Carrier pigeon foot ring data reporting scheduling method and system, electronic equipment and storage medium

Abstract

The invention provides a carrier pigeon foot ring data reporting scheduling method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining carrier pigeon foot ring data, carrying out the data preprocessing, generating a carrier pigeon density distribution thermodynamic diagram, and evaluating the motion state of a carrier pigeon according to the current position data and historical competition data of the carrier pigeon; according to the homing pigeon foot ring data and the base station historical data, predicting the real-time load of each base station through the constructed multi-modal space-time fusion prediction model; determining an access conflict occurrence probability according to the position and the motion state of the foot ring of the carrier pigeon and a base station load prediction result, carrying out space-time two-dimensional grouping on the foot ring of the carrier pigeon, and generating and issuing a foot ring network access and information reporting strategy according to grouping information; a base station load state and a carrier pigeon state in a competition process are monitored in real time, and a foot ring network access and information reporting strategy is dynamically adjusted based on a dynamic parameter optimization mechanism. According to the invention, terminal access congestion is avoided, the terminal access speed is effectively improved, and the terminal access power consumption is reduced.

Description

Technical Field

[0001] This application belongs to the field of Internet of Things (IoT) terminal networking technology, and in particular relates to a method, system, electronic device and storage medium for scheduling data reporting of pigeon leg rings. Background Technology

[0002] The pigeon ankle bracelet will connect to the 4G base station in real time for data reporting and configuration download. When a large number of ankle bracelets are gathered in a small area, the base station will use its own conflict and collision handling mechanism and use a random backoff method for access. The theoretical access limit of the 4G base station is 500 UE / cell.

[0003] The existing solution does not consider the geographical clustering characteristics of terminals, and has the following main drawbacks: (1) When the number of devices requesting access exceeds the concurrent carrying capacity of the base station, it will cause terminal access blockage. The base station will refuse the access request of the terminal that exceeds the carrying capacity, resulting in a decrease in the terminal access success rate. On the one hand, the signals interfere with each other and cause the loss of access information. On the other hand, the base station refuses the access request that exceeds the carrying capacity, resulting in a serious decrease in the terminal access success rate.

[0004] (2) Due to the base station access blockage caused by the start-up storm, the terminal will continuously retry base station access, continuously send access requests, and wait for access confirmation, causing the terminal to be unable to enter the low power mode and consuming a lot of power. Therefore, for low power IoT devices, the battery capacity of the pigeon real-time positioning ankle bracelet is limited, but in order to ensure that the positioning information can be continuously collected and transmitted during the competition, the battery life must be maintained for a certain period of time. The large amount of power consumed by access will reduce the battery life and affect the continuity of the competition trajectory.

[0005] (3) Due to access collision rollback, when there are many access terminals, the number of failures and rollback time will increase, which will cause most ankle bracelets to be unable to transmit data for a long time. Because the rollback time is too long, most ankle bracelets will be unable to transmit positioning data for a long time, resulting in poor data timeliness and loss of user experience. Summary of the Invention

[0006] In view of this, this application aims to propose a pigeon leg band data reporting and scheduling method to solve the problem of cellular network access conflict for high-density IoT devices in spatially confined areas.

[0007] To achieve the above objectives, the technical solution of this application is implemented as follows: Firstly, this application provides a method for scheduling pigeon leg band data reporting, including: Data on pigeon leg bands was acquired and preprocessed to generate a heat map of pigeon density distribution. Based on the current location data and historical race data of the pigeons, the movement status of the pigeons was evaluated using a Kalman filter algorithm. Based on the pigeon leg band data and base station historical data, a multimodal spatiotemporal fusion prediction model is constructed to predict the real-time load of each base station; wherein, the multimodal spatiotemporal fusion prediction model is trained using the historical race data; Based on the location and movement status of the pigeon's leg band and the base station load prediction results, the probability of access conflict is determined. The pigeon leg bands are divided into two-dimensional spatiotemporal groups. Based on the group information, the leg band network access and information reporting strategy is generated and distributed. By monitoring the base station load and pigeon status in real time during the competition, and based on a dynamic parameter optimization mechanism, the leg ring network access and information reporting strategies are dynamically adjusted in two phases: intensive and sparse.

