A Smart City Lighting Advertising Management Method and System

Abstract

This invention provides a smart city lighting and advertising management method and system, comprising: collecting state data of lighting equipment, advertising resources, and environmental scenes through a multi-source sensing network; preprocessing and fusing the data using edge computing nodes to obtain a standardized data stream; based on the standardized data stream, bucketing the lighting equipment and advertising resources using local sensitive hashing, and optimizing the data stream through iterative local search to obtain resource grouping results; constructing a randomized regular graph model, optimizing the graph structure using a Markov chain Monte Carlo algorithm, extracting the mapping relationship between lighting equipment and advertising resources, and generating a resource allocation strategy; based on the resource allocation strategy, combining user behavior data and environmental scene state data, generating a collaborative control strategy through a multi-objective optimization algorithm; and controlling the lighting equipment and advertising resources and collecting feedback data based on the collaborative control strategy. This invention achieves intelligent collaborative management of urban lighting and advertising resources.

Description

Technical Field

[0001] This invention relates to the field of smart city management technology, and in particular to a smart city lighting advertising management method and system. Background Technology

[0002] Smart city construction is a significant trend in current urban development. Urban lighting systems and outdoor advertising displays, as crucial components of a city's image, play a vital role in enhancing urban quality and economic value. Smart city lighting and advertising management involves core technology areas such as large-scale distributed equipment control, resource optimization, and diversified content management.

[0003] Traditional urban lighting systems are typically managed through centralized control or preset schedules, such as time-zone-based lighting control systems or simple light-sensing technology for brightness adjustment. Traditional outdoor advertising management relies primarily on manual inspections and periodic maintenance, with content updates dependent on manual intervention, lacking intelligent and automated management methods.

[0004] Currently, more advanced smart city lighting systems are beginning to combine IoT technology and big data analytics with lighting systems. By establishing a centralized management platform, they enable remote monitoring and control of urban lighting equipment. These systems typically employ a hierarchical data processing architecture, performing simple cluster analysis on collected environmental data and equipment status, and then making control decisions based on preset rules. However, in terms of advertising resource management, they use fixed resource allocation strategies and cannot dynamically optimize advertising content based on regional characteristics and pedestrian traffic.

[0005] However, existing technologies suffer from problems such as low data processing efficiency and insufficient resource allocation optimization. Particularly when processing large-scale urban lighting equipment data, traditional clustering algorithms have high computational complexity, making it difficult to meet real-time requirements. Simultaneously, the lack of an effective resource allocation model leads to conflicts between advertising display effects and urban lighting needs, preventing coordinated optimization between the two. Furthermore, existing systems typically treat lighting and advertising as independent systems, lacking a unified resource scheduling mechanism, resulting in energy waste and visual pollution.

[0006] Therefore, how to construct an efficient smart city lighting advertising management method and system to achieve synergistic optimization of urban lighting equipment and advertising resources is a technical problem that urgently needs to be solved. Summary of the Invention

[0007] The purpose of this invention is to provide a smart city lighting advertising management method and system, which aims to solve the technical problems of low data processing efficiency, insufficient resource allocation optimization, and lack of a unified resource scheduling mechanism in the existing technology, and to realize the collaborative optimization management of urban lighting equipment and advertising resources.

[0008] To achieve the above objectives, the technical solution provided by the present invention is as follows:

[0009] A smart city lighting advertising management method includes:

[0010] The status data of urban lighting equipment, advertising resources and environmental scenes are collected by a multi-source sensing network, and the status data is preprocessed and fused by edge computing nodes to obtain a standardized data stream.

[0011] Based on the standardized data stream, locality-sensitive hashing is used to bucket lighting devices and advertising resources, and iterative local search is used to optimize the resource grouping results.

[0012] Based on the resource grouping results, a random regular graph model is constructed, and the graph structure is optimized using the Markov chain Monte Carlo algorithm to extract the mapping relationship between lighting equipment and advertising resources and generate a resource allocation strategy.

[0013] Based on the resource allocation strategy, combined with user behavior data and the state data of the environmental scenario, a collaborative control strategy is generated through a multi-objective optimization algorithm.

[0014] Based on the aforementioned collaborative control strategy, lighting equipment and advertising resources are controlled; simultaneously, operational status data and display effect data of both the lighting equipment and the advertising resources are collected, and the operational status data and display effect data are fed back to the multi-source sensing network as optimization inputs to achieve smart city lighting advertising management.

[0015] Optionally, the step of collecting state data of urban lighting equipment, advertising resources, and environmental scenes through a multi-source sensing network, and preprocessing and fusing the state data using edge computing nodes to obtain a standardized data stream includes:

[0016] The system collects status data of the lighting equipment, the advertising resources, and the environmental scene, and transmits the status data to the edge computing node through a preset data acquisition protocol; wherein the status data includes power parameters, brightness parameters, advertising screen size parameters, resolution parameters, and environmental sensor operating data;

[0017] Based on the state data, data preprocessing and spatiotemporal synchronization are performed at the edge computing node to generate a data stream;

[0018] The data stream is cleaned and its features are extracted to obtain the standardized data stream.

[0019] Optionally, the step of using locality-sensitive hashing to bucket lighting devices and advertising resources, and optimizing through iterative local search to obtain resource grouping results, includes:

[0020] Based on the standardized data stream, feature vectors are constructed using dimensionality reduction and feature extraction techniques.

[0021] Based on the feature vector, the lighting devices and advertising resources are initially bucketed using a locality-sensitive hash function.

[0022] The initial bucketing results are iteratively optimized using local search to obtain the resource grouping results.

[0023] Optionally, the iterative local search optimization of the initial bucketing results to obtain the resource grouping results includes:

[0024] Using each bucket of the initial bucketing result as an initial cluster, a clustering objective function is constructed; wherein, the clustering objective function includes intra-cluster similarity, spatial continuity, energy load balancing, and advertising coverage efficiency;

[0025] Based on the clustering objective function, a local search is performed on the data points within each initial cluster, and the clustering objective function is updated using incremental computation techniques.

[0026] When the clustering objective function is determined to be convergent, the resource grouping result is obtained.

[0027] Optionally, constructing the random regular graph model includes:

[0028] Based on the resource grouping results and combined with the state data of the environmental scene, a heterogeneous graph structure of lighting device nodes, advertising resource nodes, and environmental scene nodes is constructed.

[0029] The heterogeneous graph structure is subjected to degree distribution regularization and edge weight randomization to obtain the randomized regular graph model.

[0030] Optionally, the optimization of the graph structure using the Markov chain Monte Carlo algorithm includes:

[0031] Based on the aforementioned stochastic regular graph model, an energy function is constructed that incorporates resource allocation balance and system responsiveness.

[0032] The random regular graph model is iteratively optimized using a block-parallel Markov chain Monte Carlo algorithm to obtain an optimized graph structure.

[0033] Based on the optimized graph structure and the energy function, the mapping relationship between device nodes and resource nodes is extracted to generate the resource configuration strategy.

[0034] Optionally, the step of generating a cooperative control strategy through a multi-objective optimization algorithm includes:

[0035] Based on the resource allocation strategy, a multi-objective function including energy efficiency and visual comfort is constructed;

[0036] The Pareto optimal solution for resource allocation is obtained by optimizing the multi-objective function.

[0037] The cooperative control strategy is generated based on the Pareto optimal solution.

[0038] Optionally, the combination of user behavior data and the state data of the environmental scene includes:

[0039] Collect ambient light intensity, weather conditions, and pedestrian density data to obtain the state data of the environmental scene;

[0040] The user behavior data is obtained by acquiring user dwell time and attention heatmaps based on video analytics technology.

[0041] Spatiotemporal correlation analysis is performed on the state data of the environmental scene and the user behavior data to obtain the scene feature vector.

[0042] Optionally, controlling the lighting equipment and advertising resources based on the collaborative control strategy includes:

[0043] The collaborative control strategy is converted into the brightness parameters of the lighting equipment and the display parameters of the advertising resources;

[0044] The brightness of the lighting equipment and the content displayed by the advertising resources are adjusted through an Internet of Things (IoT) control system.

[0045] Collect energy consumption data of the lighting equipment and display effect data of the advertising resources.

