Multi-terminal-oriented advertisement putting cloud cooperation control system and method

Abstract

The application discloses a multi-terminal-oriented advertisement putting cloud end cooperative control system and method, relates to the technical field of cloud platforms, and collects resource state parameters and advertisement task information of multiple advertisement terminals to construct a basic data set; obtains the geographical position and coverage radius of the advertisement terminals to construct a weighted conflict relation graph; based on the weighted conflict relation graph and the basic data set, performs conflict detection on advertisement tasks between different terminals, identifies advertisement task combinations with conflicts, and obtains a potential conflict task set; based on the potential conflict task set, generates a plurality of task scheduling combination schemes; calculates a scheme comprehensive evaluation index, screens an optimal scheduling scheme, and adjusts according to the optimal scheduling scheme, effectively reduces advertisement conflicts, improves advertisement putting accuracy, and improves system resource utilization and load balancing capability.

Description

Technical Field

[0001] This invention relates to the field of cloud platform technology, specifically to a cloud-based collaborative control system and method for multi-terminal advertising delivery. Background Technology

[0002] With the development of digital advertising technology, in scenarios where outdoor display terminals and various smart display devices are widely deployed, advertising placement is gradually shifting from a single-terminal mode to multi-terminal collaborative placement. Existing advertising placement scheduling is usually based on preset rules or simple strategies. However, in multi-terminal scenarios, the coverage areas of different terminals may overlap. When multiple terminals play the same or similar advertisements at similar times, it is easy to generate redundant advertising exposure and reduce placement efficiency. Some solutions determine coverage relationships by dividing the distance between terminals or regions to avoid advertising conflicts, but these are mostly coarse-grained judgments that are difficult to accurately reflect the degree of overlap of coverage areas between terminals, resulting in inaccurate conflict identification. Existing scheduling methods mostly adopt single-objective or simple priority strategies, lacking comprehensive consideration of coverage effect, conflict degree and terminal resource utilization. Moreover, spatial information has not been effectively used in scheduling optimization. Therefore, how to accurately characterize the coverage relationship between terminals, and on this basis, realize conflict quantification and multi-factor comprehensive scheduling optimization to improve advertising effectiveness and resource utilization efficiency, has become a technical problem that needs to be solved. Summary of the Invention

[0003] The purpose of this invention is to provide a cloud-based collaborative control system and method for multi-terminal advertising delivery, in order to solve the problems raised in the prior art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: a cloud-based collaborative control method for multi-terminal advertising delivery, the method comprising: Step S100: Collect resource status parameters and advertising task information from multiple advertising terminals to construct a basic dataset; Step S200: Obtain the geographical location and coverage radius of the advertising terminal, divide the terminal coverage area into spatial grids, calculate the coverage overlap area ratio between any two terminals, and construct a weighted conflict relationship graph; Step S300: Based on the weighted conflict relationship graph and the basic data set, perform conflict detection on advertising tasks across different terminals, identify conflicting combinations of advertising tasks, and obtain a set of potential conflicting tasks. Step S400: Based on the set of potential conflicting tasks, generate several sets of task scheduling combination schemes; Step S500: Calculate the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme. Based on the total conflict score and resource utilization balance index, calculate the scheme comprehensive evaluation index, select the optimal scheduling scheme, and adjust according to the optimal scheduling scheme.

[0005] Furthermore, the construction of the basic data set in step S100 includes: In the advertising platform, the operating status information of each advertising terminal is obtained by collecting cloud data. The operating status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operating status information is preprocessed, including removing abnormal data and supplementing missing values. Obtain advertising task information for each advertising terminal, including advertising delivery time information, target coverage area, and resource requirement parameters, including bandwidth and storage requirements. Based on the preprocessed running status information and advertising task information, construct a basic data set. By combining the operational status and advertising task information of each advertising terminal, a basic data set is constructed to comprehensively understand the resource utilization, providing accurate input for the intelligent scheduling algorithm, thereby optimizing advertising delivery efficiency and resource allocation; The constructed dataset can be directly used for data mining, task prediction, and advertising strategy optimization. It is scalable and versatile, providing data support for intelligent management.

[0006] Furthermore, step S200, which involves constructing a weighted conflict graph, includes: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, the coverage radius corresponding to each advertising terminal is collected, and a two-dimensional coverage area corresponding to each advertising terminal is generated according to the coordinates and coverage radius. In the plane coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage areas of each advertising terminal are mapped to the set of grid units. Based on the spatial positional relationship between the grid units and the two-dimensional coverage areas, the set of advertising terminals corresponding to each grid unit is determined. Based on the set of advertising terminals in the grid unit, the number of advertising terminals covered by each grid unit is counted, and a correspondence table between grid units and advertising terminals is established. For any two advertising terminals, the number of grid units covered by both advertising terminals is multiplied by the area of ​​a single grid unit based on the correspondence table, which is used as the approximate intersection area of ​​the coverage areas of the two terminals. The number of grid units covered by any terminal is multiplied by the area of ​​a single grid unit, which is used as the approximate union area. The ratio of the approximate intersection area to the approximate union area is determined as the percentage of overlapping coverage area between the two advertising terminals. Using each advertising terminal as a graph node, when the percentage of overlapping area between any two advertising terminals is greater than a preset threshold, a connection edge is established between the corresponding nodes, and the percentage of overlapping area is used as the weight of the edge to construct a weighted conflict relationship graph. The entire coverage area is discretized by using a grid partitioning method, which transforms the continuous spatial problem into a discrete grid statistical problem. This avoids complex geometric intersection calculations, significantly reduces computational complexity while ensuring a certain level of accuracy, and improves processing efficiency. It is suitable for large-scale terminal scenarios. By setting the grid resolution, a flexible trade-off can be made between computational accuracy and computational overhead, enabling the system to be adjusted according to different application requirements, and exhibiting good adaptability and scalability. By statistically analyzing the intersection and union areas of grid cells and calculating the percentage of overlapping coverage areas, the coverage conflict between advertising terminals is transformed from a qualitative judgment to a quantitative assessment, thereby improving the accuracy and objectivity of conflict identification.