[0008] Secondly, based on the same inventive concept, this application also provides a pigeon leg ring data reporting and scheduling system, comprising: The data acquisition and processing module is configured to acquire pigeon leg band data and perform data preprocessing to generate a pigeon density distribution heat map, and evaluate the pigeon's movement status based on the pigeon's current location data and historical race data, using a Kalman filter algorithm. The load prediction module is configured to predict the real-time load of each base station based on the pigeon leg band data and the base station historical data, using a constructed multimodal spatiotemporal fusion prediction model; wherein the multimodal spatiotemporal fusion prediction model is trained using the historical race data. The leg band grouping module is configured to determine the probability of access conflict based on the leg band position, movement status, and base station load prediction results of the homing pigeons, and to group the homing pigeons into two-dimensional spatiotemporal groups. Based on the grouping information, it generates and distributes leg band network access and information reporting strategies. The dynamic adjustment module is configured to dynamically adjust the leg ring network access and information reporting strategy by monitoring the base station load status and pigeon status in real time during the competition, and based on a dynamic parameter optimization mechanism, and divided into intensive and sparse phases.

[0009] Thirdly, based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the first aspect.

[0010] Fourthly, based on the same inventive concept, this application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in the first aspect.

[0011] Compared with existing technologies, the pigeon leg band data reporting and scheduling method, system, electronic device, and storage medium described in this application have the following advantages: The pigeon leg band data reporting and scheduling method described in this application greatly improves the pigeon access success rate, effectively reduces the average energy consumption of pigeon leg bands, and effectively shortens the timeliness of location data return, fully meeting the short-term high-density access requirements of pigeon races. Attached Figure Description

[0012] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a pigeon leg band data reporting and scheduling method according to an embodiment of this application; Figure 2 This is a schematic diagram of the grouping of pigeons after release, as described in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of a pigeon leg band data reporting and scheduling system as described in an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of the electronic device described in an embodiment of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0014] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application 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 after 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 only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0015] As described in the background technology above, the existing solution does not consider the geographical clustering characteristics of terminals. When 10,000 ankle bracelets simultaneously initiate access requests within a 500m radius, 99% of the devices will need to retry more than three times; moreover, the backoff waiting time will become increasingly longer. Such multiple invalid transmissions and receptions not only lead to unnecessary power consumption, but also seriously affect the timeliness of data during the start of the competition, resulting in a loss of user experience.

[0016] Therefore, this embodiment provides a pigeon leg ring data reporting and scheduling method based on the interaction between the pigeon leg ring, cloud platform and base station, which avoids terminal access blocking, effectively improves terminal access speed and reduces terminal access power consumption.

[0017] The embodiments of this application are described in detail below with reference to the accompanying drawings.

[0018] Please see Figure 1 As shown, this embodiment provides a method for scheduling pigeon leg band data reporting, which specifically includes the following steps: Step S101: Obtain pigeon leg band data and perform data preprocessing to generate a pigeon density distribution heat map. Based on the current location data and historical race data of the pigeons, the movement status of the pigeons is evaluated using the Kalman filter algorithm.

[0019] Specifically, in this embodiment, the pigeon leg band reports the collected pigeon location, altitude, and speed information to the cloud server. The cloud server performs spatiotemporal reference transformation on the raw data, converting latitude and longitude into UTM projection coordinates in batches, aligning them to the grid of the base station coverage area, counting the number of pigeons in each grid, and generating a pigeon density distribution heat map. The Kalman filter algorithm is used to predict the pigeon location and speed data to determine that the pigeons are in a "static clustered state," confirming that all pigeons are within the predetermined coverage area of ​​the base station and that there is no premature departure.

[0020] It should be noted that the Kalman filtering algorithm described in this embodiment is a conventional processing algorithm in the field, and this application has not improved it, so it will not be described in detail here.