[0046] The present invention also provides a smart city lighting advertising management device, comprising:

[0047] The data acquisition module is used to collect status data of urban lighting equipment, advertising resources and environmental scenes through a multi-source sensing network, and to preprocess and fuse the status data using edge computing nodes to obtain a standardized data stream.

[0048] The resource grouping module is used to bucket lighting devices and advertising resources based on the standardized data stream using locality-sensitive hashing, and to optimize the resource grouping results through iterative local search.

[0049] The resource mapping module is used to construct a random regular graph model based on the resource grouping results, optimize the graph structure through the Markov chain Monte Carlo algorithm, extract the mapping relationship between lighting equipment and advertising resources, and generate a resource allocation strategy.

[0050] The control decision module is used to generate a collaborative control strategy based on the resource allocation strategy, combined with user behavior data and the state data of the environmental scenario, through a multi-objective optimization algorithm.

[0051] The execution feedback module is used to control the lighting equipment and advertising resources based on the collaborative control strategy; at the same time, it collects the operating status data and display effect data of both the lighting equipment and the advertising resources, and feeds the operating status data and the display effect data as optimization inputs to the multi-source sensing network to realize smart city lighting advertising management.

[0052] The beneficial effects of this invention are:

[0053] 1. By combining an improved Locality Sensitive Hashing (LSH) method with iterative local search, a near-linear time clustering with O(nlog n) time complexity is achieved, solving the real-time problem of large-scale urban lighting equipment and advertising resource analysis.

[0054] 2. A stochastic regular graph model that transcends uniqueness was designed, and the Markov chain Monte Carlo (MCMC) algorithm with fast mixing properties was introduced to realize the optimal mapping relationship between lighting equipment and advertising resources.

[0055] 3. A four-layer architecture system based on the Internet of Things sensing layer, edge computing layer, cloud platform layer and application service layer was constructed, realizing the collection, processing and fusion of multi-source heterogeneous data.

[0056] 4. A multi-objective optimization decision-making system integrating energy efficiency, visual comfort, advertising display effect and environmental impact was developed, realizing the coordinated control of lighting and advertising.

[0057] An adaptive execution and feedback mechanism was designed, and a closed-loop optimization was formed through a real-time performance evaluation model, which improved the system's adaptability and resource utilization efficiency. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 A flowchart of a smart city lighting advertising management method provided in an embodiment of the present invention;

[0060] Figure 2 A flowchart of the near-linear time clustering algorithm provided in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of the structure of the collaborative optimization decision-making system provided in an embodiment of the present invention;

[0062] Figure 4 This is a diagram showing the module composition of the smart city lighting and advertising management system provided in an embodiment of the present invention. Detailed Implementation

[0063] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0064] like Figure 1 As shown, the smart city lighting advertising management method provided in this embodiment of the invention includes:

[0065] Step S1: Collect status data of urban lighting equipment, advertising resources and environmental scenes through a multi-source sensing network, and preprocess and fuse the status data using edge computing nodes to obtain a standardized data stream;

[0066] Data is collected through a multi-source sensing network deployed throughout the city. This network includes environmental sensors, lighting equipment sensors, pedestrian detectors, and advertising screen status monitors. Environmental sensors primarily collect environmental parameters such as light intensity, temperature, humidity, and air quality; lighting equipment sensors collect parameters such as lamp power, brightness, color temperature, and operating time; pedestrian detectors collect data on pedestrian numbers, flow direction, and density; and advertising screen status monitors collect information such as screen resolution, brightness, and content switching frequency. This multi-source heterogeneous data is transmitted via wired or wireless networks to edge computing nodes distributed throughout the city.

[0067] After receiving heterogeneous data from multiple sources, edge computing nodes first perform data cleaning to remove outliers, duplicates, and missing values. Then, they perform data standardization to convert data with different dimensions to a unified numerical range. Next, they synchronize data from different devices in time and space to ensure consistency across these dimensions. Finally, the edge computing nodes perform preliminary data fusion and feature extraction to generate a standardized data stream. This edge computing approach not only reduces data transmission volume and alleviates the computational burden on the cloud platform but also improves the system's response speed to local events.

[0068] Standardized data streams are transmitted from edge computing nodes to the cloud platform data center, where further in-depth processing and storage take place. The cloud platform utilizes big data technology to analyze historical data, uncovering potential patterns and trends. Simultaneously, the cloud platform has established a knowledge graph, structurally representing entities such as devices, resources, environment, and users, along with their relationships, providing knowledge support for subsequent intelligent decision-making. This multi-layered data processing architecture effectively handles heterogeneous data at the city scale, providing a solid data foundation for intelligent lighting and advertising management.

[0069] Step S2: Based on the standardized data stream, locality-sensitive hashing is used to bucket the lighting devices and advertising resources, and optimization is performed through iterative local search to obtain the resource grouping results;

[0070] In the specific implementation process, the system first constructs feature vectors from the standardized data stream. Through dimensionality reduction and feature extraction techniques, high-dimensional heterogeneous data is transformed into low-dimensional feature vectors, reducing computational complexity while preserving key data information. Then, based on these feature vectors, the system uses an improved Locality Sensitive Hash (LSH) algorithm to initially bucket lighting equipment and advertising resources. The core idea of ​​LSH is to design a set of hash functions that ensures a high probability of similar objects in the feature space being mapped to the same "bucket." This invention improves upon the traditional LSH algorithm by designing an adaptive family of hash functions, which can better adapt to the uneven distribution characteristics of urban spatial data.

[0071] After initial bucketing, the system optimizes the bucketing results using an iterative local search algorithm. The core of iterative local search is a "perturbation-improvement" cyclical strategy: in the improvement phase, each data point is considered for moving from its current cluster to a neighboring cluster; if this move improves the objective function value, the move is accepted. In the perturbation phase, a certain proportion of points are randomly selected for redistribution to escape local optima. To achieve near-linear time complexity, this invention employs optimization techniques such as proximity graph-based search range limitation, incremental computation, and early stopping strategies. Through this method, the system can efficiently generate resource grouping results, providing a foundation for subsequent resource mapping relationship construction.

[0072] Step S3: Based on the resource grouping results, construct a random regular graph model, optimize the graph structure using the Markov chain Monte Carlo algorithm, extract the mapping relationship between lighting equipment and advertising resources, and generate a resource allocation strategy.

[0073] Based on the resource grouping results, a multi-layered heterogeneous graph structure is constructed. This graph consists of lighting device nodes, advertising resource nodes, and environmental scene nodes, with edges between nodes representing physical proximity, functional complementarity, and visual impact. Then, the graph structure is subjected to stochastic regularization, including degree distribution regularization, edge weight randomization, and global structure balancing. This stochastic regularization method, which transcends uniqueness, enables the model to better reflect the uncertainty and variability of the real-world environment.

[0074] Next, the system applies the Markov Chain Monte Carlo algorithm, which features fast mixing, to optimize the graph structure. The algorithm first defines an energy function, comprehensively considering multiple aspects such as resource allocation balance, control system responsiveness, visual coordination, and energy efficiency. Then, it employs techniques such as a block-parallel MCMC strategy, adaptive proposal distribution, and temperature scheduling mechanism to improve the algorithm's convergence speed and optimization effect. The algorithm's fast mixing characteristic is mainly achieved through a nonlocal proposal mechanism, enabling MCMC to complete "long-distance jumps" in the state space in one step, thereby significantly improving the mixing speed.

[0075] After optimization, the system extracts the mapping relationship between lighting devices and advertising resources from a stable graph structure. This process employs techniques such as community detection, bipartite graph matching, and hyperedge mapping, enabling it to handle complex one-to-one, one-to-many, and many-to-many mapping relationships. Simultaneously, the system also incorporates a time-window-based mapping update mechanism and an event-triggered update mechanism to ensure that resource allocation strategies can dynamically adapt to environmental changes.

[0076] Step S4: Based on the resource allocation strategy, combined with user behavior data and the state data of the environmental scenario, a collaborative control strategy is generated through a multi-objective optimization algorithm;

[0077] A multi-objective function is constructed, encompassing dimensions such as energy efficiency, visual comfort, advertising display effectiveness, and environmental impact. The energy efficiency objective evaluates the ratio of the system's energy consumption to its service effectiveness; the visual comfort objective, based on a human visual perception model, assesses the combined visual effect of lighting and advertising; the advertising display effectiveness objective evaluates the visibility of advertising content and the efficiency of information delivery; and the environmental impact objective assesses the degree of the system's impact on the ecological environment.