[0007] Furthermore, the set of potential conflicting tasks obtained in step S300 includes: Based on the weighted conflict relationship graph, obtain any two pairs of advertising terminals with connected edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set to extract the set of advertising tasks being executed and to be executed on each advertising terminal. Obtain the weight of the connection edge corresponding to any two connected advertising terminals. If the weight of the connection edge reaches the preset conflict threshold, it is determined that there is a potential spatial conflict relationship between the corresponding terminals. Then, the advertising tasks of the two terminals are combined in pairs to construct a set of candidate task pairs. In any pair of candidate tasks, the time interval for placing ads is obtained based on the ad placement time information in the ad task information. When the time intervals for placing ads of two ad tasks overlap, they are determined to be a pair of candidate tasks with time conflicts. In the time-conflict candidate task pairs, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the advertising task target area in the overlapping area. The spatial conflict degree index is calculated according to the following formula: ; Where A represents the spatial conflict index, B represents the overlapping area between advertising terminals, B1 and B2 represent the coverage areas of the first and second advertising terminals in the time conflict candidate task pair, respectively, and C1 and C2 represent the distribution ratio of the target areas of the advertising tasks being executed by the first and second advertising terminals in the time conflict candidate task pair within the overlapping area. Extract the resource requirement parameters of the advertising task, normalize the resource requirement parameters, and perform a weighted summation of the normalized resource requirement parameters to calculate the resource requirement index. Obtain the bandwidth occupancy rate and storage space utilization rate of the corresponding advertising terminal and perform normalization calculation. Calculate the resource conflict degree index as D=a1×(E1×b1+E2×b2)+a2×(F1×b1+F2×b2), where D represents the resource conflict degree index, E1 and E2 represent the normalized bandwidth occupancy rate of the first and second advertising terminals in the time conflict candidate task pair, F1 and F2 represent the normalized storage space utilization rate of the first and second advertising terminals in the time conflict candidate task pair, b1 and b2 represent the resource requirement index of the advertising task being executed by the first and second advertising terminals in the time conflict candidate task pair, and a1 and a2 represent the weights of the bandwidth occupancy rate and storage space utilization rate, respectively. The spatial conflict index is jointly determined by the degree of spatial overlap of the coverage area and the degree of distribution of advertising tasks in the overlapping area. The degree of spatial overlap of the coverage area is reflected by the ratio of the overlapping area to the total coverage area, which reflects the degree of spatial overlap between the two advertising terminals. The degree of distribution of advertising tasks in the overlapping area is reflected by the proportion of each task in the overlapping area, and the combined effect of the two is reflected by the product form. Thus, the conflict index is significantly improved only when both tasks are concentrated in the overlapping area. When calculating the degree of resource conflict, the resource utilization rate of the terminal is multiplied by the resource demand index of the corresponding advertising task to reflect the actual pressure that the task puts on the system under the current resource status. This can reflect the combined effect of the degree of resource utilization and the intensity of task demand. That is, only when the resource utilization rate and task demand are both high will a large resource conflict occur. The above results are calculated and summed for the two advertising terminals in the time conflict candidate task pair to represent the overall resource competition situation. Weights are set for bandwidth and storage resources to obtain the final resource conflict degree index. The spatial conflict index and resource demand index are normalized separately. The normalized spatial conflict index and resource demand index are then weighted and summed to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than a preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then aggregated to obtain a potential conflict task set. Based on the weighted conflict relationship graph, it is further refined to the advertising task level. By extracting the set of tasks being executed and to be executed by each terminal and constructing candidate task pairs, it achieves in-depth analysis from coarse-grained terminal conflict identification to fine-grained task conflict identification, significantly improving the accuracy of conflict detection. By setting a threshold for the percentage of overlapping coverage area, terminal pairs with potential spatial conflicts are prioritized for screening. Only tasks that meet the conditions are combined for analysis, which effectively reduces the computational load of irrelevant task pairs and improves the overall processing efficiency of the system. At the same time, time, space and resource dimensions are introduced to build a multi-dimensional conflict judgment mechanism, avoiding misjudgments caused by a single indicator and improving the comprehensiveness and accuracy of conflict identification. By calculating the overlapping area of ​​advertising terminals and the distribution ratio of the target area of ​​advertising tasks within the overlapping area, a spatial conflict index is constructed, transforming spatial conflict from empirical judgment into a quantifiable indicator, thereby improving the objectivity and interpretability of the assessment.

[0008] Furthermore, step S400 generates several sets of task scheduling combination schemes, including: A preset scheduling analysis time interval is set, and the scheduling analysis time interval is divided into several continuous time segments. Based on the original delivery time information of each advertising task in the potential conflict task set, each advertising task is matched to the corresponding time segment to obtain the distribution of advertising tasks in the time dimension. For advertising tasks within the set of potential conflicting tasks, a scheduling adjustment method is generated. This scheduling adjustment method includes adjustments to the delivery time, the delivery terminal, and the task structure. The process of adjusting the delivery time is as follows: for any advertising task, the delivery time is increased by an offset according to a preset time step, the offset delivery time interval is calculated, the delivery time interval is used as a candidate delivery time interval, and the time overlap detection is performed between each candidate delivery time interval and other advertising tasks in the potential conflict task set. When the time overlap between the adjusted delivery time interval and other advertising tasks is lower than a preset threshold, the corresponding candidate delivery time interval is determined as a feasible delivery time interval. The process of adjusting the advertising terminal is as follows: A set of candidate advertising terminals is obtained from other advertising terminals besides the original ones. For any candidate advertising terminal, the bandwidth occupancy rate, storage space utilization rate, and coverage area information within the advertising task's delivery time interval are obtained. The allocated bandwidth occupancy rate and storage space utilization rate within the advertising task's delivery time interval are subtracted from the total bandwidth occupancy rate and storage space utilization rate of the advertising terminal to calculate the remaining bandwidth occupancy rate and storage space utilization rate. The remaining bandwidth occupancy rate and storage space utilization rate are then normalized and weighted to obtain a remaining resource demand index. Based on the coverage area information, the coverage ratio of the advertising task's target area is calculated. When the remaining resource demand index is greater than the advertising task's resource demand index, and the coverage ratio is higher than a preset coverage threshold, the candidate advertising terminal is determined as a feasible alternative advertising terminal. The task structure adjustment method includes task splitting and task merging. The task splitting process includes dividing an advertising task into multiple sub-tasks and assigning them to different time segments and different advertising terminals. The task merging process includes merging multiple advertising tasks with similar placement times and target areas to form a new combined task. Based on the feasible delivery time interval, feasible alternative delivery terminals, and task structure adjustment scheme, calculate the delivery time interval of each advertising task in the adjusted delivery time interval, count the time overlap length between any two tasks, calculate the total time overlap length, divide the total time overlap length by the total delivery time, and calculate the adjusted time conflict degree. Based on the adjusted advertising terminal allocation, the coverage overlap area between each advertising terminal is calculated and summed to obtain the total coverage overlap area. Based on the task target coverage area of ​​the advertising task, the effective overlap area belonging to the task target coverage area in the coverage overlap area is calculated. The effective overlap area is divided by the total coverage overlap area to calculate the adjusted spatial overlap ratio. The resource usage of each advertising terminal within the corresponding time segment is statistically analyzed. The adjusted task resource requirements are then superimposed with the remaining resources of the terminal to calculate the resource load value of each terminal. Based on the adjustment conditions, the scheduling adjustment results are filtered to obtain several sets of task scheduling combination schemes. The adjustment conditions include: the degree of time conflict is lower than the preset time conflict threshold; the spatial overlap ratio is lower than the preset spatial conflict threshold; and the resource load of each advertising terminal does not exceed the carrying capacity threshold. By pre-setting and analyzing time intervals and dividing them into continuous time segments, advertising tasks are mapped to the time dimension for analysis. This enables the system to dynamically adjust task arrangements based on time distribution characteristics, achieving refined scheduling and adaptive optimization of potentially conflicting tasks. At the same time, three scheduling methods are introduced: adjustment of delivery time, adjustment of delivery terminal, and adjustment of task structure. This allows for collaborative optimization from multiple levels, including time, space, and task organization structure, avoiding the limitations of a single adjustment method and improving the flexibility and effectiveness of conflict resolution. By setting time steps to gradually shift the ad delivery time and combining time overlap detection to filter feasible delivery time intervals, the task can effectively avoid high conflict periods while ensuring delivery effectiveness, thereby improving the efficiency of time resource utilization. During the adjustment of the delivery terminal, the remaining bandwidth utilization and storage space utilization of the candidate terminals are calculated. Combined with normalization and weighted processing, terminals with sufficient resources and reasonable coverage are selected to achieve reasonable migration of tasks between different terminals, reduce the risk of local resource overload, and introduce coverage ratio constraints during the terminal replacement process to ensure that the adjusted advertising tasks can still meet the coverage requirements of the target area, thereby reducing conflicts while ensuring that the advertising effect is not significantly affected. After scheduling adjustments, the schemes are quantitatively evaluated from three dimensions: the degree of time conflict, the proportion of spatial overlap, and the resource load. This enables a comprehensive analysis of the scheduling results, improving the scientific and rational nature of scheme selection. By setting time conflict thresholds, spatial conflict thresholds, and terminal carrying capacity thresholds, scheduling schemes are screened to ensure that the final schemes meet the constraints in all dimensions, thereby avoiding resource overload and conflict accumulation problems and improving system operational stability.