[0021] Step S102: Based on the pigeon leg band data and base station historical data, predict the real-time load of each base station using the constructed multimodal spatiotemporal fusion prediction model; wherein, the multimodal spatiotemporal fusion prediction model is trained using historical race data.

[0022] Specifically, in this embodiment, the cloud server confirms the gathering location of the leg rings based on the pigeon collection information or predicts the location of the leg rings based on the location information transmitted back by the leg rings. It then confirms the area where the leg rings are located and the coverage of the base stations by using nearby base station IDs and location information. Finally, it predicts and calculates terminal congestion and preemption based on the capacity and load of individual base stations. This embodiment uses historical race data to train the network model to predict the real-time load rate of each base station.

[0023] In some implementations, the raw data reported by the pigeon leg band and the historical data from the base station are normalized. Spatial and temporal features are extracted using spatiotemporal convolutional gating units, and feature fusion is performed based on the spatial and temporal features to obtain a spatiotemporal feature tensor. Multi-scale load prediction is performed on the spatiotemporal feature tensor to output the base station load prediction value.

[0024] Specifically, this embodiment adopts the Multimodal Spatiotemporal Fusion Prediction Model (MSTF-Net), whose core innovation lies in constructing a three-layer progressive feature extraction architecture: original input layer (location, altitude, speed information reported by pigeon leg bands and historical data of base station load) → spatiotemporal feature extraction layer → dynamic weight fusion layer → prediction output layer.

[0025] First, the information reported by the pigeon leg bands is normalized using a multi-source data normalization engine, and historical base station load data is integrated.

[0026] Secondly, local spatiotemporal correlations are extracted through parallel spatiotemporal convolution, and a gating mechanism is used to dynamically weight feature importance. In this embodiment, a spatiotemporal convolutional gating unit (ST-GCU) is employed. Original input layer: Determine the heat map of pigeon density distribution based on the position, altitude, and speed information reported by the pigeon leg bands. In the formula, This indicates the spatial density division of homing pigeons within the base station coverage area. Indicates the width of the heatmap. Indicates the height of the heatmap. Representing a dimension as A real matrix of size H; Determine the historical load time series of base stations based on historical base station load data. In the formula, This represents historical data on base station load; it is a two-dimensional matrix. Indicates the length of the time series. The number of channels in a time series. Representing a dimension as A real matrix; The spatiotemporal feature extraction layer includes spatial feature extraction and temporal feature extraction, which includes: Spatial feature extraction: Spatial features are And perform spatial convolution operations; In the formula, It is calculated from the BeiDou positioning data of the pigeon's leg band. For example, the leg band reports its location (latitude and longitude) at a sampling rate of 10Hz. The base station coverage area (radius 500m) is divided into 50m×50m grids (a total of 10×10=100 grids). The number of pigeons in each grid is counted to generate the result. .

[0027] The convolution kernel parameters are calculated from the model training data (historical race data). For example, the convolution kernel size k = 3×3 (covering a range of 150m, matching the typical spatial scale of pigeon gathering), and the output feature dimension D = 64 (trained with data from 10 races of 1200 pigeons, achieving a prediction accuracy of 92% in this dimension), and the weights... Bias Obtained through training.

[0028] Temporal feature extraction: Temporal features are It is a bidirectional long short-term memory network; In the formula, This is a historical load time series of the base station, calculated from subsequent monitoring data (for example): The base station load rate over the past 24 hours is taken as (actual number of access terminals / 500), recorded every 15 minutes, for a total of N = 96 time steps. The load characteristic dimensions are C = 3 (number of access terminals, resource utilization, signal strength). Therefore... For example, if the load rate at 07:00 is 30% (150 terminals connected), then... = 0.3 (representing the 28th time step, the 1st feature).

[0029] For BiLSTM parameters, the model training data (example): the hidden layer dimension is set to 64 (consistent with the spatial feature dimension, which facilitates subsequent fusion), and the input weights, hidden layer weights, and other parameters are obtained from historical 24-hour load data training.