[0078] Simultaneously, the system collects real-time environmental data, such as light intensity, weather conditions, and pedestrian density, through an environmental sensor network. Using video analytics, the system can also acquire behavioral data such as user dwell time in front of the advertising screen, gaze direction, and attention distribution. This environmental and user data is then analyzed in a spatiotemporal correlation to transform it into scene feature vectors, serving as crucial inputs for control decisions.

[0079] Based on a multi-objective function and scene feature vectors, the system applies an improved multi-objective evolutionary algorithm (MOEA) to find a Pareto optimal solution set. From these non-dominated solutions, the system selects the most suitable solution according to the current environmental state and management strategy, and transforms it into a specific cooperative control strategy. This method not only considers the efficiency and effectiveness of resource utilization, but also dynamically adjusts the control strategy according to real-time conditions, achieving intelligent and personalized urban lighting advertising management.

[0080] Step S5: Based on the collaborative control strategy, control the lighting equipment and advertising resources; simultaneously collect the operating status data and display effect data of both the lighting equipment and the advertising resources, and use the operating status data and display effect data as optimization inputs to feed back to the multi-source sensing network to realize smart city lighting advertising management.

[0081] A command parsing engine translates high-level control policies into device-level control commands. This engine comprises three modules: policy parsing, parameter mapping, and command generation. It accurately translates abstract control policies into a set of executable commands for the device. These commands are then reliably distributed to various terminal devices via a distributed message queue system.

[0082] Upon receiving control commands, the intelligent LED lighting system adjusts the brightness, color temperature, and illumination angle of the lamps to achieve precise control of the lighting environment. The digital advertising screen's content management system then adjusts parameters such as the display schedule, brightness, and animation effects of the advertising content. A collaborative mechanism is established between the two systems to ensure that the lighting effects and advertising displays work together to create the best visual experience.

[0083] Meanwhile, the system continuously collects energy consumption data from lighting equipment and display effect data from advertising resources through its built-in monitoring module. This data, after initial processing at edge computing nodes, is transmitted to the cloud platform for in-depth analysis. The analysis results are used to evaluate the effectiveness of control strategies, identify potential problems, and serve as optimization inputs fed back to the system's data acquisition module, forming a complete closed-loop optimization mechanism. Through this adaptive execution and feedback mechanism, the system can continuously learn and improve, enhancing the intelligence level and service quality of urban lighting advertising management.

[0084] In one embodiment of the present invention, step S1 involves collecting state data of urban lighting equipment, advertising resources, and environmental scenes through a multi-source sensing network, and preprocessing and fusing the state data using edge computing nodes to obtain a standardized data stream, including:

[0085] Step S1.1: Collect status data of the lighting equipment, the advertising resources, and the environmental scene, and transmit the status data to the edge computing node through a preset data acquisition protocol; wherein the status data includes power parameters, brightness parameters, advertising screen size parameters, resolution parameters, and environmental sensor operating data;

[0086] Step S1.2: Based on the state data, perform data preprocessing and spatiotemporal synchronization at the edge computing node to generate a data stream;

[0087] Step S1.3: Perform data cleaning and feature extraction on the data stream to obtain the standardized data stream.

[0088] In step S1.1, the multi-source sensing network is deployed on the city's existing infrastructure and includes environmental sensors, lighting equipment sensors, pedestrian flow detectors, and advertising screen status monitors to collect real-time data related to urban lighting and advertising displays. These sensors are connected to pre-set edge computing nodes via wired or wireless means, and transmit data to the edge nodes for preliminary processing using standard IoT communication protocols (such as MQTT, CoAP, etc.).

[0089] In step S1.2, edge computing nodes are deployed in key areas of the city, responsible for preprocessing and spatiotemporally synchronizing raw data from different sensors. Preprocessing includes data format conversion, outlier detection, and preliminary data fusion. Spatiotemporal synchronization ensures the consistency of data from different sensors in both time and space, providing a foundation for subsequent analysis.

[0090] In step S1.3, the data stream processed by edge computing is transmitted to the cloud platform data center, where further data cleaning (such as noise removal and handling of missing values) and feature extraction (such as temporal features, spatial features, and device features) are performed. Through this series of processes, the original heterogeneous data is transformed into a standardized data stream, providing a unified data foundation for subsequent analysis.

[0091] In one embodiment of the present invention, such as Figure 2 As shown, in step S2, locality-sensitive hashing is used to bucket lighting devices and advertising resources, and iterative local search is used for optimization to obtain resource grouping results, including:

[0092] Step S2.1: Based on the standardized data stream, construct a feature vector using dimensionality reduction and feature extraction techniques;

[0093] In step S2.1, after obtaining the standardized data stream from step S1, the system first constructs feature vectors. This process aims to convert high-dimensional heterogeneous data into low-dimensional feature representations, reducing computation while preserving key information. Specifically, a two-stage feature processing method is used: the first stage performs feature selection and normalization, and the second stage applies a deep autoencoder for nonlinear feature extraction.

[0094] In the feature selection and normalization stage, the information content and correlation of data in each dimension are first analyzed. Principal Component Analysis (PCA) is used to identify the principal components that contribute most to data variation; Non-negative Matrix Factorization (NMF) is used to extract key patterns from the data. These techniques help the system select the most representative feature dimensions from the original high-dimensional data. Simultaneously, the system normalizes data of different dimensions, such as mapping different types of data like location coordinates, power parameters, and brightness values ​​to a unified numerical range, facilitating subsequent processing.

[0095] In the deep autoencoder stage, a multi-layer encoder-decoder network structure was constructed. The encoder part consists of multiple dimensionality reduction layers, progressively compressing the input data into low-dimensional latent representations; the decoder part attempts to reconstruct the original input from these latent representations. By minimizing the reconstruction error, the network learns a non-linear representation of the data. To maintain the similarity of geographically proximate devices in the feature space, a spatial sensitivity constraint was added to the loss function of the autoencoder: L = L_reconstruction + λ·L_space, where L_space measures the consistency between the feature space distance and the actual geographical distance.

[0096] This feature vector construction method not only effectively reduces data dimensionality and subsequent computational load, but also preserves key information and spatial relationships within the data, providing high-quality input for locality-sensitive hashing and clustering analysis. Ultimately, 32-128 dimensional feature vectors were generated for each lighting device and advertising resource. These vectors capture both the physical attributes and functional characteristics of the devices while preserving spatial distribution information.

[0097] Step S2.2: Based on the feature vector, perform initial bucketing of lighting devices and advertising resources using a locality-sensitive hash function;

[0098] In step S2.2, based on the feature vector constructed in step S2.1, the system uses an improved Locality Sensitive Hash (LSH) algorithm to initially bucket the lighting equipment and advertising resources. The core idea of ​​the LSH algorithm is to construct a set of hash functions such that the probability of points that are close in the feature space being mapped to the same "bucket" is much higher than that of points that are far apart, thereby achieving fast coarse classification of the data.

[0099] To address the uneven distribution of urban spatial data, this invention designs an adaptive hash function family. The basic hash function employs a random projection method: h(v) = floor((v·r+b) / w), where v is the feature vector, r is a random vector, b is a random offset, and w is the quantization width. The system dynamically adjusts the parameter w by observing the data distribution characteristics, ensuring that w is larger in sparse data regions (increasing sensitivity) and smaller in dense data regions (reducing hash collisions). This adaptive adjustment allows the hash function to better adapt to the uneven distribution of urban data.

[0100] To further improve hash quality, the system employs a multi-level LSH strategy, which involves constructing L hash tables containing k hash functions. By adjusting the values ​​of L and k, the system can achieve a balance between recall and precision. In practice, the system first uses a smaller k value and a larger L value for coarse-grained binning, and then uses a larger k value and a smaller L value for fine-grained binning within each bin, forming a hierarchical grouping structure.

[0101] In addition, an automatic threshold adjustment mechanism based on local density is introduced. By calculating the distribution density of points in the feature space, the system automatically adjusts the parameters of the hash function, avoiding the performance degradation problem of traditional LSH when dealing with unevenly distributed data. This adaptive mechanism enables the system to obtain better initial binning results under different urban areas and different densities of device distributions.