[0009] Furthermore, step S500 generates several sets of task scheduling combination schemes, including: In any set of task scheduling combination schemes, the advertising task pairs are collected, and the adjusted advertising task delivery time interval, target area information and corresponding advertising terminal coverage area are collected. The spatial conflict degree index and resource conflict degree index between advertising tasks are calculated. Based on the spatial conflict degree index and resource conflict degree index, the comprehensive conflict score of a single advertising task pair is calculated. The comprehensive conflict scores of all advertising task pairs in the task scheduling combination scheme are accumulated to obtain the total conflict score of the task scheduling combination scheme. Based on the adjusted advertising task allocation, the bandwidth occupancy and storage space utilization of each advertising terminal in each time segment are extracted. The bandwidth occupancy and storage space utilization are normalized and weighted, and then summed to calculate the resource utilization rate. The resource utilization of each time segment is statistically analyzed, and the resource utilization balance index is calculated according to the following formula: ; Where K represents the resource utilization balance index, Ri represents the resource utilization rate of the i-th time segment, R1 represents the average resource utilization rate of all time segments, and n represents the number of time segments; The total conflict score and resource utilization balance index are normalized and calculated. The weights of the total conflict score and resource utilization balance index are preset. The normalized total conflict score and resource utilization balance index are weighted and summed to obtain the comprehensive evaluation index of the scheme. All task scheduling combination schemes are traversed, and the task scheduling combination scheme with the smallest comprehensive evaluation index is selected as the optimal scheduling scheme. According to the optimal scheduling scheme, an advertising task scheduling execution instruction is generated and sent to the corresponding advertising terminal for advertising placement control. By analyzing advertising task pairs in any task scheduling combination scheme, taking into account the delivery time interval, target area information and terminal coverage area, spatial conflict index and resource conflict index are constructed, and a comprehensive conflict score for a single task pair is further formed, so that scheduling conflict is transformed from qualitative judgment to quantifiable evaluation, thereby improving the accuracy of analysis. By summing the comprehensive conflict scores of all advertising task pairs in the scheme, the total conflict score of the task scheduling combination scheme is obtained. The scheduling conflict is uniformly measured from the overall level, avoiding the global conflict problem caused by local optimization and improving the global consistency of the scheduling scheme. Based on the bandwidth occupancy and storage space utilization of each advertising terminal in different time segments, the resource utilization rate is calculated, and a resource utilization balance index is further introduced to quantify the uniformity of resource distribution in the time dimension, thereby making the resource scheduling status more intuitive and controllable. By traversing all task scheduling combinations and selecting the optimal solution based on comprehensive evaluation indicators, the scheduling results are automatically filtered, reducing manual intervention and improving scheduling efficiency and accuracy. By achieving an optimal balance between conflict control and resource balancing, advertising tasks are more rationally allocated across terminals and time segments, thereby improving resource utilization efficiency and ensuring the continuity and stability of advertising delivery.

[0010] To better implement the above methods, a cloud-based collaborative control system for multi-terminal advertising delivery is also proposed. The system includes a basic data set module, a weighted conflict relationship graph module, a potential conflict task set module, a task scheduling combination scheme module, and a scheduling adjustment module. Basic Data Set Module: Collects resource status parameters and advertising task information from multiple advertising terminals to construct a basic data set; Weighted Conflict Relationship Graph Module: Obtain the geographical location and coverage radius of the advertising terminal, divide the terminal coverage area into spatial grids, calculate the coverage overlap area ratio between any two terminals, and construct a weighted conflict relationship graph; Potential conflict task set module: Based on the weighted conflict relationship graph and the basic data set, the module performs conflict detection on advertising tasks across different terminals, identifies conflicting combinations of advertising tasks, and obtains a potential conflict task set. Task scheduling combination scheme module: Based on the set of potentially conflicting tasks, generate several sets of task scheduling combination schemes; The scheduling and adjustment module calculates the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme. Based on the total conflict score and resource utilization balance index, it calculates the comprehensive evaluation index of the scheme, selects the optimal scheduling scheme, and adjusts according to the optimal scheduling scheme.

[0011] Furthermore, the basic data set module includes a runtime status information unit and an advertising task information unit: Operational Status Information Unit: In the advertising platform, operational status information of each advertising terminal is obtained by collecting cloud data. The operational status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operational status information is preprocessed, including removing abnormal data and supplementing missing values. Advertising task information unit: Acquires advertising task information from each advertising terminal. The advertising task information includes advertising placement time information, target coverage area, and resource requirement parameters. The resource requirement parameters include bandwidth requirements and storage requirements. Based on the preprocessed running status information and advertising task information, a basic data set is constructed.

[0012] Furthermore, the weighted conflict graph module includes an advertising terminal collection unit and a unit for constructing the weighted conflict graph: Advertising terminal set unit: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, collect the coverage radius corresponding to each advertising terminal, and generate the two-dimensional coverage area corresponding to each advertising terminal according to the coordinates and coverage radius. Under the planar coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage area of ​​each advertising terminal is mapped to the set of grid units. Based on the spatial positional relationship between the grid unit and the two-dimensional coverage area, the advertising terminal set corresponding to each grid unit is determined. Constructing a weighted conflict graph unit: Based on a set of advertising terminals in grid units, count the number of advertising terminals covered by each grid unit and establish a correspondence table between grid units and advertising terminals. Traverse any two advertising terminals, and based on the correspondence table, determine the number of grid units simultaneously covered by both advertising terminals multiplied by the area of ​​a single grid unit as the approximate intersection area of ​​the coverage areas of the two terminals. Count the number of grid units covered by any terminal multiplied by the area of ​​a single grid unit as the approximate union area. Determine the ratio of the approximate intersection area to the approximate union area as the coverage overlap ratio between the two advertising terminals. Using each advertising terminal as a graph node, when the coverage overlap ratio between any two advertising terminals is greater than a preset threshold, establish a connection edge between the corresponding nodes and use the coverage overlap ratio as the edge weight to construct a weighted conflict graph.

[0013] Furthermore, the potential conflict task set module includes a candidate task pair set unit and a potential conflict task set generation unit: Candidate task pair set unit: Based on the weighted conflict relationship graph, obtain any two advertising terminal pairs with connecting edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set, extract the set of advertising tasks being executed and to be executed on each advertising terminal, obtain the weight of the connecting edge corresponding to any two advertising terminals with connecting relationship, if the weight of the connecting edge reaches the preset conflict threshold, it is determined that the corresponding terminals have a potential spatial conflict relationship, and the advertising tasks of the two terminals are combined in pairs to construct a candidate task pair set; The process involves generating a potential conflict task set unit: In any pair of candidate tasks, based on the advertising delivery time information in the advertising task information, the delivery time interval is obtained. When the delivery time intervals of two advertising tasks overlap, they are identified as a time-conflict candidate task pair. Within this pair, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the advertising task's target area within the overlapping area. A spatial conflict index is calculated, and the resource requirement parameters of the advertising task are extracted and normalized. The calculation involves weighted summation of normalized resource demand parameters to obtain resource demand indicators. The bandwidth occupancy and storage space utilization of the corresponding advertising terminal are obtained and normalized. A resource conflict degree indicator is calculated. The spatial conflict degree indicator and resource demand indicator are normalized separately. The normalized spatial conflict degree indicator and resource demand indicator are then weighted and summed to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than a preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then aggregated to obtain a potential conflict task set.