[0030] Dynamic weight fusion layer: The feature fusion formula is: ; In the formula, For learnable weight matrix, It is the Sigmoid activation function. For bias Prediction output layer: This embodiment employs multi-scale prediction (Multi-Scale Forecaster module), including: Short-term forecast branch: ; In the formula, For time-dimensional sub-features, the fusion feature F is extracted. For example, the time-dimensional data is separated from F, and the fusion feature of the most recent 30 minutes (12 time steps) is taken. This corresponds to the dense gathering stage before the homing pigeons are released.

[0031] The parameters are GRU parameters, historical short-term load data, for example: hidden layer dimension 64, weight parameters are trained from the load data of the past 10 races "30 minutes before release", to adapt to the short-term load fluctuations of dense influx of homing pigeons.

[0032] Long-term forecast branch: ; In the formula, For spatial dimension sub-features, feature F is fused for extraction. For example, spatial dimension data is separated from F, and raster features are taken one hour after release (24 time steps). This corresponds to the stage of the homing pigeons dispersing and flying.

[0033] for The parameters are historical long-term load data, such as the number of attention heads h=8. The weight parameters are trained from the load data of the past 10 races "1 hour after release", which is adapted to the gradual change in load caused by the spatial dispersion of homing pigeons.

[0034] Prediction Output Layer: The final prediction output formula is: ; In the formula, For dynamic weighting coefficients, This indicates a multi-head self-attention mechanism. This represents the final predicted value. This represents a short-term forecast. This represents a long-term forecast.

[0035] Step S103: Determine the probability of access conflict based on the location and movement status of the pigeon's leg band and the base station load prediction results. Group the pigeon leg bands into two-dimensional spatiotemporal groups and generate and distribute the leg band network access and information reporting strategies based on the group information.

[0036] Specifically, in this embodiment, the probability of access conflict is determined based on the location and movement status of the pigeon's leg ring and the base station load prediction results. The leg rings are then grouped in a spatiotemporal dimension: Spatial dimension: Based on the base station GIS data, the devices are divided into groups (each group ≤ 500), including a central dense area grid and its surroundings, and an outer scattered area grid; Temporal dimension: Gradient access delays are set (e.g., group 1: 08:00:00, group 2: 08:00:30...), and the first access time for each group is determined based on the release time and the set minimum time interval between components. The cloud server sends the group information to the pigeon leg ring via a communication protocol. After receiving the information, the pigeon leg ring stores the scheduling instructions and waits for a specified time to initiate an access request. In addition, the cloud server formulates an information reporting strategy and determines the reporting interval (when the predicted load rate is >80%, the startup time = base time + random(0, number of groups × interval)).

[0037] The initial grouping strategy is as follows: ; in, ; In the formula, This indicates the total number of racing pigeons, determined based on the number of race entries. This indicates the maximum access capacity of the base station, sourced from the 4G base station communication standard (3GPP TS 36.304). η represents the density of homing pigeons, and η represents the coverage area of ​​the base station. The final number of groups is: (Take the maximum value between the minimum number of groups and the optimal number of groups, taking into account both capacity and spatial distribution).

[0038] This embodiment also determines the access time for each group based on time-sharing access scheduling calculations, specifically including: (1) Calculation of initial access time ; In the formula, Indicates the first Group first access time, ; Indicates the base time; This indicates the minimum time interval for the component.

[0039] (2) Calculation of reconnection time (in response to access failure) ; In the formula, For the first Group 1 Reconnection time , To determine the maximum number of reconnections, we set it to 3. This is the reconnection interval within the same group.

[0040] Step S104: By monitoring the base station load status and pigeon status during the competition in real time, and based on the dynamic parameter optimization mechanism, the leg ring network access and information reporting strategy are dynamically adjusted in two phases: intensive phase and dispersed phase.

[0041] Specifically, in this embodiment, during the competition, the access strategy is dynamically adjusted by monitoring the base station load status and the pigeon status in real time. It is executed in two stages: the dense period (08:00-09:00) and the dispersed period (after 09:00). The dense period is during the 08:00-09:00 period when the pigeons have not flown out of the base station coverage area, and the dispersed period is during the 09:00 period when the pigeons have flown out of the base station coverage area.