[0102] Through this improved LSH algorithm, preliminary data binning can be completed within the time complexity of O(n). The output initial grouping results retain the local proximity relationship of the data, providing a good starting point for subsequent iterative local search optimization. This preliminary grouping can transform the O(n2) complexity problem that may originally require global comparison into a nearly linear time problem that only needs to be optimized within a local range.

[0103] Step S2.3: Perform iterative local search optimization on the initial binning results to obtain the resource grouping results.

[0104] In step S2.3, the system performs iterative local search optimization on the initial binning results of step S2.2 to further improve the clustering quality. Traditional clustering optimization usually adopts a global search strategy with high computational complexity; while simple greedy optimization is prone to falling into local optimal solutions. The present invention designs an iterative local search algorithm with a time complexity of O(n log n), effectively controlling the computational complexity while ensuring the clustering quality.

[0105] The core of iterative local search is the "perturbation - improvement" loop strategy. In the improvement stage, the system considers moving each data point from the current cluster to a neighboring cluster. If this move can improve the clustering objective function value, the move is accepted. The clustering objective function comprehensively considers multiple factors such as intra - cluster similarity, spatial continuity, energy load balance, and advertising coverage efficiency. To avoid falling into local optimality, in the perturbation stage, the system randomly selects a certain proportion of points (usually 5 - 10%) for re - allocation, creating opportunities to jump out of local optima.

[0106] The near - linear time complexity of the algorithm is mainly achieved through three technological innovations: Step one, construct a proximity graph using the LSH results of step S2.2, restricting the search range of each point within its K - nearest neighbors (K << n), significantly reducing the number of comparison operations; Step two, adopt incremental calculation technology, and only update the affected part of the objective function each time a point moves, avoiding global recalculation; Step three, introduce an early - stopping strategy, and terminate the iteration in advance when the improvement in consecutive rounds is lower than a preset threshold.

[0107] To verify the effectiveness of the clustering results, a multi-index adaptive clustering verification method was employed. This method constructs a comprehensive evaluation index system, including internal indicators (such as the silhouette coefficient, Davies-Bouldin index, and Calinski-Harabasz index) and external indicators (such as energy load balance and advertising coverage uniformity). Simultaneously, the system designs a Bayesian optimization-based parameter adaptive adjustment mechanism, capable of automatically searching for the optimal combination of clustering parameters. Furthermore, the system introduces a domain-knowledge-based constraint verification mechanism to ensure that the clustering results meet practical engineering needs such as urban planning area division.

[0108] This multi-dimensional, adaptive clustering validation and optimization method ultimately outputs fully validated and optimized resource grouping results, providing a high-quality grouping foundation for subsequent construction of random regular graph models.

[0109] In step S2.1, feature vectors are constructed from the standardized data stream from step S1. This process employs a two-stage feature processing method: first, feature selection and normalization are performed using Principal Component Analysis (PCA) and Non-negative Matrix Factorization (NMF) techniques to identify and retain the most informative feature dimensions; then, a deep autoencoder network is applied to compress the high-dimensional input into a low-dimensional latent representation. To ensure that geographically proximate devices also exhibit similarity in the feature space, a location-based regularization term is added to the autoencoder's loss function. The final output is a low-dimensional feature vector (typically 32-128 dimensions) for each lighting device and advertising resource.

[0110] In step S2.2, based on the feature vector constructed in step S2.1, an improved Locality Sensitive Hash (LSH) algorithm is designed for initial data bucketing. The core of the improved LSH is to construct a set of hash functions such that the probability of nearby points in the feature space being mapped to the same "bucket" is much higher than that of points that are far apart. To address the uneven distribution of urban spatial data, an adaptive family of hash functions is designed. By dynamically adjusting the parameter w, sensitivity is increased in sparse data areas, and hash collisions are reduced in dense data areas. Simultaneously, a multi-level LSH strategy is adopted to construct L hash tables containing k hash functions. By adjusting the values ​​of L and k, a balance is achieved between recall and precision.

[0111] In step S2.3, based on the initial bucketing results from step S2.2, the initial bucketing results are iteratively optimized using local search to obtain the final resource grouping results.

[0112] In one embodiment of the present invention, step S2.3 involves iterative local search optimization of the initial bucketing results to obtain the resource grouping results, including:

[0113] Step S2.3.1: Using each bucket of the initial bucketing result as an initial cluster, construct a clustering objective function; wherein, the clustering objective function includes intra-cluster similarity, spatial continuity, energy load balancing, and advertising coverage efficiency;

[0114] Step S2.3.2: Based on the clustering objective function, perform a local search on the data points within each initial cluster, and update the clustering objective function using incremental computation techniques;

[0115] Step S2.3.3: When it is determined that the clustering objective function has converged, the resource grouping result is obtained.

[0116] In step S2.3.1, based on the initial binning results, a clustering objective function suitable for urban lighting and advertising resources is constructed. This function comprehensively considers multiple factors such as intra-cluster similarity, spatial continuity, energy load balance, and advertising coverage efficiency. Formalized as: F(C)=α·F

[0117] (C)+β·F(C)+γ·F(C)+δ·F(C), where C represents the clustering result, F1 to F4 represent the sub-objective functions respectively, and α, β, γ and δ are weight coefficients.

[0118] In step S2.3.2, an iterative local search algorithm is used to optimize the clustering. The core of this algorithm is the "perturbation-improvement" cyclical strategy: in the improvement phase, each data point is considered for movement from its current cluster to a neighboring cluster; if this movement improves the objective function value, the movement is accepted. In the perturbation phase, a certain proportion of points are randomly selected for redistribution, creating opportunities to escape local optima. To achieve near-linear time complexity, three technical innovations are employed: a proximity graph is constructed using LSH results, limiting the search range of each point to its K nearest neighbors; incremental computation is used, updating only the affected objective function portion with each point movement; and an early stopping strategy is introduced, terminating the iteration prematurely when the improvement falls below a preset threshold for multiple consecutive rounds.

[0119] In step S2.3.3, when the change value of the clustering objective function is lower than the preset threshold multiple times consecutively, the algorithm is considered to have converged, and the final resource grouping result is output. To verify the effectiveness of the clustering results, a multi-index adaptive clustering verification method is also adopted, including internal indices such as the silhouette coefficient, Davies-Bouldin index, and Calinski-Harabasz index, as well as external indices such as energy load balance and advertising coverage uniformity.

[0120] In one embodiment of the present invention, step S3, constructing a random regular graph model, includes:

[0121] Step S3.1: Based on the resource grouping results and combined with the state data of the environmental scene, construct a heterogeneous graph structure of lighting device nodes, advertising resource nodes, and environmental scene nodes;

[0122] Step S3.2: Perform degree distribution regularization and edge weight randomization on the heterogeneous graph structure to obtain the random regular graph model.

[0123] In step S3.1, a multi-layered heterogeneous graph structure is constructed based on the resource grouping results output in step S2. This graph structure consists of three basic types of nodes: lighting device nodes, advertising resource nodes, and environmental scene nodes. Each type of node has a different set of attributes. For example, lighting device nodes include attributes such as location, power, and illumination radius; advertising resource nodes include attributes such as size, resolution, and viewing angle; and environmental scene nodes include attributes such as regional features and pedestrian density. Edges between nodes represent potential relationships, including physical proximity, functional complementarity, and visual impact. To improve the computational efficiency of the graph structure, sparse matrix representation and spatial index optimization techniques are used to reduce the edge construction complexity from O(n²) to O(nlog n).

[0124] In step S3.2, the initial heterogeneous graph structure undergoes random regularization. Unlike traditional regular graphs that require all nodes to have the same degree, this model allows node degrees to vary randomly within a certain range while maintaining the overall degree distribution in accordance with preset statistical characteristics. Specifically, a three-stage random regularization process is employed: the first stage performs degree distribution regularization, setting a probability model for the degree distribution based on the functional characteristics of different types of nodes, and assigning a target degree to each node through Monte Carlo sampling; the second stage randomizes edge weights, introducing appropriate random perturbations to each edge in the graph based on the type of relationship it represents; the third stage performs global structural balancing, applying a global optimization algorithm to restore the key structural characteristics of the graph.

