Multi-source heterogeneous spatial data retrieval method and system based on R tree improvement

Abstract

The invention provides a multi-source heterogeneous spatial data retrieval method and system based on R tree improvement. Wherein the data packet carrying the geographic label in the network data flow is processed. The data packets comprise at least two different types of spatial information, and the priority of the splitting dimension of the R tree is adjusted by analyzing the space-time aggregation degree of the data packets. An initial split path is generated by adopting a genetic algorithm, and co-evolution is executed at network edge nodes to optimize cross-source retrieval efficiency. And updating network topology adaptation parameters in each iteration, and constructing an adaptability evaluation function according to the dynamic priority parameters and the transmission delay. And triggering preloading of the heterogeneous data packets based on the path offset, and verifying compatibility by using a gene expression template. Finally, parameters are dynamically adjusted and optimized according to the cross-source retrieval conflict index and the preloaded data verification result, and the retrieval efficiency is improved. According to the technical scheme provided by the invention, the efficiency and accuracy of multi-source heterogeneous spatial data retrieval can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of multi-source heterogeneous spatial data retrieval, and in particular to a multi-source heterogeneous spatial data retrieval method and system based on an improved R-tree. Background Art

[0002] With the popularization of the Internet of Things and smart devices, the amount of multi-source heterogeneous spatial data generated has exploded. These data come from different sensor networks, mobile devices, and various geographic information systems. Each data source may use a different coordinate system, sampling frequency, or data format. In application scenarios such as smart cities, environmental monitoring, and disaster response, how to efficiently and accurately retrieve and integrate these spatial information from different sources has become a key technical requirement. Especially in scenarios with high real-time requirements, such as dynamic path planning in emergency rescue operations or real-time environmental change monitoring, the ability to quickly obtain cross-source heterogeneous spatial data is particularly important.

[0003] At present, in the scenario of processing multi-source heterogeneous spatial data, a targeted existing solution is to use a spatial data management and query system based on a cloud computing platform. This system uploads spatial data from various sources to the cloud, and uses the powerful computing power and elastic resource allocation mechanism of cloud service providers to store and process these data. Users can access these data through application programming interfaces or specific application software and perform complex query operations. This solution supports input and output of multiple data formats, can provide a certain degree of data integration capabilities, and has good scalability for large-scale data sets.

[0004] However, this approach also has some significant drawbacks. First, since all data needs to be transmitted to the cloud for processing, it leads to a high dependence on network bandwidth, especially in application scenarios with high real-time requirements, network latency may become a bottleneck. Second, although cloud computing platforms provide powerful computing resources, they are not optimized for efficient retrieval of heterogeneous data, especially when facing data with different coordinate systems and sampling frequencies. There is a lack of effective spatiotemporal aggregation analysis, making it difficult to achieve optimal data retrieval path planning. In addition, the centralized processing mode in the cloud may raise privacy and security issues, especially when it comes to sensitive geographic information, and the transfer of data ownership and control to third-party service providers will bring additional risks. These issues limit its applicability and reliability in certain specific application scenarios. Summary of the invention

[0005] The present application provides a multi-source heterogeneous spatial data retrieval method and system based on an improved R-tree, so as to solve the problems of low efficiency and poor accuracy of multi-source heterogeneous spatial data retrieval in the prior art.

[0006] In a first aspect, the present application provides a multi-source heterogeneous spatial data retrieval method based on an improved R-tree, comprising: Parse multi-source heterogeneous data packets carrying geographic tags in network data streams. Each data packet contains spatial information of at least two different coordinate systems, sampling frequencies or data formats. Generate dynamic priority parameters that constrain the R-tree splitting dimension based on the spatiotemporal aggregation of each data packet on the transmission path. Extracting a benchmark code of a split path corresponding to a cross-source query pattern in a historical record of a network data flow, wherein the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and using the benchmark code as a gene expression template of the split path in an initial population of a genetic algorithm; Executing the co-evolution of the split path at the network edge node, driving the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generating a path offset and updating the network topology adaptation parameters of the gene expression template in each iteration, and constructing an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; Triggering heterogeneous data packet preloading according to the direction and magnitude of the path offset, the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifying the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; Feedback parameters are generated by the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and the mapping relationship between the dynamic priority parameter and the gene expression template is dynamically corrected.