[0042] The dynamic parameter optimization mechanism includes: In response to sudden changes in base station load (such as local clustering caused by temporary return of some homing pigeons), the system optimizes the grouping by detecting sudden load changes and adjusting grouping parameters to dynamically adjust the foot ring network access and information reporting strategies, thereby ensuring access stability. The load mutation detection is based on the absolute value of the difference between the actual load rate and the predicted load rate, and compared with the standard deviation of the prediction error. This embodiment is configured as follows: If... This is determined to be a load mutation, where... This represents the actual load rate. To predict load factor, The standard deviation of the prediction error; Grouping parameter adjustment involves optimizing the number of groups and the minimum time interval between components. During sudden load changes, it's necessary to increase the number of groups and shorten the inter-group interval to reduce the pressure on individual group accesses. This includes: Among them, the optimized number of groups The formula is: ; In the formula, This indicates the maximum number of pigeons in a single group after adjustment. Indicates the actual load rate; Optimized component minimum time interval The formula is: ; In the formula, the original total access window duration = the original number of groups × the original interval between groups, keeping the total window duration unchanged and shortening the interval between groups.

[0043] Compared with existing technologies, this application has significant advantages in access success rate, average access speed and energy consumption through the system strategy design of "load prediction + leg band grouping + time-sharing access". Taking the number of racing pigeons as 12,000 and the surrounding base station capacity as 4 base stations (total capacity 2,000) as an example, the improvement effect is shown in Table 1.

[0044] Table 1 Comparison of the improvement effects of existing technology and this application on various indicators Based on the above solution, this embodiment provides the following specific implementation: like Figure 2 As shown, assuming there are 1200 pigeons preparing for a race, and only one base station nearby, in a scenario with 1200 pigeons participating and only one 4G base station (maximum access capacity 500 UE / cell, coverage radius 500m), the execution steps can be divided into four core stages: data collection and state estimation, base station load prediction, grouping strategy generation, and real-time monitoring and strategy adjustment, as detailed below: Phase 1: Collection of pigeon leg band locations and estimation of movement status (pre-race preparation, 07:00-07:50) The core of this stage is to collect spatiotemporal data through pigeon leg bands, providing a foundation for subsequent load prediction and grouping. The execution steps are as follows: Step A1, Leg Ring Data Collection: The Beidou positioning leg rings worn by 1200 homing pigeons report raw data such as latitude, longitude, altitude, and speed in real time at a sampling rate of 10Hz, and transmit them to the cloud scheduling server through the CAT1 / NB-IoT module; Step A2, Data Preprocessing: The cloud server performs spatiotemporal benchmark transformation on the raw data, converts latitude and longitude into UTM projection coordinates in batches, aligns them to a 50m×50m grid in the base station coverage area (the base station coverage area is divided into 10×10=100 grids), counts the number of pigeons in each grid, and generates a heat map of pigeon density distribution (e.g., the central grid (5,5) has 150 pigeons, and the surrounding grids have an average of 50 pigeons). Step A3, motion state estimation: The position and speed data of the homing pigeons are predicted by Kalman filtering (based on probability distribution) to determine that the homing pigeons are in a "static gathering state" (pre-race pigeon gathering stage, speed <0.5m / s), and to confirm that all homing pigeons are within the 500m coverage area of ​​the base station and there is no premature departure.

[0045] Phase Two: Dynamic Prediction of Base Station Load (10 minutes before the match, 07:50-08:00) Based on the collected pigeon data and historical base station data (base station location data and base station load status), the real-time base station load is predicted using the MSTF-Net model. The steps are as follows: Step B1, Data Input Preparation: Spatial data: Heat map of pigeon density distribution generated in Phase 1 ( This reflects the spatial aggregation characteristics of homing pigeons; Time data: Retrieve the base station's historical load sequence for the past 24 hours (1 data point every 15 minutes, for a total of 96 time steps). Features include the number of access terminals, resource utilization, and signal strength. Pigeon collection information: It has been confirmed that all 1,200 homing pigeons are concentrated in the base station coverage area, and no external homing pigeons have entered.