[0125] In one embodiment of the present invention, step S3 involves optimizing the graph structure using a Markov chain Monte Carlo algorithm, including:

[0126] Step S3.3: Based on the stochastic regular graph model, construct an energy function that includes resource allocation balance and control system responsiveness;

[0127] Step S3.4: The random regular graph model is iteratively optimized using the block-parallel Markov chain Monte Carlo algorithm to obtain the optimized graph structure;

[0128] Step S3.5: Based on the optimized graph structure and the energy function, extract the mapping relationship between device nodes and resource nodes to generate the resource configuration strategy.

[0129] In step S3.3, an energy function E(G) is defined to evaluate the quality of the graph structure G. This energy function comprehensively considers multiple aspects such as resource allocation balance, control system responsiveness, visual coordination, and energy efficiency: E(G) = w·E(G) + w·E(G) + w·E(G) + w·E(G), where E1 to E4 correspond to sub-energy functions for different optimization objectives, and w1 to w are weighting coefficients.

[0130] In step S3.4, based on the energy function defined in step S3.3, the Markov Chain Monte Carlo (MCMC) algorithm, which features fast mixing, is used to optimize the graph structure. To address the challenges of large-scale graph structure optimization, three technical innovations are introduced: a block-parallel MCMC strategy is adopted, decomposing the large graph into multiple overlapping subgraphs and performing MCMC iterations in parallel on each subgraph; an adaptive proposal distribution is designed, dynamically adjusting the proposal step size based on historical acceptance rates; and a temperature scheduling mechanism is introduced, using high temperature in the initial stage to promote extensive exploration, and gradually cooling down in the later stage to achieve fine-grained optimization. The key to the fast mixing feature lies in the design of the nonlocal proposal mechanism, which combines local proposals (such as single-edge flipping and node attribute fine-tuning) and nonlocal proposals (such as subgraph reconstruction, pattern substitution, and cross-community exchange), enabling the algorithm to achieve "long-distance jumps" in the state space in one step, significantly improving the mixing speed.

[0131] In step S3.5, based on the optimized graph structure, the optimal mapping relationship between lighting equipment and advertising resources is extracted. This process employs a multi-level resource mapping relationship extraction method: first, based on graph community structure analysis, closely related equipment-resource clusters are identified; then, a bipartite graph matching algorithm considering node type constraints is applied to find the optimal mapping relationship; finally, constraint satisfaction checks are performed to ensure the feasibility of resource mapping in actual implementation. To cope with dynamic environmental changes, a time window-based mapping relationship update mechanism and an event-triggered update mechanism are also designed.

[0132] In one embodiment of the present invention, such as Figure 3 As shown, in step S4, a cooperative control strategy is generated using a multi-objective optimization algorithm, including:

[0133] Step S4.1: Based on the resource allocation strategy, construct a multi-objective function that includes energy efficiency and visual comfort;

[0134] Based on the resource allocation strategy output in step S3, a multi-objective function for collaborative optimization is constructed. These objective functions are not simple, single evaluation indicators, but rather a set of potentially conflicting evaluation dimensions, comprehensively reflecting the complexity of smart city lighting advertising management. The system defines four core objective functions: energy efficiency objective function, visual comfort objective function, advertising display effect objective function, and environmental impact objective function.

[0135] The energy efficiency objective function evaluates the ratio of system energy consumption to service effectiveness. This function considers factors such as the power parameters of lighting equipment and advertising screens, operating hours, and service coverage, calculating energy utilization efficiency through a nonlinear model. To improve accuracy, the system also incorporates a time factor, setting different weighting coefficients for different time periods (e.g., peak, off-peak, and low-peak periods). The visual comfort objective function, based on a human visual perception model, evaluates the impact of the combined visual effects of lighting and advertising on human comfort. This function integrates sub-indicators such as color temperature consistency, brightness uniformity, contrast suitability, and visual interference, quantifying these through a neural network model trained on perceptual experimental data.

[0136] The objective function for advertising display effectiveness evaluates the visibility, attractiveness, and information delivery efficiency of advertising content. This function combines the location characteristics of the advertising screen, surrounding environmental conditions, and historical viewing data to construct a predictive model for advertising effectiveness. It also considers the interaction between the advertising content and the surrounding lighting environment, such as the impact of lighting conditions on color perception. The environmental impact objective function assesses the system's impact on the ecological environment, primarily focusing on light pollution, energy consumption, and visual disturbance. This function sets evaluation indicators based on environmental standards and urban planning requirements to ensure that the system's operation does not place an excessive burden on the urban ecological environment.

[0137] These objective functions are integrated into a unified multi-objective optimization problem through weighted coefficients: F = {f1(x), f2(x), f3(x), f4(x)}, where x represents a vector of decision variables, including lighting parameters and advertising display parameters. By constructing this multi-objective function, comprehensive evaluation and optimization can be performed across multiple dimensions, including efficiency, comfort, advertising effectiveness, and environmental protection.

[0138] Step S4.2: Optimize the multi-objective function to obtain the Pareto optimal solution for resource allocation;

[0139] The multi-objective function is optimized to find a Pareto optimal solution set for resource allocation. Due to inherent conflicts between objective functions (e.g., increasing advertising brightness may increase energy consumption and light pollution), traditional single-objective optimization methods struggle to find solutions that simultaneously optimize all objectives. The concept of Pareto optimality provides a framework for handling such problems: finding a set of solutions where improvement in any objective necessarily leads to the deterioration of at least one other objective.

[0140] This complex problem is solved using an improved multi-objective evolutionary algorithm (MOEA). Specifically, NSGA-III (Non-dominated sorting genetic algorithm III) is chosen as the base algorithm. This algorithm effectively maintains the diversity of the solution set in the high-dimensional objective space through non-dominated sorting and reference point-assisted selection mechanisms. To adapt to the characteristics of smart city scenarios, three improvements are made to the standard NSGA-III: First, an initial population generation strategy based on problem knowledge is designed to distribute initial solutions in more promising search space regions; second, adaptive crossover and mutation operators are developed to dynamically adjust parameters according to population diversity and convergence status; finally, a local search strategy is introduced to finely optimize the leading solutions and improve the quality of the solutions.

[0141] To address the computational challenges of city-scale optimization problems, a distributed computing architecture and GPU acceleration technology were employed. A region decomposition strategy was used to break down the large-scale optimization problem into multiple sub-problems for parallel solving, followed by a boundary coordination mechanism to integrate the results. This divide-and-conquer approach enables the system to complete city-scale multi-objective optimization solutions within an acceptable timeframe (typically on the order of minutes).

[0142] The algorithm ultimately outputs a set of Pareto optimal solutions, each representing a resource allocation scheme with different trade-offs among the objective functions. These solutions form a Pareto front in the objective space, providing decision-makers with a diverse range of choices, allowing them to select the most suitable allocation scheme based on specific management needs and preferences.

[0143] Step S4.3: Generate the cooperative control strategy based on the Pareto optimal solution.

[0144] Based on the Pareto optimal solution set obtained in step S4.2, and combined with the actual operating environment and management strategy, an executable collaborative control strategy is generated. This process does not simply select a Pareto optimal solution, but rather generates specific and feasible control instructions through a complex decision-making mechanism that considers the current system state, management preferences, and execution constraints.

[0145] First, a solution set selection mechanism is applied to choose the solution most suitable for the current situation from the Pareto front. This selection process combines multi-criteria decision-making methods and context-aware techniques. Multi-criteria decision-making methods, such as TOPSIS (Topology for Ranking Approximate Ideal Solutions) and AHP (Analytic Hierarchy Process), help the system weigh multiple objectives; context-aware techniques consider time factors (such as date and time of day), spatial factors (such as regional characteristics), and activity factors (such as special events and holidays). For example, during peak hours in a commercial area, the system may prioritize advertising display effectiveness; while in a residential area at night, it may prioritize visual comfort and minimizing light pollution.

[0146] Next, decision variable mapping is performed, transforming abstract optimization variables into specific control parameters. This process involves device characteristic mapping (considering the performance parameters and control interfaces of different devices), environmental adaptation mapping (adjusting control parameters according to real-time environmental conditions), and time-series planning mapping (an execution plan for design parameters that change over time). For example, the system might map the abstract "50% lighting intensity" to the specific power value and PWM dimming parameters of a particular LED luminaire.