[0014] Compared with existing technologies, the beneficial effects of this invention are that by constructing a weighted conflict relationship graph, time conflicts, spatial conflicts, and resource conflicts in advertising are modeled in a unified manner, and combined with a multi-strategy scheduling mechanism and comprehensive evaluation indicators, global optimization scheduling of advertising tasks across multiple terminals is achieved. This not only effectively reduces advertising conflicts and improves the accuracy of advertising, but also significantly improves the system's resource utilization and load balancing capabilities, enhances the system's intelligence level and operational stability, and is suitable for complex multi-terminal advertising scenarios. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the cloud-based collaborative control system for multi-terminal advertising delivery according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 This invention provides a technical solution: a cloud-based collaborative control method for multi-terminal advertising delivery, the method comprising: Step S100: Collect resource status parameters and advertising task information from multiple advertising terminals to construct a basic dataset; The construction of the basic data set in step S100 includes: In the advertising platform, the operating status information of each advertising terminal is obtained by collecting cloud data. The operating status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operating status information is preprocessed, including removing abnormal data and supplementing missing values. Obtain advertising task information for each advertising terminal, including advertising delivery time information, target coverage area, and resource requirement parameters, including bandwidth and storage requirements. Based on the preprocessed running status information and advertising task information, construct a basic data set. For example, in the advertising platform, a total of 5 advertising terminal devices are connected, numbered T1, T2, T3, T4, and T5 respectively. The system collects operational status data from the cloud every 5 minutes. The raw operational status information collected at a certain moment is as follows: Terminal T1: Bandwidth utilization 65%, storage space utilization 72%, currently executing advertising tasks A101 and A102; Terminal T2: Bandwidth utilization 92%, storage space utilization 88%, currently executing advertising task A103; Terminal T3: Bandwidth utilization -5%, storage space utilization 55%, currently executing advertising task A104; Terminal T4: Bandwidth utilization is 48%, storage space utilization is missing, currently executing advertising task A105; Terminal T5: Bandwidth utilization 75%, storage space utilization 81%, currently executing advertising tasks A106 and A107; The above operational status information is preprocessed as follows: Terminal T3's bandwidth utilization rate was -5%, which is not within the actual range and is therefore considered an outlier and removed. The storage space utilization rate of terminal T4 is missing. It is supplemented by the average value of terminals in the same area. The average storage utilization rate of T1, T2 and T5 is calculated to be (72%+88%+81%) / 3≈80%. Therefore, the storage space utilization rate of T4 is supplemented to 80%. The corresponding advertising task information is as follows: Advertising Task A101: Ad banner "Beverage Brand A Spring Promotion", campaign period 2026-04-19 08:00 to 2026-04-19 20:00, target area "commercial center", resource requirements parameters: bandwidth requirement 5Mbps, storage requirement 500MB; Advertising Task A102: Ad banner "New Mobile Phone Launch", running time from 10:00 on April 19, 2026 to 22:00 on April 25, 2026, target area "subway station", resource requirements parameters: bandwidth requirement 8Mbps, storage requirement 1.2GB; Advertising Task A103: Ad brand “Car Brand B Ad”, campaign period from 09:00 on April 18, 2026 to 21:00 on April 30, 2026, target area “highway intersection”, resource requirements: bandwidth 10Mbps, storage 2GB. Advertising Task A104: Ad label "Restaurant Chain Promotion", placement time 2026-04-19 11:00 to 2026-04-19 14:00, target area "office buildings", resource requirements parameters: bandwidth requirement 4Mbps, storage requirement 300MB; Advertising Task A105: Ad label "E-commerce platform promotion", campaign period 2026-04-19 00:00 to 2026-04-20 23:59, target area "residential area", resource requirements parameters: bandwidth requirement 6Mbps, storage requirement 800MB; Advertising Task A106: Ad Label "Tourism Promotion Ad", Placement Time: 2026-04-19 09:00 to 2026-04-28 18:00, Target Area: "Airport", Resource Requirements: Bandwidth Requirement: 7Mbps, Storage Requirement: 1GB; Ad Task A107: Ad label "Education and Training Courses", campaign period from 13:00 on April 19, 2026 to 21:00 on May 10, 2026, target area "Schools and surrounding areas", resource requirements: bandwidth 3Mbps, storage 400MB; Based on the preprocessed runtime status information and advertising task information described above, the following basic dataset is constructed: (T1, 65%, 72%, A101, A102, "Business Center / Metro Station", total bandwidth requirement 13Mbps, total storage requirement 1.7GB) (T2, 92%, 88%, A103, "Highway Intersection", bandwidth requirement 10Mbps, storage requirement 2GB) (T3, bandwidth missing, 55%, A104, "office building", bandwidth requirement 4Mbps, storage requirement 300MB) (T4, 48%, 80%, A105, "Residential Area", bandwidth requirement 6Mbps, storage requirement 800MB) (T5, 75%, 81%, A106, A107, “Airport / School Area”, total bandwidth requirement 10Mbps, total storage requirement 1.4GB).

[0018] Step S200: Obtain the geographical location and coverage radius of the advertising terminal, divide the terminal coverage area into spatial grids, calculate the coverage overlap area ratio between any two terminals, and construct a weighted conflict relationship graph; The construction of the weighted conflict relationship graph in step S200 includes: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, the coverage radius corresponding to each advertising terminal is collected, and a two-dimensional coverage area corresponding to each advertising terminal is generated according to the coordinates and coverage radius. In the plane coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage areas of each advertising terminal are mapped to the set of grid units. Based on the spatial positional relationship between the grid units and the two-dimensional coverage areas, the set of advertising terminals corresponding to each grid unit is determined. Based on the set of advertising terminals in the grid unit, the number of advertising terminals covered by each grid unit is counted, and a correspondence table between grid units and advertising terminals is established. For any two advertising terminals, the number of grid units covered by both advertising terminals is multiplied by the area of ​​a single grid unit based on the correspondence table, which is used as the approximate intersection area of ​​the coverage areas of the two terminals. The number of grid units covered by any terminal is multiplied by the area of ​​a single grid unit, which is used as the approximate union area. The ratio of the approximate intersection area to the approximate union area is determined as the percentage of overlapping coverage area between the two advertising terminals. Using each advertising terminal as a graph node, when the percentage of overlapping area between any two advertising terminals is greater than a preset threshold, a connection edge is established between the corresponding nodes, and the percentage of overlapping area is used as the weight of the edge to construct a weighted conflict relationship graph. For example, based on the aforementioned advertising terminals T1 to T5, the geographical location information of each terminal is obtained and mapped to a unified planar coordinate system, resulting in the following coordinates: Terminal T1: Original location longitude 116.397, latitude 39.908, mapped coordinates (0.0, 0.0); Terminal T2: Original location longitude 116.407, latitude 39.918, mapped coordinates (1.1, 1.2). Terminal T3: Original location longitude 116.392, latitude 39.915, mapped coordinates (-0.6, 0.8); Terminal T4: Original location longitude 116.385, latitude 39.905, mapped coordinates (-1.2, -0.3). Terminal T5: Original location longitude 116.410, latitude 39.900, mapped coordinates (1.4, -0.8). The coverage radii of each advertising terminal are as follows: T1 coverage radius 0.9km, T2 coverage radius 0.8km, T3 coverage radius 0.7km, T4 coverage radius 0.6km, and T5 coverage radius 1.0km; Construct two-dimensional coverage areas for each terminal. T1 coverage area: center (0.0, 0.0), radius 0.9; T2 coverage area: center (1.1, 1.2), radius 0.8; T3 coverage area: center (-0.6, 0.8), radius 0.7; T4 coverage area: center (-1.2, -0.3), radius 0.6; T5 coverage area: center (1.4, -0.8), radius 1.0; Calculate the union range of all covered areas to obtain the overall boundary: X range: [-1.8, 2.4], Y range: [-1.8, 2.0]. Set the grid resolution to 0.2km × 0.2km, then the area of ​​a single grid cell is 0.04 km². Divide the above range into grids to obtain approximately 420 grid cells.

[0019] Step S300: Based on the weighted conflict relationship graph and the basic data set, perform conflict detection on advertising tasks across different terminals, identify conflicting combinations of advertising tasks, and obtain a set of potential conflicting tasks. The set of potential conflicting tasks obtained in step S300 includes: Based on the weighted conflict relationship graph, obtain any two pairs of advertising terminals with connected edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set to extract the set of advertising tasks being executed and to be executed on each advertising terminal. Obtain the weight of the connection edge corresponding to any two connected advertising terminals. If the weight of the connection edge reaches the preset conflict threshold, it is determined that there is a potential spatial conflict relationship between the corresponding terminals. Then, the advertising tasks of the two terminals are combined in pairs to construct a set of candidate task pairs. In any pair of candidate tasks, the time interval for placing ads is obtained based on the ad placement time information in the ad task information. When the time intervals for placing ads of two ad tasks overlap, they are determined to be a pair of candidate tasks with time conflicts. In the time-conflict candidate task pairs, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the advertising task target area in the overlapping area. The spatial conflict degree index is calculated according to the following formula: ; Where A represents the spatial conflict index, B represents the overlapping area between advertising terminals, B1 and B2 represent the coverage areas of the first and second advertising terminals in the time conflict candidate task pair, respectively, and C1 and C2 represent the distribution ratio of the target areas of the advertising tasks being executed by the first and second advertising terminals in the time conflict candidate task pair within the overlapping area. Extract the resource requirement parameters of the advertising task, normalize the resource requirement parameters, and perform a weighted summation of the normalized resource requirement parameters to calculate the resource requirement index. Obtain the bandwidth occupancy rate and storage space utilization rate of the corresponding advertising terminal and perform normalization calculation. Calculate the resource conflict degree index as D=a1×(E1×b1+E2×b2)+a2×(F1×b1+F2×b2), where D represents the resource conflict degree index, E1 and E2 represent the normalized bandwidth occupancy rate of the first and second advertising terminals in the time conflict candidate task pair, F1 and F2 represent the normalized storage space utilization rate of the first and second advertising terminals in the time conflict candidate task pair, b1 and b2 represent the resource requirement index of the advertising task being executed by the first and second advertising terminals in the time conflict candidate task pair, and a1 and a2 represent the weights of the bandwidth occupancy rate and storage space utilization rate, respectively. The spatial conflict degree index and the resource demand index are normalized and calculated separately. The normalized spatial conflict degree index and the resource demand index are then weighted and summed to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than the preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then summarized to obtain a set of potential conflict tasks.