[0007] Optionally, the co-evolution of the split path is performed at the network edge node, driving the split path to iterate in the direction of minimizing cross-source retrieval conflicts, each iteration generates a path offset and updates the network topology adaptation parameters of the gene expression template, including: The split path is encoded into a multi-dimensional vector, where each dimension corresponds to the adjustable interval of the network topology adaptation parameter. The topological overlap is calculated based on the spatiotemporal area covered by the current split path. Constructing a conflicting neighborhood table according to the topological overlap, recording the heterogeneous data packet transmission paths and network topology adaptation parameter differences that overlap with the current split path in time and space, and generating candidate offsets; A dual probability selection mechanism is used to select a target offset from the candidate offsets, wherein a first probability distribution is based on a success rate of the candidate offset in reducing conflicts in historical iterations, and a second probability distribution is based on a compatibility constraint between the target offset and a format conversion rule in a gene expression template; Applying the target offset to the current split path, dynamically adjusting the parameter value of the corresponding dimension in the multidimensional vector, and updating the topological overlap threshold of the conflicting neighbor table to trigger the elimination of low-priority paths; The time-space coverage matching degree between the updated split path and the preloaded data packet set is recalculated, and the change in the time-space coverage matching degree is fed back to the weight calculation of the dynamic priority parameter to achieve an iterative closed loop.

[0008] Optionally, the adopting a dual probability selection mechanism to select a target offset from the candidate offsets includes: Based on the success rate of reducing conflicts by the candidate offset in historical iterations, the ratio of the number of successes of each candidate offset to the number of path priority enhancements within a preset time window is counted and normalized to an initial weight of the first probability distribution; Extracting the protocol field bound to the format conversion rule in the gene expression template, calculating the overlap ratio between the parameter dimension corresponding to the candidate offset and the preset field value range, and mapping it to the constraint weight of the second probability distribution; Nonlinearly superimposing the initial weight of the first probability distribution and the constraint weight of the second probability distribution, and dynamically adjusting the superposition coefficient according to the proportion of low-priority paths in the conflict neighborhood table, so that the compatibility constraint has a higher weight in the conflict-intensive area; Roulette screening is performed based on the comprehensive selection probability to verify the non-conflict between the candidate offset and the core dimension parameters in the gene expression template. If there is a conflict, the second-best candidate offset is traced back until the target offset is selected after the constraints are met.

[0009] Optionally, the nonlinear superposition of the initial weight of the first probability distribution and the constraint weight of the second probability distribution, wherein the superposition coefficient is dynamically adjusted according to the proportion of the low priority path in the conflict neighborhood table, includes: Based on the comparison between the proportion of low-priority paths in the conflict neighborhood table and the historical maximum path capacity, the conflict density index is calculated through a piecewise function; A dynamic superposition coefficient is generated according to the conflict density index, and an adjustment mechanism triggered by a segmented threshold is adopted, and a historical sliding mean is combined as a dynamic attenuation factor to suppress the sudden change of the dynamic superposition coefficient, wherein the growth rate of the dynamic superposition coefficient is adjusted as the conflict density index changes within the threshold interval; Associating the initial weight of the first probability distribution with the dynamic superposition coefficient, and superimposing the association result with the constraint weight of the second probability distribution, performing logarithmic weighted summation and truncation processing on the superposition coefficient; The rate of the dynamic attenuation factor is adjusted according to the difference between the superposition coefficient and the historical sliding mean. If it exceeds the tolerance interval, the rate is increased, and the updated rate is applied to the superposition coefficient calculation of the next iteration.