[0147] Finally, a control instruction set is generated, formatting the control parameters into executable instructions for the devices, and adding necessary execution timing and safety check logic. The control instruction set not only includes basic operating instructions for the lighting equipment and advertising screens but also includes collaborative control logic to ensure that the two types of equipment operate in coordination, creating a harmonious and unified urban visual experience. For example, when advertising content changes, the system may synchronously adjust the brightness and color temperature of the surrounding lighting to enhance the advertising effect; or under specific environmental conditions, it may dynamically adjust the parameters of the lighting and advertising screens to adapt to the constantly changing environment.

[0148] Through this sophisticated collaborative control strategy generation mechanism, the system can transform theoretical optimization results into practically executable control schemes, enabling intelligent collaborative management of lighting equipment and advertising resources.

[0149] In step S4.1, based on the resource allocation strategy output in step S3, a multi-objective function for collaborative optimization is constructed. This multi-objective function comprehensively considers multiple dimensions such as energy efficiency, visual comfort, advertising display effect, and environmental impact. The energy efficiency objective function evaluates the ratio of system energy consumption to service effect; the visual comfort objective function, based on a human visual perception model, evaluates the impact of the combined visual effect of lighting and advertising on human comfort; the advertising display effect objective function evaluates the visibility, attractiveness, and information delivery efficiency of advertising content; and the environmental impact objective function evaluates the degree of impact of the system on the ecological environment (such as light pollution).

[0150] In step S4.2, the multi-objective function is optimized to find the Pareto optimal solution set. Since conflicts may exist between objective functions, traditional single-objective optimization methods are not applicable. This scheme employs an improved multi-objective evolutionary algorithm (MOEA), such as NSGA-III (Non-dominated sorting genetic algorithm III), to find a balance among multiple objectives by maintaining a diverse solution set. A resource allocation constraint handling mechanism is introduced into the algorithm design to ensure that the generated solution meets the physical constraints of the actual system. Simultaneously, to improve algorithm efficiency, an adaptive mutation strategy and parallel computing architecture are adopted, keeping the solution time for large-scale problems within an acceptable range.

[0151] In step S4.3, based on the Pareto optimal solution set obtained in step S4.2, and combined with the actual operating environment and management strategy, an executable collaborative control strategy is generated. This process includes three steps: solution set selection, decision variable mapping, and control instruction generation. Solution set selection selects the most suitable solution from the Pareto front based on the current system state and decision preferences; decision variable mapping converts abstract optimization variables into specific control parameters; and control instruction generation formats the control parameters into a set of instructions executable by the device, and adds necessary execution timing and safety check logic.

[0152] In one embodiment of the present invention, step S4, combining user behavior data and the state data of the environmental scene, includes:

[0153] Step S4.4: Collect ambient light intensity, weather conditions, and pedestrian density data to obtain the state data of the environmental scene;

[0154] Step S4.5: Obtain user dwell time and attention heatmap based on video analysis technology to obtain the user behavior data;

[0155] Step S4.6: Perform spatiotemporal correlation analysis on the state data of the environmental scene and the user behavior data to obtain the scene feature vector.

[0156] Spatiotemporal correlation analysis is performed on environmental scenario state data and user behavior data to explore the complex relationship between environmental factors and user behavior. This analysis reveals the changing patterns of user behavior under different environmental conditions, helping the system build predictive models and proactively adapt to environmental changes. The spatiotemporal correlation analysis employs various data mining techniques, including spatiotemporal autocorrelation analysis, conditional random field models, and deep learning methods.

[0157] First, data preprocessing and alignment are performed to ensure consistency across time and space for data from different sources. This includes time synchronization (adjusting data from different sampling frequencies to a unified time axis), spatial registration (mapping location information from different coordinate systems to a unified reference system), and data interpolation (filling in missing values ​​and smoothing outliers). Then, the system applies spatiotemporal autocorrelation analysis methods, such as Moran's I and LISA statistics, to detect spatiotemporal clustering patterns and hotspots in the data. These analyses reveal the temporal and spatial distribution patterns of environmental variables and user behavior, such as the spatial correlation between pedestrian traffic and lighting brightness, or the temporal correlation between advertising attention and weather conditions.

[0158] Next, an environment-behavior relationship model is constructed to reveal the mechanism by which environmental factors influence user behavior. The system uses a Conditional Random Field (CRF) model to capture local dependencies, a Long Short-Term Memory (LSTM) network to model temporal patterns, and a Graph Neural Network (GNN) to express spatial interactions. These models can answer complex questions such as "how changes in lighting brightness affect user dwell time" and "how advertising content preferences are adjusted in rainy weather."

[0159] Finally, scene feature vectors are generated as important inputs for collaborative optimization decisions. Scene feature vectors are compact representations of key features of environmental states and user behavior, typically composed of 100-200 dimensional numerical vectors. The vector design considers the information content, complementarity, and stability of the features, effectively supporting subsequent decision optimization. Through this in-depth spatiotemporal correlation analysis, the system establishes a bridge between the environment, users, and control strategies, enabling the decision-making process to consider both objective environmental conditions and the user's subjective experience.

[0160] In step S4.4, environmental scene status data such as ambient light intensity, weather conditions, temperature and humidity, and pedestrian density are collected in real time through an environmental sensor network. This data, after standardization and spatiotemporal synchronization, is used to characterize the real-time state of the current urban environment. Environmental status data directly impacts lighting control and advertising display effects; for example, under different weather and lighting conditions, it is necessary to adjust lighting brightness and the contrast of advertising content to ensure optimal results.

[0161] In step S4.5, using cameras and sensors deployed near the advertising screen and employing advanced video analytics, behavioral data such as user dwell time, gaze direction, and attention distribution in front of the advertising screen are acquired. This data is processed using privacy-preserving algorithms to remove personally identifiable information, retaining only statistical characteristics. Deep learning-based visual algorithms can generate attention heatmaps, visually displaying which areas and content are more likely to attract user attention.

[0162] In step S4.6, spatiotemporal correlation analysis is performed on environmental scene state data and user behavior data to explore the relationship between environmental factors and user behavior. This analysis employs spatiotemporal data mining techniques, such as spatiotemporal autocorrelation analysis and conditional random field models, to identify key environmental factors and spatiotemporal patterns affecting user behavior. The analysis results form a scene feature vector, which serves as an important input for subsequent collaborative optimization decisions, enabling the system to dynamically adjust lighting and advertising control strategies based on environmental changes and user responses.

[0163] In one embodiment of the present invention, step S5, based on the collaborative control strategy, controls the lighting equipment and advertising resources, including:

[0164] Step S5.1: Convert the collaborative control strategy into the brightness parameters of the lighting device and the display parameters of the advertising resources;

[0165] The instruction parsing engine converts the high-level collaborative control strategy output in step S4 into specific device-level control instructions. This process is akin to translating an abstract conductor's intention into concrete playing instructions for each instrument in an orchestra, ensuring that all devices execute the overall strategy in a coordinated manner. The instruction parsing engine consists of three core modules: a strategy parsing module, a parameter mapping module, and an instruction generation module.

[0166] The strategy parsing module first understands the semantics and intent of the high-level control strategy. Control strategies are typically expressed as a target state (e.g., "Provide a comfortable and bright environment in the main commercial street area while highlighting holiday advertisements") or a transition command (e.g., "Gradually increase lighting brightness as sunset, while adjusting the color saturation of advertisements"). The strategy parsing module uses semantic analysis techniques to decompose these expressions into specific control objectives, constraints, and priority rules. Then, the parameter mapping module converts the parsed control objectives into operating parameters for specific devices. This process considers differences in device characteristics, the influence of environmental conditions, and synergistic effect requirements.

[0167] For example, the abstract goal of "comfortable lighting" is mapped to a combination of brightness percentage, color temperature, and illumination angle of LED lights in a specific area; the advertising requirement of "highlighting" is converted into parameters such as brightness, contrast, and content refresh rate of the display screen. During the parameter mapping process, the system uses a device characteristic database to store the performance parameters, control range, and response curves of each device model, ensuring that the generated parameter values ​​are suitable for the technical capabilities of the specific device.