[0020] Step S400: Based on the set of potential conflicting tasks, generate several sets of task scheduling combination schemes; Among them, the generation of several task scheduling combination schemes in step S400 includes: A preset scheduling analysis time interval is set, and the scheduling analysis time interval is divided into several continuous time segments. Based on the original delivery time information of each advertising task in the potential conflict task set, each advertising task is matched to the corresponding time segment to obtain the distribution of advertising tasks in the time dimension. For advertising tasks within the set of potential conflicting tasks, a scheduling adjustment method is generated. This scheduling adjustment method includes adjustments to the delivery time, the delivery terminal, and the task structure. The process of adjusting the delivery time is as follows: for any advertising task, the delivery time is increased by an offset according to a preset time step, the offset delivery time interval is calculated, the delivery time interval is used as a candidate delivery time interval, and the time overlap detection is performed between each candidate delivery time interval and other advertising tasks in the potential conflict task set. When the time overlap between the adjusted delivery time interval and other advertising tasks is lower than a preset threshold, the corresponding candidate delivery time interval is determined as a feasible delivery time interval. The process of adjusting the advertising terminal is as follows: A set of candidate advertising terminals is obtained from other advertising terminals besides the original ones. For any candidate advertising terminal, the bandwidth occupancy rate, storage space utilization rate, and coverage area information within the advertising task's delivery time interval are obtained. The allocated bandwidth occupancy rate and storage space utilization rate within the advertising task's delivery time interval are subtracted from the total bandwidth occupancy rate and storage space utilization rate of the advertising terminal to calculate the remaining bandwidth occupancy rate and storage space utilization rate. The remaining bandwidth occupancy rate and storage space utilization rate are then normalized and weighted to obtain a remaining resource demand index. Based on the coverage area information, the coverage ratio of the advertising task's target area is calculated. When the remaining resource demand index is greater than the advertising task's resource demand index, and the coverage ratio is higher than a preset coverage threshold, the candidate advertising terminal is determined as a feasible alternative advertising terminal. The task structure adjustment method includes task splitting and task merging. The task splitting process includes dividing an advertising task into multiple sub-tasks and assigning them to different time segments and different advertising terminals. The task merging process includes merging multiple advertising tasks with similar placement times and target areas to form a new combined task. Based on the feasible delivery time interval, feasible alternative delivery terminals, and task structure adjustment scheme, calculate the delivery time interval of each advertising task in the adjusted delivery time interval, count the time overlap length between any two tasks, calculate the total time overlap length, divide the total time overlap length by the total delivery time, and calculate the adjusted time conflict degree. Based on the adjusted advertising terminal allocation, the coverage overlap area between each advertising terminal is calculated and summed to obtain the total coverage overlap area. Based on the task target coverage area of ​​the advertising task, the effective overlap area belonging to the task target coverage area in the coverage overlap area is calculated. The effective overlap area is divided by the total coverage overlap area to calculate the adjusted spatial overlap ratio. The resource usage of each advertising terminal within the corresponding time segment is statistically analyzed. The adjusted task resource requirements are then superimposed with the remaining resources of the terminal to calculate the resource load value of each terminal. Based on the adjustment conditions, the scheduling adjustment results are filtered to obtain several sets of task scheduling combination schemes. The adjustment conditions include: the degree of time conflict is lower than the preset time conflict threshold; the spatial overlap ratio is lower than the preset spatial conflict threshold; and the resource load of each advertising terminal does not exceed the carrying capacity threshold. For example, the preset scheduling analysis time interval is from 08:00 to 12:00 on April 19, 2026, and it is divided into continuous time segments of 30 minutes, forming a total of 8 time segments, namely: Time slot S1: 08:00-08:30; Time slot S2: 08:30-09:00; Time slot S3: 09:00-09:30; Time slot S4: 09:30-10:00; Time slot S5: 10:00-10:30; Time slot S6: 10:30-11:00; Time slot S7: 11:00-11:30; Time slot S8: 11:30-12:00; The potential conflict task set includes the following advertising tasks: Task A1: Original deployment terminal T1, deployment time 09:00-10:00, target area R1, resource requirement index 0.62; Task A2: Original deployment terminal T2, deployment time 09:20-10:20, target area R1, resource requirement index 0.58; Task A3: Original deployment terminal T3, deployment time 09:10-09:50, target area R2, resource requirement index 0.46; Task A4: Original deployment terminal T5, deployment time 10:00-11:00, target area R3, resource requirement index 0.55; Task A5: Original deployment terminal T4, deployment time 09:30-10:30, target area R2, resource requirement index 0.40; After matching the original delivery time to the corresponding time segment, the task time distribution is as follows: S3: A1, A2, A3; S4: A1, A2, A3, A5; S5: A2, A4, A5; S6: A4; S7: A4; Set the time step to 10 minutes, shift the time of task A2, and generate candidate delivery time intervals: Candidate Solution 1: The overlap with A1 is 30 minutes from 09:30 to 10:30, and with A5 is 60 minutes. The overall overlap rate is 0.75, which exceeds the threshold of 0.40 and is therefore not feasible. Candidate Solution 2: From 09:40 to 10:40, it overlaps with A1 for 20 minutes and with A5 for 50 minutes, with a combined overlap rate of 0.58, which exceeds the threshold and is therefore not feasible. Candidate Option 3: 10:00-11:00 overlaps with A1 for 0 minutes and with A4 for 60 minutes, but A4 is located in T5, which is a different spatial region. The time conflict rate is 0.20, so it is feasible. Candidate Option 4: 08:40-09:40 overlaps with A1 for 40 minutes and with A3 for 30 minutes, with a comprehensive overlap rate of 0.61, which is not feasible. Therefore, Candidate Option 3 is determined as the feasible deployment time interval for A2. For Task A1, alternative terminals are selected from other terminals T2, T3, T4, and T5. Assuming Task A1 is deployed between 09:00 and 10:00, the resource status of each terminal is as follows: T2 has a total bandwidth of 100Mbps, of which 55Mbps has been allocated; total storage is 80GB, of which 42GB has been used. T3 has a total bandwidth of 100Mbps, of which 38Mbps has been allocated; total storage is 80GB, of which 30GB has been used. T4 has a total bandwidth of 100Mbps, of which 28Mbps has been allocated; total storage is 80GB, of which 22GB has been used. T5 has a total bandwidth of 100Mbps, with 60Mbps already allocated; total storage is 80GB, with 40GB already used. T2 has 45% remaining bandwidth and 47.5% remaining storage. T3 has 62% remaining bandwidth and 62.5% remaining storage. T4 has 72% remaining bandwidth and 72.5% remaining storage. T5 has 40% remaining bandwidth and 50% remaining storage. Calculate the remaining resource demand index: The remaining resource demand index for T2 is: 0.6 × 0.45 + 0.4 × 0.475 = 0.460; The remaining resource demand index for T3 is: 0.6 × 0.62 + 0.4 × 0.625 = 0.622; The remaining resource demand index for T4 is: 0.6 × 0.72 + 0.4 × 0.725 = 0.722; The remaining resource demand index for T5 is: 0.6 × 0.40 + 0.4 × 0.50 = 0.440; Calculate the coverage ratio R1 of the target area: T2 coverage ratio of R1: 0.92; T3 coverage ratio for R1: 0.78; T4 coverage ratio for R1: 0.55; T5 coverage ratio for R1: 0.81; Assuming the coverage threshold is 0.75 and the resource requirement index for task A1 is 0.62, then T2: insufficient resources, not feasible; T3: resources are sufficient and coverage meets the standard, feasible; T4: insufficient coverage, not feasible; T5: insufficient resources, not feasible. Therefore, T3 is determined to be a feasible alternative deployment terminal for A1.