[0010] Optionally, constructing an adaptability evaluation function based on the association between the dynamic priority parameter and the data packet transmission delay includes: Based on the spatiotemporal distribution of the dynamic priority parameters, calculating the spatiotemporal coupling coefficient of data packet transmission delay and geographic distribution density; Generating a dynamic weight factor according to the spatiotemporal coupling coefficient, associating the spatiotemporal aggregation of high-delay regions with low-density regions, and constraining the dynamic weight factor based on a network topology adaptation parameter of a gene expression template; The dynamic priority parameter, dynamic weight factor and spatiotemporal coupling coefficient are integrated to construct an adaptability evaluation function, and a penalty term related to the difference in network topology adaptation parameters is introduced, wherein the dynamic priority parameter is used as the main decision variable and the dynamic weight factor is used as the conflict sensitivity adjustment coefficient.

[0011] Optionally, generating a dynamic weight factor according to the spatiotemporal coupling coefficient to associate the spatiotemporal aggregation of the high-delay area with the low-density area includes: Based on the spatiotemporal coupling coefficient and the inverse correlation between the transmission path fluctuation and the change in the geographical distribution gradient, the spatiotemporal aggregation difference index between the high-latency area and the low-density area is calculated; Generate an initial dynamic weight factor according to the spatiotemporal aggregation difference index, associate the path conflict sensitivity of the high-delay area with the spatiotemporal coverage sparsity of the low-density area, and constrain the growth rate of the initial dynamic weight factor through the network topology adaptation parameter of the gene expression template; Based on the historical correction record of the network topology adaptation parameter, a dynamic attenuation factor is applied to the initial dynamic weight factor to suppress the asymmetric associated overload between the high-latency area and the low-density area, thereby generating a dynamic weight factor.

[0012] Optionally, the generating of feedback parameters by using the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result includes: Calculate the conflict and compatibility coupling coefficient based on the cross-source retrieval conflict index output by the adaptability evaluation function and the protocol compatibility ratio in the preloaded data verification result; generating an initial feedback parameter according to the conflict and compatibility coupling coefficient, associating the protocol compatibility loss in the high-conflict region with the path offset, and applying a dynamic attenuation constraint to the initial feedback parameter based on the historical correction record of the gene expression template; Based on the initial feedback parameters, the conflict and compatibility coupling coefficient, the cross-source retrieval conflict index and the protocol compatibility ratio are integrated to construct a multidimensional feedback model, and a penalty term related to the dynamic priority parameter difference is introduced to generate feedback parameters, wherein the cross-source retrieval conflict index serves as the main regulating variable of the multidimensional feedback model.

[0013] In a second aspect, the present application provides a multi-source heterogeneous spatial data retrieval system based on an improved R-tree, comprising: A parsing module is used to parse multi-source heterogeneous data packets carrying geographic tags in network data streams. Each data packet contains spatial information of at least two different coordinate systems, sampling frequencies or data formats. The dynamic priority parameters constraining the R-tree splitting dimension are generated according to the spatiotemporal aggregation of each data packet on the transmission path. A matching module extracts a benchmark code of a split path corresponding to a cross-source query pattern in a historical record of a network data flow, wherein the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and uses the benchmark code as a gene expression template of the split path in an initial population of a genetic algorithm; An update module executes the co-evolution of the split path at the network edge node, drives the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generates a path offset in each iteration and updates the network topology adaptation parameters of the gene expression template, and constructs an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; A trigger module triggers the preloading of heterogeneous data packets according to the direction and magnitude of the path offset, wherein the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifies the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; A generation module generates feedback parameters through the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and dynamically corrects the mapping relationship between the dynamic priority parameter and the gene expression template.

[0014] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a multi-source heterogeneous spatial data retrieval method based on an R-tree improvement as described in the first aspect above.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a multi-source heterogeneous spatial data retrieval method based on an improved R-tree as described in the first aspect.

