the West Stand is 75%, it calculates that the number of similar requests in that area within 5 minutes has reached 180 and has not exceeded the threshold. When the halftime break begins, the number of requests surges to 320 within 10 minutes. The system automatically switches the parameter weight, adjusting the weight of the event progress from 0.3 to 0.6, triggering a tiered strategy: requests from Gold members are prioritized, requests from ordinary users are redirected to the North Stand catering area, and a 5-second delay is set for non-urgent queries.

[0118] Real-time adjustments and optimizations were made. The multi-state monitoring module detected a 15% drop in the network transmission rate of the West Grandstand and an increase in the comprehensive load index to 0.82, exceeding the safety threshold of 0.8. The intelligent adjustment module immediately activated cross-regional traffic diversion, directing 20% ​​of requests to the South Grandstand server node with lower load, ultimately ensuring that the system response latency was ≤1 second during the event and that the user request success rate reached 99.2%.

[0119] Therefore, this invention adopts the multi-dimensional flow restriction method, system and terminal applied to the stadium management system. By setting up three ways to cover requests and integrating multiple features to generate dynamic rules with associated spatiotemporal constraints, it balances experience and stability with a hierarchical strategy. It can monitor and adjust parameters in real time and activate cross-regional flow diversion, thus promoting the upgrade of stadium management to proactive and intelligent control.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-dimensional flow restriction method applied to stadium management systems, characterized in that, Includes the following steps: S1. Obtain multi-dimensional client requests and extract the information carried in the client requests; S2. Based on the combined features carrying information, match the multi-dimensional rate limiting rules corresponding to the combined features. The multi-dimensional rate limiting rules include the correlation constraints between spatial dimension thresholds and time dimension thresholds. S3. Dynamically limit client requests based on multi-dimensional rate limiting rules. Dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types. S1 specifically refers to: The system acquires multi-dimensional client requests for accessing the stadium management system. The acquisition methods include directly receiving requests initiated by the client through the request receiving module, generating requests triggered by changes in the status of IoT devices within the stadium, and predicting potential requests based on historical user behavior data. Extract information carried in client requests, including any two or more of the following: IP address, URL path, token information, user-associated physical region label, and event identifier; S2 includes the following steps: S21. Pre-construct a dynamic mapping model carrying information combination characteristics and rate limiting rules, and construct a multi-parameter linkage computing framework by integrating physical area constraints, network resource status and user behavior characteristics; The combined characteristics of the information carried include spatial characteristics, network characteristics, user characteristics, and temporal characteristics, among which spatial characteristics include physical area capacity. and user density within the region , Total number of users S The area is the region; network characteristics include available network bandwidth. Bandwidth utilization and network latency L ,in This refers to the actual bandwidth used; user characteristics include the number of concurrent users. U User activity A and request frequency F Temporal characteristics include time period factors and period coefficients. ; The process of constructing a multi-parameter linkage computational framework is as follows: First, establish the basic relationship between physical constraints and network resources, including: Network support user limit ,in Average bandwidth consumption per user; maximum number of users supported by space. ,in Spatial redundancy coefficient; upper limit of foundation bearing capacity. ; Then, real-time fluctuation factors are introduced to correct the basic correlation, including: Bandwidth fluctuation correction This is used to reflect the impact of bandwidth congestion on actual available resources; user behavior correction. This allows for more redundancy to be reserved for highly active user groups; Time period correction ; Finally, the weights of each feature are dynamically adjusted through training with real-time data. ； in, for t Time characteristics i The normalized value, for t Time characteristics i The prediction error The learning rate; S22. Based on the extracted carried information and combined features, the corresponding multi-dimensional rate limiting rules are generated and obtained through real-time calculation of the dynamic mapping model, including basic rate limiting thresholds, dynamic adjustment coefficients, and tiered rate limiting strategies. The specific process is as follows: The basic threshold is calculated by integrating historical data and real-time status. : ； in, This is the optimal threshold for the same period in history; Historical weighting coefficients; The system health coefficient is dynamically adjusted based on CPU or memory load. The scaling factor is dynamically adjusted based on real-time load conditions. K The formula is as follows: ； The tiered rate limiting strategy is implemented from the dimensions of user level, business type, and time and space. For the user level dimension, differentiated thresholds are set according to user value. ； in, For level coefficients, core users r 1=1.5, Regular User r 2=1.0, Temporary User r 3 = 0.6; Traffic quotas are allocated based on business type according to business importance, while the time and space dimension is dynamically adjusted based on region and time period.