[0021] Plan P1: A1: Adjust from T1 to T3, 09:00-10:00; A2: Adjust to 10:00-11:00, terminal T2; A35: 09:20-10:20, terminal T3 / T4; A4 to be split for execution; The time conflict level is as follows: A1 and A35 overlap for 40 minutes, A2 and A4-1 overlap for 30 minutes, and the remaining tasks overlap for 0 minutes. The total overlap time is 70 minutes, the total task time is 260 minutes, and the time conflict level is 70 / 260=0.269. The spatial overlap ratios for the main terminal combinations after adjustment are as follows: T2 and T3 overlap ratio is 0.031, T3 and T4 overlap ratio is 0.025, T2 and T5 overlap ratio is 0.024, and the maximum spatial overlap ratio is 0.031. Resource load values ​​for the 09:30-10:00 time slice: T2 load 0.48, T3 load 0.74, T4 load 0.51, T5 load 0.43; Time slice 10:00-10:30: T2 load 0.69, T3 load 0.36, T4 load 0.28, T5 load 0.58, all terminals are below the carrying capacity threshold of 0.85; If the screening criteria are set as follows: time conflict threshold is 0.35, space conflict threshold is 0.05, and resource carrying capacity threshold is 0.85, then scheme P1 is satisfied.

[0022] Step S500: Calculate the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme. Based on the total conflict score and resource utilization balance index, calculate the scheme comprehensive evaluation index, select the optimal scheduling scheme, and adjust according to the optimal scheduling scheme. Among them, the generation of several task scheduling combination schemes in step S500 includes: In any set of task scheduling combination schemes, the advertising task pairs are collected, and the adjusted advertising task delivery time interval, target area information and corresponding advertising terminal coverage area are collected. The spatial conflict degree index and resource conflict degree index between advertising tasks are calculated. Based on the spatial conflict degree index and resource conflict degree index, the comprehensive conflict score of a single advertising task pair is calculated. The comprehensive conflict scores of all advertising task pairs in the task scheduling combination scheme are accumulated to obtain the total conflict score of the task scheduling combination scheme. Based on the adjusted advertising task allocation, the bandwidth occupancy and storage space utilization of each advertising terminal in each time segment are extracted. The bandwidth occupancy and storage space utilization are normalized and weighted, and then summed to calculate the resource utilization rate. The resource utilization of each time segment is statistically analyzed, and the resource utilization balance index is calculated according to the following formula: ; Where K represents the resource utilization balance index, Ri represents the resource utilization rate of the i-th time segment, R1 represents the average resource utilization rate of all time segments, and n represents the number of time segments; The total conflict score and resource utilization balance index are normalized and calculated. The weights of the total conflict score and resource utilization balance index are preset. The normalized total conflict score and resource utilization balance index are weighted and summed to obtain the comprehensive evaluation index of the scheme. All task scheduling combination schemes are traversed, and the task scheduling combination scheme with the smallest comprehensive evaluation index is selected as the optimal scheduling scheme. According to the optimal scheduling scheme, the advertising task scheduling execution instruction is generated and sent to the corresponding advertising terminal for advertising placement control.

[0023] To better implement the above methods, a cloud-based collaborative control system for multi-terminal advertising delivery is also proposed. The system includes a basic data set module, a weighted conflict relationship graph module, a potential conflict task set module, a task scheduling combination scheme module, and a scheduling adjustment module. Basic Data Set Module: Collects resource status parameters and advertising task information from multiple advertising terminals to construct a basic data set; The basic data set module includes a running status information unit and an advertising task information unit: Operational Status Information Unit: In the advertising platform, operational status information of each advertising terminal is obtained by collecting cloud data. The operational status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operational status information is preprocessed, including removing abnormal data and supplementing missing values. Advertising task information unit: Acquires advertising task information from each advertising terminal. The advertising task information includes advertising placement time information, target coverage area, and resource requirement parameters. The resource requirement parameters include bandwidth requirements and storage requirements. Based on the preprocessed running status information and advertising task information, a basic data set is constructed.

[0024] Weighted Conflict Relationship Graph Module: Obtain the geographical location and coverage radius of the advertising terminal, divide the terminal coverage area into spatial grids, calculate the coverage overlap area ratio between any two terminals, and construct a weighted conflict relationship graph; The weighted conflict graph module includes an advertising terminal collection unit and a unit for constructing the weighted conflict graph: Advertising terminal set unit: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, collect the coverage radius corresponding to each advertising terminal, and generate the two-dimensional coverage area corresponding to each advertising terminal according to the coordinates and coverage radius. Under the planar coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage area of ​​each advertising terminal is mapped to the set of grid units. Based on the spatial positional relationship between the grid unit and the two-dimensional coverage area, the advertising terminal set corresponding to each grid unit is determined. Constructing a weighted conflict graph unit: Based on a set of advertising terminals in grid units, count the number of advertising terminals covered by each grid unit and establish a correspondence table between grid units and advertising terminals. Traverse any two advertising terminals, and based on the correspondence table, determine the number of grid units simultaneously covered by both advertising terminals multiplied by the area of ​​a single grid unit as the approximate intersection area of ​​the coverage areas of the two terminals. Count the number of grid units covered by any terminal multiplied by the area of ​​a single grid unit as the approximate union area. Determine the ratio of the approximate intersection area to the approximate union area as the coverage overlap ratio between the two advertising terminals. Using each advertising terminal as a graph node, when the coverage overlap ratio between any two advertising terminals is greater than a preset threshold, establish a connection edge between the corresponding nodes and use the coverage overlap ratio as the edge weight to construct a weighted conflict graph.

[0025] Potential conflict task set module: Based on the weighted conflict relationship graph and the basic data set, the module performs conflict detection on advertising tasks across different terminals, identifies conflicting combinations of advertising tasks, and obtains a potential conflict task set. The potential conflict task set module includes a candidate task pair set unit and a potential conflict task set generation unit: Candidate task pair set unit: Based on the weighted conflict relationship graph, obtain any two advertising terminal pairs with connecting edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set, extract the set of advertising tasks being executed and to be executed on each advertising terminal, obtain the weight of the connecting edge corresponding to any two advertising terminals with connecting relationship, if the weight of the connecting edge reaches the preset conflict threshold, it is determined that the corresponding terminals have a potential spatial conflict relationship, and the advertising tasks of the two terminals are combined in pairs to construct a candidate task pair set; The process involves generating a potential conflict task set unit: In any pair of candidate tasks, based on the advertising delivery time information in the advertising task information, the delivery time interval is obtained. When the delivery time intervals of two advertising tasks overlap, they are identified as a time-conflict candidate task pair. Within this pair, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the advertising task's target area within the overlapping area. A spatial conflict index is calculated, and the resource requirement parameters of the advertising task are extracted and normalized. The calculation involves weighted summation of normalized resource demand parameters to obtain resource demand indicators. The bandwidth occupancy and storage space utilization of the corresponding advertising terminal are obtained and normalized. A resource conflict degree indicator is calculated. The spatial conflict degree indicator and resource demand indicator are normalized separately. The normalized spatial conflict degree indicator and resource demand indicator are then weighted and summed to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than a preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then aggregated to obtain a potential conflict task set.