[0016] In an embodiment of the present application, multi-source heterogeneous data packets carrying geographic tags in a network data stream are parsed, each data packet contains spatial information of at least two different coordinate systems, sampling frequencies or data formats, and a dynamic priority parameter constraining the R-tree split dimension is generated according to the spatiotemporal aggregation of each data packet on the transmission path; a benchmark code of the split path corresponding to the cross-source query mode in the historical record of the network data stream is extracted, the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and the benchmark code is used as a gene expression template for the split path in the initial population of the genetic algorithm; the co-evolution of the split path is performed at the network edge node, and the split path is driven to iterate in the direction of minimizing cross-source retrieval conflicts, and each The path offset is generated in the iteration and the network topology adaptation parameters of the gene expression template are updated, and an adaptability evaluation function is constructed based on the correlation between the dynamic priority parameter and the data packet transmission delay; heterogeneous data packet preloading is triggered according to the direction and amplitude of the path offset, and the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and the compatibility of the data packets with the network transmission protocol is verified through the format conversion rules in the gene expression template; feedback parameters are generated through the cross-source retrieval conflict index output by the adaptability evaluation function and the preloaded data verification result, and the mapping relationship between the dynamic priority parameter and the gene expression template is dynamically corrected.

[0017] The technical solution of this application has the following beneficial effects: This application parses data packets carrying geographic tags, identifies spatial information of different coordinate systems, sampling frequencies or data formats, and generates dynamic priority parameters based on spatiotemporal aggregation, thereby optimizing the selection of R-tree split dimensions and improving retrieval efficiency. The split path benchmark code corresponding to the cross-source query pattern is extracted from historical records and used as the initial population template of the genetic algorithm to promote the effective initialization of the split path. The co-evolution of the split path is performed at the edge node of the network, and the split path is iteratively driven to develop in the direction of minimizing cross-source retrieval conflicts. The network topology adaptation parameters are updated each iteration, the adaptability evaluation function is optimized, and the retrieval accuracy and speed are enhanced. The preloading of heterogeneous data packets is triggered based on the direction and amplitude of the path offset, and the compatibility is verified through the gene expression template to ensure the consistency of the data transmission protocol and improve data access efficiency. Feedback parameters are generated based on the cross-source retrieval conflict index output by the adaptability evaluation function and the preloaded data verification result, and the mapping relationship between the dynamic priority parameters and the gene expression template is dynamically corrected to achieve continuous optimization.

[0018] Furthermore, the split path is encoded into a multidimensional vector, the topological overlap is calculated to construct a conflict neighborhood table, a dual probability selection mechanism is used to select the target offset, the network topology adaptation parameter value is dynamically adjusted, and the threshold of the conflict neighborhood table is updated to eliminate low-priority paths. The spatiotemporal coverage matching degree of the updated split path and the preloaded data packet set is recalculated, and its change is fed back to the weight calculation of the dynamic priority parameter to form an iterative closed loop. The effect of this mechanism is to significantly improve the accuracy and efficiency of multi-source heterogeneous spatial data retrieval, reduce cross-source retrieval conflicts by continuously optimizing the split path, and enhance the overall performance and response speed of the system.

[0019] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 A flowchart of a multi-source heterogeneous spatial data retrieval method based on an improved R-tree provided by the present application is shown; Figure 2 A schematic diagram of the structure of a multi-source heterogeneous spatial data retrieval system based on an improved R-tree provided by the present application is shown; Figure 3 A schematic diagram of the structure of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0023] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0024] The solution analyzes the spatiotemporal aggregation of each data packet on the transmission path by parsing multi-source heterogeneous data packets carrying geographic tags in the network data stream, and generates dynamic priority parameters that constrain the R-tree splitting dimension. This method aims to optimize the efficiency of spatial data retrieval. Its core lies in identifying and utilizing the relationship between spatial information in different coordinate systems, sampling frequencies or data formats to improve retrieval speed and accuracy.