[0168] Finally, the instruction generation module formats the mapped parameters into a set of instructions executable by the device. These instructions conform to device communication protocol specifications (such as the DALI lighting control protocol, DMX512 digital multiplexing protocol, etc.) and include necessary address information, command codes, and parameter values. The system also adds timestamps, execution order, and security check logic to the instructions to ensure the reliability and security of the control process. The generated instructions are reliably distributed through a distributed message queue system (such as Apache Kafka or RabbitMQ), ensuring that instructions are correctly delivered to the target device even in the event of network fluctuations or partial node failures.

[0169] Step S5.2: Adjust the brightness of the lighting equipment and the display content of the advertising resources through the Internet of Things control system;

[0170] The IoT control system receives and distributes control commands, precisely controlling lighting equipment and advertising resources. This process involves multi-layered control logic and coordination mechanisms to ensure the system can create a unified and harmonious urban visual experience as intended. The lighting control system employs environmentally aware adaptive lighting technology, dynamically adjusting the brightness, color temperature, and illumination angle of LED lights based on real-time environmental conditions and control strategies.

[0171] It supports a variety of advanced lighting control functions, such as gradual transition (smoothly adjusting lighting parameters to avoid visual discomfort caused by sudden changes), scene presets (quickly switching lighting schemes according to different scene requirements), intelligent grouping (logically grouping lamps to achieve precise regional control), and timed scheduling (executing lighting changes according to preset plans). In terms of energy efficiency, the system automatically adjusts the lighting intensity according to ambient light conditions, such as reducing artificial lighting output when natural light is sufficient and enhancing lighting effects on cloudy days or at night, achieving rational use of energy.

[0172] The advertising content management system adjusts parameters such as display schedule, brightness, contrast, and animation effects of advertising content according to control instructions. The system supports intelligent content scheduling, such as selecting the most suitable advertising content based on time of day, weather, and audience characteristics; it supports dynamic rendering, enabling real-time adjustments to the visual effects of advertisements to adapt to environmental changes; and it supports interactive displays, adjusting content presentation methods or triggering specific interactive modules based on audience feedback. The advertising system also has content adaptation capabilities, automatically optimizing content layout and display effects based on the size, resolution, and color characteristics of different display devices.

[0173] A two-way communication and coordination mechanism has been established between the lighting system and the advertising system to ensure that they work together to create the best visual experience. For example, when the advertising screen displays bright content, the ambient lighting may be appropriately reduced in intensity to enhance the visual impact of the advertisement; when the advertising content changes, the lighting tone may change accordingly to enhance the atmosphere; in special events or emergencies, the two systems can respond synchronously to jointly convey important information. This coordinated control not only improves the advertising effect and lighting experience but also optimizes overall energy use and reduces visual interference.

[0174] Step S5.3: Collect the energy consumption data of the lighting equipment and the display effect data of the advertising resources.

[0175] The system continuously collects energy consumption data from lighting equipment and display effect data from advertising resources through its built-in monitoring module. This real-time feedback data is crucial for the system's self-evaluation and optimization, forming a complete closed loop from control to monitoring and optimization. Energy consumption data is collected through smart meters and the equipment's built-in power monitoring unit, recording parameters such as voltage, current, power factor, and cumulative power consumption of the lighting equipment and advertising screens. The system processes this raw data to calculate energy efficiency indicators (such as lighting power density per square meter and energy consumption per viewer) and detect anomalies (such as sudden changes in equipment power consumption and early warnings of reduced energy efficiency).

[0176] Advertising display performance data is collected through multiple channels, including display device operating parameters (actual brightness, contrast ratio, refresh rate), environmental influencing factors (ambient light interference, viewing angle obstruction), and audience response indicators (number of viewers, dwell time, interaction rate). The system pays particular attention to the difference between expected and actual results, such as the deviation between planned and actual brightness, and the ratio of estimated to actual viewers. This discrepancy data helps identify inaccurate assumptions in the system model or unexpected changes in the environment.

[0177] The collected data undergoes initial processing at edge computing nodes, including data compression (reducing transmission load), anomaly detection (marking potential faults or errors), and feature extraction (calculating key performance indicators). The processed data is then transmitted to the cloud platform via a secure channel for more in-depth analysis, such as trend analysis (identifying long-term patterns of change), correlation analysis (discovering dependencies between factors), and comparative analysis (comparing with historical data or similar facilities).

[0178] The analysis results are used in multiple aspects: performance evaluation (assessing the effectiveness of the current control strategy), problem diagnosis (identifying bottlenecks or defects in the system), model calibration (updating the parameters of the predictive model), and strategy optimization (adjusting control rules and parameters). This feedback data ultimately flows back to the system's data acquisition module as input for the next round of decision optimization, forming a complete closed-loop optimization mechanism. Through this continuous data acquisition and feedback analysis, the system can continuously learn and improve, adapt to environmental changes, improve resource utilization efficiency, and achieve truly intelligent management.

[0179] In step S5.1, the high-level collaborative control strategy output in step S4 is converted into device-level control commands through an instruction parsing engine. This engine includes a strategy parsing module, a parameter mapping module, and an instruction generation module. The strategy parsing module understands the semantics of the high-level strategy; the parameter mapping module maps abstract control parameters to the operating parameters of specific devices, such as converting lighting effect requirements into the brightness, color temperature, and power parameters of specific LED lights; the instruction generation module generates standardized control commands according to the device communication protocol requirements. The generated commands are reliably distributed through a distributed message queue system (such as Kafka or RabbitMQ) to ensure that the control commands can be delivered to each device in a timely and accurate manner.

[0180] In step S5.2, the IoT control system receives control commands and precisely controls the lighting equipment and advertising resources. The lighting control system employs environmentally perceptive adaptive lighting technology, dynamically adjusting the brightness, color temperature, and illumination angle of the LED lights based on real-time environmental conditions and control strategies. The advertising content management system, in turn, adjusts parameters such as the display plan, brightness, contrast, and animation effects of the advertising content according to the control commands. A collaborative mechanism is established between the two systems to ensure that the lighting effects and advertising displays work together to create the best visual experience.

[0181] In step S5.3, the built-in monitoring module continuously collects energy consumption data (such as power, current, and operating time) from lighting equipment and display effect data (such as actual brightness, content switching frequency, and number of viewers) from advertising resources. This data undergoes preliminary processing and aggregation via edge computing nodes, and is then transmitted to the cloud platform for in-depth analysis. The analysis results are used to evaluate the effectiveness of the control strategy, identify potential problems, and provide feedback for strategy optimization. Through this closed-loop feedback mechanism, continuous learning and improvement are possible, enhancing resource utilization efficiency and service quality.

[0182] like Figure 4 As shown, this embodiment of the invention also provides a smart city lighting advertising management device, comprising:

[0183] The data acquisition module 10 is used to collect status data of urban lighting equipment, advertising resources and environmental scenes through a multi-source sensing network, and to preprocess and fuse the status data using edge computing nodes to obtain a standardized data stream.

[0184] Resource grouping module 20 is used to bucket lighting devices and advertising resources based on the standardized data stream using locality-sensitive hashing, and to optimize the resource grouping results through iterative local search.

[0185] The resource mapping module 30 is used to construct a random regular graph model based on the resource grouping results, optimize the graph structure through the Markov chain Monte Carlo algorithm, extract the mapping relationship between lighting equipment and advertising resources, and generate a resource configuration strategy.

[0186] Control decision module 40 is used to generate a collaborative control strategy based on the resource allocation strategy, combined with user behavior data and the state data of the environmental scenario, through a multi-objective optimization algorithm.

[0187] The execution feedback module 50 is used to control the lighting equipment and advertising resources based on the collaborative control strategy; at the same time, it collects the operating status data and display effect data of both the lighting equipment and the advertising resources, and feeds the operating status data and the display effect data as optimization inputs to the multi-source sensing network to realize smart city lighting advertising management.

[0188] Preferably, the data acquisition module 10 includes a multi-source sensing submodule and an edge computing submodule, wherein the multi-source sensing submodule is used to collect status data of urban lighting equipment, advertising resources and environmental scenes, and the edge computing submodule is used to preprocess and fuse the status data to obtain a standardized data stream.

[0189] Preferably, the resource grouping module 20 includes a feature vector construction submodule, a locality-sensitive hashing submodule, and an iterative local search submodule. The feature vector construction submodule is used to construct feature vectors based on the standardized data stream, the locality-sensitive hashing submodule is used to perform initial bucketing based on the feature vectors, and the iterative local search submodule is used to optimize the initial bucketing results to obtain resource grouping results.