[0026] Task scheduling combination scheme module: Based on the set of potentially conflicting tasks, generate several sets of task scheduling combination schemes; The scheduling and adjustment module calculates the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme. Based on the total conflict score and resource utilization balance index, it calculates the comprehensive evaluation index of the scheme, selects the optimal scheduling scheme, and adjusts according to the optimal scheduling scheme.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A cloud-based collaborative control method for multi-terminal advertising delivery, characterized in that: The methods include: Step S100: Collect resource status parameters and advertising task information from multiple advertising terminals to construct a basic dataset; Step S200: Obtain the geographical location and coverage radius of the advertising terminal, divide the terminal coverage area into spatial grids, calculate the coverage overlap area ratio between any two terminals, and construct a weighted conflict relationship graph; The construction of the weighted conflict relationship graph in step S200 includes the following steps: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, the coverage radius corresponding to each advertising terminal is collected, and a two-dimensional coverage area corresponding to each advertising terminal is generated according to the coordinates and coverage radius. In the plane coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage areas of each advertising terminal are mapped to the set of grid units. Based on the spatial positional relationship between the grid units and the two-dimensional coverage areas, the set of advertising terminals corresponding to each grid unit is determined. Based on the set of advertising terminals in the grid unit, the number of advertising terminals covered by each grid unit is counted, and a correspondence table between grid units and advertising terminals is established. For any two advertising terminals, the number of grid units covered by both advertising terminals is multiplied by the area of ​​a single grid unit based on the correspondence table, which is used as the approximate intersection area of ​​the coverage areas of the two terminals. The number of grid units covered by any terminal is multiplied by the area of ​​a single grid unit, which is used as the approximate union area. The ratio of the approximate intersection area to the approximate union area is determined as the percentage of overlapping coverage area between the two advertising terminals. Using each advertising terminal as a graph node, when the percentage of overlapping area between any two advertising terminals is greater than a preset threshold, a connection edge is established between the corresponding nodes, and the percentage of overlapping area is used as the weight of the edge to construct a weighted conflict relationship graph. Step S300: Based on the weighted conflict relationship graph and the basic data set, perform conflict detection on advertising tasks across different terminals, identify conflicting combinations of advertising tasks, and obtain a set of potential conflicting tasks. Step S400: Based on the set of potential conflicting tasks, generate several sets of task scheduling combination schemes; Step S500: Calculate the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme. Based on the total conflict score and resource utilization balance index, calculate the scheme comprehensive evaluation index, select the optimal scheduling scheme, and adjust according to the optimal scheduling scheme.

2. The cloud-based collaborative control method for multi-terminal advertising delivery according to claim 1, characterized in that, The construction of the basic data set in step S100 includes the following steps: In the advertising platform, the operating status information of each advertising terminal is obtained by collecting cloud data. The operating status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operating status information is preprocessed, including removing abnormal data and supplementing missing values. The advertising task information of each advertising terminal is obtained. The advertising task information includes advertising placement time information, target coverage area and resource requirement parameters. The resource requirement parameters include bandwidth requirement and storage requirement. Based on the preprocessed running status information and advertising task information, a basic data set is constructed.

3. The cloud-based collaborative control method for multi-terminal advertising delivery according to claim 1, characterized in that, The step S300, which obtains a set of potential conflicting tasks, includes the following steps: Based on the weighted conflict relationship graph, obtain any two pairs of advertising terminals with connected edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set to extract the set of advertising tasks being executed and to be executed on each advertising terminal. Obtain the weight of the connection edge corresponding to any two connected advertising terminals. If the weight of the connection edge reaches the preset conflict threshold, it is determined that there is a potential spatial conflict relationship between the corresponding terminals. Then, the advertising tasks of the two terminals are combined in pairs to construct a set of candidate task pairs. In any pair of candidate tasks, the time interval for placing ads is obtained based on the ad placement time information in the ad task information. When the time intervals for placing ads of two ad tasks overlap, they are determined to be a pair of candidate tasks with time conflicts. In the time-conflict candidate task pairs, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the advertising task target area in the overlapping area. The spatial conflict degree index is calculated according to the following formula: ； Where A represents the spatial conflict index, B represents the overlapping area between advertising terminals, B1 and B2 represent the coverage areas of the first and second advertising terminals in the time conflict candidate task pair, respectively, and C1 and C2 represent the distribution ratio of the target areas of the advertising tasks being executed by the first and second advertising terminals in the time conflict candidate task pair within the overlapping area. Extract the resource requirement parameters of the advertising task, normalize the resource requirement parameters, and perform a weighted summation of the normalized resource requirement parameters to calculate the resource requirement index. Obtain the bandwidth occupancy rate and storage space utilization rate of the corresponding advertising terminal and perform normalization calculation. Calculate the resource conflict degree index as D=a1×(E1×b1+E2×b2)+a2×(F1×b1+F2×b2), where D represents the resource conflict degree index, E1 and E2 represent the normalized bandwidth occupancy rate of the first and second advertising terminals in the time conflict candidate task pair, F1 and F2 represent the normalized storage space utilization rate of the first and second advertising terminals in the time conflict candidate task pair, b1 and b2 represent the resource requirement index of the advertising task being executed by the first and second advertising terminals in the time conflict candidate task pair, and a1 and a2 represent the weights of the bandwidth occupancy rate and storage space utilization rate, respectively. The spatial conflict degree index and the resource demand index are normalized and calculated separately. The normalized spatial conflict degree index and the resource demand index are then weighted and summed to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than the preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then summarized to obtain a set of potential conflict tasks.

4. The cloud-based collaborative control method for multi-terminal advertising delivery according to claim 1, characterized in that, Step S400 generates several sets of task scheduling combination schemes, including the following steps: A preset scheduling analysis time interval is set, and the scheduling analysis time interval is divided into several continuous time segments. Based on the original delivery time information of each advertising task in the potential conflict task set, each advertising task is matched to the corresponding time segment to obtain the distribution of advertising tasks in the time dimension. For advertising tasks within the set of potential conflicting tasks, a scheduling adjustment method is generated. This scheduling adjustment method includes adjustments to the delivery time, the delivery terminal, and the task structure. The process of adjusting the delivery time is as follows: for any advertising task, the delivery time is increased by an offset according to a preset time step, the offset delivery time interval is calculated, the delivery time interval is used as a candidate delivery time interval, and the time overlap detection is performed between each candidate delivery time interval and other advertising tasks in the potential conflict task set. When the time overlap between the adjusted delivery time interval and other advertising tasks is lower than a preset threshold, the corresponding candidate delivery time interval is determined as a feasible delivery time interval. The process of adjusting the advertising terminal is as follows: A set of candidate advertising terminals is obtained from other advertising terminals besides the original ones. For any candidate advertising terminal, the bandwidth occupancy rate, storage space utilization rate, and coverage area information within the advertising task's delivery time interval are obtained. The allocated bandwidth occupancy rate and storage space utilization rate within the advertising task's delivery time interval are subtracted from the total bandwidth occupancy rate and storage space utilization rate of the advertising terminal to calculate the remaining bandwidth occupancy rate and storage space utilization rate. The remaining bandwidth occupancy rate and storage space utilization rate are then normalized and weighted to obtain a remaining resource demand index. Based on the coverage area information, the coverage ratio of the advertising task's target area is calculated. When the remaining resource demand index is greater than the advertising task's resource demand index, and the coverage ratio is higher than a preset coverage threshold, the candidate advertising terminal is determined as a feasible alternative advertising terminal. The task structure adjustment method includes task splitting and task merging. The task splitting process includes dividing an advertising task into multiple sub-tasks and assigning them to different time segments and different advertising terminals. The task merging process includes merging multiple advertising tasks with similar placement times and target areas to form a new combined task. Based on the feasible delivery time interval, feasible alternative delivery terminals, and task structure adjustment scheme, calculate the delivery time interval of each advertising task in the adjusted delivery time interval, count the time overlap length between any two tasks, calculate the total time overlap length, divide the total time overlap length by the total delivery time, and calculate the adjusted time conflict degree. Based on the adjusted advertising terminal allocation, the coverage overlap area between each advertising terminal is calculated and summed to obtain the total coverage overlap area. Based on the task target coverage area of ​​the advertising task, the effective overlap area belonging to the task target coverage area in the coverage overlap area is calculated. The effective overlap area is divided by the total coverage overlap area to calculate the adjusted spatial overlap ratio. The resource usage of each advertising terminal within the corresponding time segment is statistically analyzed. The adjusted task resource requirements are then superimposed with the remaining resources of the terminal to calculate the resource load value of each terminal. Based on the adjustment conditions, the scheduling adjustment results are filtered to obtain several sets of task scheduling combination schemes. The adjustment conditions include: the degree of time conflict is lower than the preset time conflict threshold; the spatial overlap ratio is lower than the preset spatial conflict threshold; and the resource load of each advertising terminal does not exceed the carrying capacity threshold.