[0046] Step B2, MSTF-Net model computation: Multi-source data normalization: Normalize the pigeon density values ​​(0-150 birds / grid) and base station load rate (0-100%) to the range of [0, 1] to eliminate dimensional differences; Spatiotemporal feature extraction: Spatial features (such as the dense concentration of pigeons in the central grid) and temporal features (such as the trend of gradually increasing pre-race load as pigeons gather) are extracted through a spatiotemporal convolutional gating unit (ST-GCU) and fused into a 64-dimensional spatiotemporal feature tensor. ); Multi-scale load forecasting: Short-term forecast (30 minutes after launch): The base station load rate at 08:00 is predicted to be 82% (corresponding to 410 terminal accesses) using the GRU branch. Long-term forecast (1 hour after launch): The load rate is predicted to be 35% (corresponding to 175 terminal accesses) through the Transformer branch. Dynamic weighted fusion: combining prediction errors (short-term errors) Long-term error ), take dynamic weights The final output is the predicted base station load value at 08:00. (72.5%, corresponding to 362 terminal accesses).

[0047] Step B3, Load Conflict Prediction: Based on the predicted load rate of 72.5% and the maximum capacity of the base station of 500, it is determined that if 1200 homing pigeons randomly access the network, approximately 838 homing pigeons (1200-362) will need to retry repeatedly, resulting in an access conflict rate of 69.8%, and a group scheduling strategy needs to be initiated.

[0048] Phase 3: Ankle band grouping strategy generation (5 minutes before the match, before 08:00) Based on the load prediction results, the leg bands of 1200 homing pigeons were grouped from a two-dimensional perspective of "space + time" to generate a leg band network access and information reporting strategy. The execution steps are as follows: Step C1, Spatial Grouping: Based on the base station GIS data (500m coverage area) and the pigeon density heat map, the 100 grids are divided into sub-regions according to "≤500 pigeons per group". Finally, combined with the load prediction results, the initial number of groups is determined to be G=6 groups (200 pigeons per group, far below the base station's capacity limit of 500). The specific grouping is as follows: Groups 1-2: Central dense area grid (5, 5) and surrounding area, 200 birds in each group (400 birds in total); Groups 3-6: Peripheral scattered area grid, 200 birds per group (800 birds in total).

[0049] Step C2, Time-based Scheduling: Set the gradient access delay, using 08:00 (launch time) as the base time. minimum interval between groups Determine the initial access time for each group: Group 1: 08:00:00; Group 2: 08:00:30; Group 3: 08:01:00; Group 4: 08:01:30; Group 5: 08:02:00; Group 6: 08:02:30.

[0050] Step C3, Policy Distribution: The cloud server distributes the group information (group number, access time, reconnection interval τ=5min) to 1200 pigeon leg rings via the MQTT protocol. After receiving the information, the leg rings store the scheduling instructions and wait for a specified time to initiate an access request.

[0051] Phase 4: Real-time load monitoring and strategy adjustment (starting at 08:00 during the competition) During the competition, the base station load and carrier pigeon status are monitored in real time, and the access strategy is dynamically adjusted, divided into two phases: a peak period (08:00-09:00) and a dispersed period (after 09:00). 1. Peak period (08:00-09:00, when homing pigeons have not yet left the base station coverage area) (1) Real-time load monitoring: The cloud server collects the base station resource utilization rate every 500ms and generates a load analysis report every 10 minutes (fixed interval T1): 08:00-08:05: Groups 1-3 completed their first access. The actual base station load rate was 70% (350 terminals), with an error of only 3.4% compared to the predicted value of 72.5%, and no sudden load changes. 08:05-08:10: Groups 4-6 completed their first access, with the base station load rate peaking at 75% (375 terminals), still below the base station load warning threshold of 80%.

[0052] (2) Reconnection scheduling: For the 20 pigeons (approximately 1.7%) that failed to connect on the first attempt, reconnection was scheduled at 5-minute intervals. Retry was performed, and all reconnections were completed between 08:05 and 08:15, with a success rate of 99.8%.