[0190] Preferably, the resource mapping module 30 includes a graph structure construction submodule, a random regularization submodule, and an MCMC optimization submodule. The graph structure construction submodule is used to construct a heterogeneous graph structure based on the resource grouping results. The random regularization submodule is used to perform random regularization processing on the heterogeneous graph structure. The MCMC optimization submodule is used to optimize the graph structure and extract the mapping relationship through the Markov chain Monte Carlo algorithm.

[0191] Preferably, the control decision module 40 includes a multi-objective function construction submodule, an environmental perception data integration submodule, a user behavior analysis submodule, and a multi-objective optimization submodule, used to generate a collaborative control strategy based on resource allocation strategies and environmental user data.

[0192] Preferably, the execution feedback module 50 includes a control instruction parsing submodule, a device control submodule, and a performance evaluation submodule, used to execute control strategies and collect feedback data to form a closed-loop optimization.

[0193] This invention constructs a four-layer architecture system to achieve the collection, processing, and fusion of multi-source heterogeneous data; it employs an improved method combining locality-sensitive hashing and iterative local search to achieve near-linear time clustering; it designs a stochastic regular graph model and a fast hybrid MCMC algorithm to achieve optimal mapping between lighting equipment and advertising resources; it develops a multi-objective optimization decision system to achieve coordinated control of lighting and advertising; and it designs an adaptive execution and feedback mechanism to form a closed-loop optimization, thereby improving the efficiency and effectiveness of smart city lighting and advertising management.

[0194] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A smart city lighting advertisement management method, characterized in that, The method comprises the following steps: Collecting state data of city lighting equipment, advertising resources and environmental scenes through a multi-source perception network, pre-processing and fusing the state data by an edge computing node to obtain standardized data flow; Based on the standardized data flow, using a local sensitive hash function to divide the lighting equipment and advertising resources into buckets, and optimizing through iterative local search to obtain resource grouping results, including: Based on the standardized data flow, using dimension reduction and feature extraction techniques to construct a feature vector; based on the feature vector, using a local sensitive hash function to perform initial bucketing on the lighting equipment and advertising resources; performing iterative local search optimization on the initial bucketing results to obtain the resource grouping results, wherein the iterative local search optimization on the initial bucketing results to obtain the resource grouping results comprises: Taking each bucket of the initial bucketing results as an initial clustering cluster, constructing a clustering objective function; wherein the clustering objective function includes intra-cluster similarity, spatial continuity, energy load balancing and advertising coverage efficiency; based on the clustering objective function, performing local search on the data points in each initial clustering cluster, and updating the clustering objective function through incremental calculation technology; when it is determined that the clustering objective function converges, the resource grouping result is obtained; Based on the resource grouping result, constructing a random regular graph model, optimizing the graph structure through a Markov chain Monte Carlo algorithm, extracting the mapping relationship between the lighting equipment and the advertising resources, and generating a resource configuration strategy; wherein the optimization of the graph structure through the Markov chain Monte Carlo algorithm comprises: based on the random regular graph model, constructing an energy function including resource configuration balance and control system responsiveness; using a block parallel Markov chain Monte Carlo algorithm to iteratively optimize the random regular graph model to obtain an optimized graph structure; based on the optimized graph structure, combining the energy function, extracting the mapping relationship between the device nodes and the resource nodes, and generating the resource configuration strategy; Based on the resource configuration strategy, combining user behavior data and state data of the environmental scene, generating a collaborative control strategy through a multi-objective optimization algorithm; Based on the collaborative control strategy, controlling the lighting equipment and advertising resources; simultaneously collecting the running state data and display effect data of the lighting equipment and the advertising resources, and feeding back the running state data and the display effect data as optimization input to the multi-source perception network to realize intelligent city lighting advertising management.

2. The method of claim 1, wherein, The method comprises the following steps: Collecting state data of city lighting equipment, advertising resources and environmental scenes through a multi-source perception network, pre-processing and fusing the state data by an edge computing node to obtain standardized data flow, including: Collecting state data of the lighting equipment, the advertising resources and the environmental scenes, and transmitting the state data to the edge computing node through a preset data collection protocol; wherein the state data includes power parameters, brightness parameters, size parameters of advertising screens, resolution parameters and working data of environmental sensors; Based on the state data, data preprocessing and space-time synchronization are performed at the edge computing node to generate a data stream; Data cleaning and feature extraction are performed on the data stream to obtain the standardized data stream.

3. The method of claim 1, wherein, The random regular graph model is constructed, including: Based on the resource grouping result, combined with the state data of the environment scene, a heterogeneous graph structure of lighting device nodes, advertising resource nodes and environment scene nodes is constructed; Degree distribution regularization and edge weight randomization processing are performed on the heterogeneous graph structure to obtain the random regular graph model.

4. The method of claim 1, wherein, The collaborative control strategy is generated through a multi-objective optimization algorithm, including: Based on the resource configuration strategy, a multi-objective function including energy efficiency and visual comfort is constructed; The multi-objective function is optimized and solved to obtain the Pareto optimal solution of resource configuration; Based on the Pareto optimal solution, the collaborative control strategy is generated.

5. The method of claim 1, wherein, The state data of the environment scene and the user behavior data are combined, including: Collecting ambient light intensity, weather conditions and crowd density data to obtain the state data of the environment scene; Based on video analysis technology, user stay time and attention heat map are obtained to obtain the user behavior data; Temporal and spatial correlation analysis is performed on the state data of the environment scene and the user behavior data to obtain scene feature vectors.

6. The method of claim 1, wherein, Based on the collaborative control strategy, the lighting device and the advertising resource are controlled, including: The collaborative control strategy is converted into the brightness parameter of the lighting device and the display parameter of the advertising resource; The brightness of the lighting device and the display content of the advertising resource are adjusted through the Internet of Things control system; The energy consumption data of the lighting device and the display effect data of the advertising resource are collected. 7.A smart city lighting advertisement management device, characterized in that, It includes: A data acquisition module is used to collect the state data of city lighting devices, advertising resources and environment scenes through a multi-source perception network, and to preprocess and fuse the state data using an edge computing node to obtain a standardized data stream; A resource grouping module is used to group the lighting devices and advertising resources based on the standardized data stream using a local sensitive hash function, and to optimize the grouping result through iterative local search, including: Based on the standardized data stream, a feature vector is constructed using dimension reduction and feature extraction technology; based on the feature vector, the lighting devices and advertising resources are initially grouped using a local sensitive hash function; the initial grouping result is optimized through iterative local search to obtain the resource grouping result, wherein the initial grouping result is optimized through iterative local search to obtain the resource grouping result, including: Each bucket of the initial grouping result is used as an initial clustering cluster to construct a clustering objective function; the clustering objective function includes intra-cluster similarity, spatial continuity, energy load balancing and advertising coverage efficiency; based on the clustering objective function, local search is performed on the data points in each initial clustering cluster, and the clustering objective function is updated through incremental calculation technology; when the clustering objective function converges, the resource grouping result is obtained; The resource mapping module is configured to construct a random regular graph model based on the resource grouping result, optimize the graph structure by using a Markov chain Monte Carlo algorithm, extract a mapping relationship between the lighting device and the advertising resource, and generate a resource configuration strategy. The optimization of the graph structure by using the Markov chain Monte Carlo algorithm includes: constructing an energy function containing resource configuration balance and control system responsiveness based on the random regular graph model; iteratively optimizing the random regular graph model by using a block parallel Markov chain Monte Carlo algorithm to obtain an optimized graph structure; and extracting a mapping relationship between a device node and a resource node based on the optimized graph structure and in combination with the energy function, and generating the resource configuration strategy. The control decision module is configured to generate a cooperative control strategy by using a multi-objective optimization algorithm based on the resource configuration strategy and in combination with user behavior data and state data of the environment scene. The execution feedback module is configured to control the lighting device and the advertising resource based on the cooperative control strategy, simultaneously collect running state data and display effect data of both the lighting device and the advertising resource, and feed back the running state data and the display effect data as optimization input to the multi-source perception network to realize intelligent city lighting advertising management.

Images