5. The cloud-based collaborative control method for multi-terminal advertising delivery according to claim 1, characterized in that, Step S500 generates several sets of task scheduling combination schemes, including the following steps: In any set of task scheduling combination schemes, the advertising task pairs are collected, and the adjusted advertising task delivery time interval, target area information and corresponding advertising terminal coverage area are collected. The spatial conflict degree index and resource conflict degree index between advertising tasks are calculated. Based on the spatial conflict degree index and resource conflict degree index, the comprehensive conflict score of a single advertising task pair is calculated. The comprehensive conflict scores of all advertising task pairs in the task scheduling combination scheme are accumulated to obtain the total conflict score of the task scheduling combination scheme. Based on the adjusted advertising task allocation, the bandwidth occupancy and storage space utilization of each advertising terminal in each time segment are extracted. The bandwidth occupancy and storage space utilization are normalized and weighted, and then summed to calculate the resource utilization rate. The resource utilization of each time segment is statistically analyzed, and the resource utilization balance index is calculated according to the following formula: ； Where K represents the resource utilization balance index, Ri represents the resource utilization rate of the i-th time segment, R1 represents the average resource utilization rate of all time segments, and n represents the number of time segments; The total conflict score and resource utilization balance index are normalized and calculated. The weights of the total conflict score and resource utilization balance index are preset. The normalized total conflict score and resource utilization balance index are weighted and summed to obtain the comprehensive evaluation index of the scheme. All task scheduling combination schemes are traversed, and the task scheduling combination scheme with the smallest comprehensive evaluation index is selected as the optimal scheduling scheme. According to the optimal scheduling scheme, the advertising task scheduling execution instruction is generated and sent to the corresponding advertising terminal for advertising placement control.

6. A cloud-based collaborative control system for multi-terminal advertising delivery, used to implement the cloud-based collaborative control method for multi-terminal advertising delivery as described in any one of claims 1-5, characterized in that, The system includes a basic data set module, a weighted conflict relationship graph module, a potential conflict task set module, a task scheduling combination scheme module, and a scheduling adjustment module; The basic data set module collects resource status parameters and advertising task information from multiple advertising terminals to construct a basic data set. The weighted conflict graph module: obtains the geographical location and coverage radius of the advertising terminal, divides the terminal coverage area into spatial grids, calculates the percentage of overlapping coverage area between any two terminals, and constructs a weighted conflict graph; The potential conflict task set module: Based on the weighted conflict relationship graph and the basic data set, it performs conflict detection on advertising tasks across different terminals, identifies combinations of advertising tasks that conflict, and obtains a potential conflict task set. The task scheduling combination scheme module generates several sets of task scheduling combination schemes based on the set of potentially conflicting tasks. The scheduling and adjustment module calculates the total conflict score and resource utilization balance index corresponding to each task scheduling combination scheme, calculates the scheme comprehensive evaluation index based on the total conflict score and resource utilization balance index, selects the optimal scheduling scheme, and adjusts according to the optimal scheduling scheme.

7. The cloud-based collaborative control system for multi-terminal advertising delivery according to claim 6, characterized in that, The basic data set module includes a running status information unit and an advertising task information unit: The operational status information unit: In the advertising platform, operational status information of each advertising terminal is obtained by collecting cloud data. The operational status information includes bandwidth utilization, storage space utilization, and information on advertising tasks being executed. The bandwidth utilization is calculated by dividing the used bandwidth by the maximum available bandwidth of the terminal. The storage space utilization is calculated by dividing the current used storage capacity by the total storage capacity. The collected operational status information is preprocessed, including removing abnormal data and supplementing missing values. The advertising task information unit acquires advertising task information from each advertising terminal. The advertising task information includes advertising placement time information, target coverage area, and resource requirement parameters. The resource requirement parameters include bandwidth requirements and storage requirements. Based on the preprocessed running status information and advertising task information, a basic data set is constructed.

8. The cloud-based collaborative control system for multi-terminal advertising delivery according to claim 6, characterized in that, The weighted conflict graph module includes an advertising terminal collection unit and a weighted conflict graph construction unit: The advertising terminal set unit: Based on the geographical location of each advertising terminal, each advertising terminal is mapped to a unified planar coordinate system to generate the coordinates of each advertising terminal, the coverage radius corresponding to each advertising terminal is collected, and a two-dimensional coverage area corresponding to each advertising terminal is generated according to the coordinates and coverage radius. Under the planar coordinate system, the union of the two-dimensional coverage areas of each advertising terminal is used as the grid division range. The grid division range is spatially divided into several grid units using a preset resolution. The two-dimensional coverage area of ​​each advertising terminal is mapped to the set of grid units, and the advertising terminal set corresponding to each grid unit is determined based on the spatial positional relationship between the grid unit and the two-dimensional coverage area. The construction of the weighted conflict graph unit involves: based on a set of advertising terminals in grid units, counting the number of advertising terminals covered by each grid unit, and establishing a correspondence table between grid units and advertising terminals. For any two advertising terminals, the number of grid units simultaneously covered by both terminals is multiplied by the area of ​​a single grid unit, which is used as the approximate intersection area of ​​the two terminals' coverage areas. The number of grid units covered by any terminal is multiplied by the area of ​​a single grid unit, which is used as the approximate union area. The ratio of the approximate intersection area to the approximate union area is determined as the percentage of overlapping coverage area between the two advertising terminals. Each advertising terminal is used as a graph node. When the percentage of overlapping coverage area between any two advertising terminals exceeds a preset threshold, a connection edge is established between the corresponding nodes, and the percentage of overlapping coverage area is used as the weight of the edge, thus constructing a weighted conflict graph.

9. The cloud-based collaborative control system for multi-terminal advertising delivery according to claim 6, characterized in that, The potential conflict task set module includes a candidate task pair set unit and a potential conflict task set generation unit: The candidate task pair set unit: Based on the weighted conflict relationship graph, obtain any two advertising terminal pairs with connecting edges. In each pair of advertising terminals, call the corresponding advertising task information in the basic data set, extract the set of advertising tasks being executed and to be executed on each advertising terminal, obtain the weight of the connecting edge corresponding to any two advertising terminals with connecting relationship, if the weight of the connecting edge reaches the preset conflict threshold, determine that the corresponding terminals have a potential spatial conflict relationship, and combine the advertising tasks of the two terminals in pairs to construct a candidate task pair set; The unit for generating a potential conflict task set: In any pair of candidate tasks, based on the advertising placement time information in the advertising task information, the placement time interval is obtained. When the placement time intervals of two advertising tasks overlap, they are determined to be a time conflict candidate task pair. In the time conflict candidate task pair, the target coverage area of ​​the advertising task is obtained, and the area in the overlapping area is calculated. The area of ​​the overlapping area is divided by the target coverage area to calculate the distribution ratio of the target area of ​​the advertising task in the overlapping area. The spatial conflict degree index is calculated, the resource requirement parameters of the advertising task are extracted, and the resource requirement parameters are normalized. The system performs weighted summation of normalized resource demand parameters to obtain resource demand indicators. It also obtains and normalizes the bandwidth occupancy and storage space utilization of the corresponding advertising terminal, calculates resource conflict degree indicators, and performs normalized calculations on both the spatial conflict degree indicator and the resource demand indicator. Finally, it performs weighted summation of the normalized spatial conflict degree indicator and the resource demand indicator to obtain a comprehensive conflict score. When the comprehensive conflict score is greater than a preset conflict judgment threshold, the corresponding candidate task pair is determined to be a conflict task pair. All advertising tasks determined to be conflict task pairs are then aggregated to obtain a set of potential conflict tasks.

Images