2. The multi-dimensional flow restriction method applied to a stadium management system according to claim 1, characterized in that, In S1, the process of triggering request generation through state changes of IoT devices within the stadium is as follows: Sa1. Collect status data through IoT devices, including smart gates, seat sensors, and mobile signal base stations; Sa2: Receive status data uploaded by IoT devices in real time; Sa3. The preset triggering conditions are divided into three levels according to priority. The preset conditions include the number of consecutive passages of the smart gate exceeding the set value, the abnormal frequency of personnel changes detected by the seat sensor, and the sudden increase in the strength of the mobile signal in a specific area. Sa4. When the status data meets the preset trigger conditions, automatically generate a client request associated with the status change; The process of predicting and generating potential requests based on user historical behavior data is as follows: Sb1. Construct a user behavior prediction model and train the model. The input parameters of the model include the user's historical access time period, access frequency, related event type and consumption record. Sb2. Based on the user behavior prediction model, generate the types and probabilities of requests that users may initiate within a preset time window, and identify requests with probabilities exceeding a set threshold as potential requests.

3. The multi-dimensional flow restriction method applied to a stadium management system according to claim 2, characterized in that, In S3, dynamic rate limiting is applied to client requests based on multi-dimensional rate limiting rules, specifically: Based on real-time crowd density data from the stadium, the weighted request volume of client requests with the same combination of characteristics within a preset time window is calculated, and it is determined whether the weighted request volume reaches the flow control threshold after dynamic adjustment. If the rate limiting threshold is reached, subsequent client requests will be processed differently according to the tiered rate limiting strategy. The tiered rate limiting strategy includes two or more of the following: request priority sorting based on user membership level, request diversion and guidance based on associated physical regions, and request delay gradient settings combined with the event progress.

4. The multi-dimensional flow restriction method applied to a stadium management system according to claim 3, characterized in that, In S3, before calculating the weighted request volume of client requests with the same combination of characteristics within a preset time window, the following is also included: The client request is marked with a spatiotemporal dimension, including the real-time crowd flow data of the stadium at the time the request was initiated and the code of the user's reserved seat area. The sliding time window algorithm is used to incrementally count client requests with the same combination of characteristics and to establish an index linking request time and spatial location.

5. The multi-dimensional flow restriction method applied to a stadium management system according to claim 4, characterized in that, The following steps are included after S3: Step 1: Monitor the real-time operation status and physical space status of the stadium management system. The operation status includes system load, resource usage, and network transmission rate. The physical space status includes the real-time number of people in each area, the efficiency of entrance and exit passages, and the occupancy of emergency passages. Step 2: Based on the preset multi-dimensional evaluation model, calculate the comprehensive load index according to the monitored operating status and physical space status; Step 3: Dynamically adjust the threshold parameters and strategy weights of the multi-dimensional rate limiting rules based on the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

6. A multi-dimensional flow control system applied in a stadium management system, used to implement the method described in any one of claims 1-5, characterized in that, It includes a request acquisition module, a rule matching module, and a rate limiting module. The request acquisition module is used to acquire client requests and extract the information carried in the client requests. The client requests are used to request access to the stadium management system. The rule matching module is used to match multi-dimensional rate limiting rules corresponding to combined features based on the information carried. The multi-dimensional rate limiting rules include the association constraints of spatial dimension thresholds and temporal dimension thresholds. The rate limiting module is used to dynamically limit client requests based on multi-dimensional rate limiting rules. Dynamic rate limiting includes automatically switching the weight of rate limiting parameters according to different event types.

7. The multi-dimensional flow restriction system applied to a stadium management system according to claim 6, characterized in that, It also includes a spatiotemporal marking module, an index counting module, a polymorphic monitoring module, a load assessment module, and an intelligent adjustment module. The spatiotemporal marking module marks client requests with spatiotemporal dimensions, including real-time stadium crowd data at the time the request is initiated and the user's reserved seat area code. The index counting module is used to incrementally count client requests with the same combined characteristics based on a sliding time window algorithm and to establish an index linking request time and spatial location. The multi-state monitoring module is used to monitor the operational status and physical space status of the stadium management system in real time. The operational status includes system load, resource usage, and network transmission rate, while the physical space status includes the real-time number of people in each area, entrance and exit passage efficiency, and emergency passage occupancy. The load assessment module is used to calculate the comprehensive load index based on the monitored operational status and physical space status using a preset multi-dimensional assessment model. The intelligent adjustment module is used to dynamically adjust the threshold parameters and strategy weights of the multi-dimensional flow restriction rules based on the comprehensive load index. When the comprehensive load index exceeds the safety threshold, the cross-regional request diversion mechanism is automatically activated.

8. A terminal, characterized in that, It includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the multi-dimensional flow restriction method applied to a stadium management system as described in any one of claims 1-5.

Images