[0053] 2. Emergency adjustment for sudden load changes (08:10 emergency) If some carrier pigeons temporarily turn back at 08:10, causing the actual load rate of the base station to surge to 85% (425 terminals), the **load surge detection** will be triggered. ), immediately implement strategy adjustments: (1) Regrouping: The original 6 groups were split into 9 groups, the number of pigeons in each group was reduced from 200 to 133, and the interval between groups was shortened to 20 seconds; (2) Access time rearrangement: With 08:11 as the new base time, the 9 access times are 08:11:00, 08:11:20, 08:11:40...08:12:40, and the total access window remains 180s; (3) Policy issuance and execution: After the new policy is issued, the base station load rate gradually drops to 60% (300 terminals), eliminating the risk of access blockage.

[0054] 3. Dispersal period (after 09:00, the homing pigeons fly away from the base station coverage area) (1) Status judgment: When the base station signal strength reported by the pigeon leg band is <-90dBm (trigger condition), it is determined that the pigeon has flown out of the base station coverage area and entered the dispersed flight stage; (2) Strategy switching: The data reporting interval is switched from 10 minutes (T1) during the peak period to 2 minutes (T2), and the ankle bracelet only triggers data reporting when the signal is stable, thereby reducing energy consumption; (3) Continuous monitoring: The cloud server tracks the reported status of the remaining pigeons (about 150) based on the pigeon location data until the end of the race, ensuring that the trajectory data is completely transmitted back.

[0055] In this scenario, the execution of the method described in this embodiment increases the success rate of accessing 1200 homing pigeons from <50% in traditional random access to 99.8%, reduces the average energy consumption of leg bands by 40% (reducing invalid retries), and shortens the timeliness of location data return from "average 30-minute delay" to "≤2-minute delay", fully meeting the short-term high-density access requirements of homing pigeon races.

[0056] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] Based on the same inventive concept, and corresponding to the methods of any of the above embodiments, the embodiments of this application also provide a pigeon leg ring data reporting and scheduling system.

[0058] like Figure 3 As shown, the pigeon leg band data reporting and scheduling system includes: The data acquisition and processing module is configured to acquire pigeon leg band data and perform data preprocessing to generate a pigeon density distribution heat map, and evaluate the pigeon's movement status based on the pigeon's current location data and historical race data, using a Kalman filter algorithm. The load prediction module is configured to predict the real-time load of each base station based on pigeon leg band data and base station historical data through a constructed multimodal spatiotemporal fusion prediction model; wherein, the multimodal spatiotemporal fusion prediction model is trained through historical race data; The leg band grouping module is configured to determine the probability of access conflict based on the leg band position, movement status, and base station load prediction results of the homing pigeons, and to group the homing pigeons into two-dimensional spatiotemporal groups. Based on the grouping information, it generates and distributes leg band network access and information reporting strategies. The dynamic adjustment module is configured to dynamically adjust the leg ring network access and information reporting strategy by monitoring the base station load status and pigeon status in real time during the competition, and based on a dynamic parameter optimization mechanism, and divided into intensive and sparse phases.

[0059] For ease of description, the above system is described by dividing it into various modules based on their functions. Of course, in implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware.

[0060] The system described in the above embodiments is used to implement the corresponding method in any of the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0061] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the methods described in any of the above embodiments.

[0062] Figure 4 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0063] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0064] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0065] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0066] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0067] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0068] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0069] The electronic devices described above are used to implement the corresponding methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0070] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform the methods described in any of the above embodiments.

[0071] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0072] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to perform the methods described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0073] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0074] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0075] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for scheduling data reporting from pigeon leg bands, characterized in that, include: Data on pigeon leg bands was acquired and preprocessed to generate a heat map of pigeon density distribution. Based on the current location data and historical race data of the pigeons, the movement status of the pigeons was evaluated using a Kalman filter algorithm. Based on the pigeon leg band data and base station historical data, a multimodal spatiotemporal fusion prediction model is constructed to predict the real-time load of each base station; wherein, the multimodal spatiotemporal fusion prediction model is trained using the historical race data; Based on the location and movement status of the pigeon's leg band and the base station load prediction results, the probability of access conflict is determined. The pigeon leg bands are divided into two-dimensional spatiotemporal groups. Based on the group information, the leg band network access and information reporting strategies are generated and distributed. By monitoring the base station load and pigeon status in real time during the competition, and based on a dynamic parameter optimization mechanism, the leg ring network access and information reporting strategies are dynamically adjusted in two phases: intensive and sparse.

2. The method according to claim 1, characterized in that: The raw data reported by the pigeon leg bands is obtained. The latitude and longitude are converted into UTM projection coordinates in batches by performing spatiotemporal reference transformation on the raw data and aligned to the grid of the base station coverage area. The number of pigeons in each grid is counted to generate a heat map of pigeon density distribution. The raw data includes location information, altitude information, and speed information.

3. The method according to claim 2, characterized in that, The multimodal spatiotemporal fusion prediction model includes: The raw data reported by the pigeon leg bands and the historical data from the base station were normalized. Spatial and temporal features are extracted using a spatiotemporal convolutional gating unit, and feature fusion is performed based on the spatial and temporal features to obtain a spatiotemporal feature tensor. Multi-scale load prediction is performed on the spatiotemporal feature tensor to output the base station load prediction value.

4. The method according to claim 1, characterized in that: In terms of spatial dimension, the pigeon leg bands are grouped according to the initial grouping strategy, including the central dense area grid and the surrounding area, and the outer scattered area grid. In the time dimension, by setting gradient access latency, taking the launch time as the base time, and the set minimum time interval between components, the first access time of each group is determined; The group information is sent to the pigeon leg ring via a communication protocol. After receiving the information, the pigeon leg ring stores the scheduling instructions and waits for a specified time to initiate an access request.

5. The method according to claim 4, characterized in that, The initial grouping strategy is as follows: ； in, ; In the formula, Indicates the total number of homing pigeons; Indicates the maximum access capacity of the base station; η represents the density of homing pigeons, and η represents the coverage area of ​​the base station. The final number of groups is: 。 6. The method according to claim 5, characterized in that, The formula for the first access time of each group is: ； In the formula, Indicates the first Group first access time, ; Indicates the base time; This indicates the minimum time interval for the component.

7. The method according to claim 5, characterized in that, The dynamic parameter optimization mechanism includes: In response to sudden changes in base station load, packet optimization is performed through load change detection and packet parameter adjustment optimization strategies to dynamically adjust the foot ring network access and information reporting strategies. The load mutation detection is based on the absolute value of the difference between the actual load rate and the predicted load rate, and is compared with the standard deviation of the prediction error. The grouping parameter adjustment is achieved by optimizing the number of groups and the minimum time interval between components, including: Among them, the optimized number of groups The formula is: ； In the formula, This indicates the maximum number of pigeons in a single group after adjustment. Indicates the actual load rate; Optimized component minimum time interval The formula is: 。 8. A pigeon leg band data reporting and scheduling system, characterized in that, include: The data acquisition and processing module is configured to acquire pigeon leg band data and perform data preprocessing to generate a pigeon density distribution heat map, and evaluate the pigeon's movement status based on the pigeon's current location data and historical race data, using a Kalman filter algorithm. The load prediction module is configured to predict the real-time load of each base station based on the pigeon leg band data and the base station historical data, using a constructed multimodal spatiotemporal fusion prediction model; wherein the multimodal spatiotemporal fusion prediction model is trained using the historical race data. The leg band grouping module is configured to determine the probability of access conflict based on the leg band position, movement status, and base station load prediction results of the homing pigeons, and to group the homing pigeons into two-dimensional spatiotemporal groups. Based on the grouping information, it generates and distributes leg band network access and information reporting strategies. The dynamic adjustment module is configured to dynamically adjust the leg ring network access and information reporting strategy by monitoring the base station load status and pigeon status in real time during the competition, and based on a dynamic parameter optimization mechanism, and divided into intensive and sparse phases.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, in, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method described in any one of claims 1-7.

Images