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0026] Figure 1 A flowchart of a multi-source heterogeneous spatial data retrieval method based on an improved R-tree is provided for an embodiment of the present application. Figure 1 As shown, the method includes: 101. Parse multi-source heterogeneous data packets carrying geographic tags in network data streams, each data packet containing spatial information of at least two different coordinate systems, sampling frequencies or data formats, and generate dynamic priority parameters constraining the R-tree splitting dimension according to the spatiotemporal aggregation of each data packet on the transmission path; In this step, the geo-tagged multi-source heterogeneous data packets refer to data units transmitted in the network, which contain geographic location information and may use different coordinate systems, sampling frequencies or data formats.

[0027] Sampling frequency refers to the time interval for data collection, that is, the number of data points obtained per unit time.

[0028] Data format refers to the way data is presented or encoded.

[0029] The spatiotemporal aggregation measures the concentration of data packets within a specific time and space range by analyzing the transmission delay and geographical distribution density of data packets.

[0030] The R-tree split dimension is a data structure used in spatial access methods. Its split dimension refers to the key factor in determining how to split nodes to maintain balance and efficiency.

[0031] The dynamic priority parameter adaptively adjusts the importance weight of the R-tree split dimension based on the spatiotemporal aggregation to optimize storage and retrieval efficiency.

[0032] In an embodiment of the present application, first, the geo-tagged data packets in the network data stream are parsed to identify their coordinate system, sampling frequency or data format. Then, the spatiotemporal aggregation of each data packet is calculated, which requires combining the transmission delay of the data packet and its geographical distribution. Next, the spatiotemporal aggregation is used to generate dynamic priority parameters that constrain the splitting dimension of the R-tree. This process involves using a spatial data analysis algorithm to evaluate the spatial concentration trend of the data packet and adjusting the R-tree structure based on the results. Ultimately, the R-tree structure is optimized through this series of steps, and the data retrieval efficiency is improved.

[0033] In an intelligent transportation system, location updates sent by vehicle sensors are received and parsed. The system analyzes the spatiotemporal aggregation of these data packets and optimizes the storage structure so that data of vehicles that frequently pass through the same road section can be stored and queried more efficiently. This not only improves query speed, but also enhances the system's responsiveness.

[0034] 102. Extracting a benchmark code of a split path corresponding to a cross-source query pattern in a historical record of network data flow, wherein the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and using the benchmark code as a gene expression template of the split path in an initial population of a genetic algorithm; In this step, the network data flow history records refer to the history records of all packets passing through the network during transmission. These records contain information such as the source, destination, transmission time, content summary, etc. of the packet, which are used for subsequent analysis and optimization.

[0035] The split path corresponding to the cross-source query mode means that when data needs to be obtained from multiple different sources, the system will select the optimal data acquisition path based on different query requirements. The split path refers to the different strategies or path selections taken in this process to improve efficiency.

[0036] Benchmark coding is a coding method used to describe the temporal and spatial coverage complementarity of different data packet transmission paths, which serves as the basic template for splitting paths in the initial population of the genetic algorithm.

[0037] The temporal and spatial coverage complementarity of different data packet transmission paths refers to whether the temporal and spatial coverage of different data packets on their transmission paths can complement each other.

[0038] The genetic algorithm starts with a set of individuals that are randomly generated or selected based on certain rules. Each individual represents a potential solution. The initial population of the genetic algorithm refers to the initial set of split paths consisting of benchmark codes.

[0039] The gene expression template contains the format conversion rules and network topology adaptation parameters of multi-source heterogeneous data packets, guiding the path selection optimization in the cross-source query mode.

[0040] In the embodiment of the present application, the split path related to the cross-source query mode is first extracted from the historical network data stream, and the corresponding benchmark code is generated. This process uses the spatiotemporal coverage analysis technology to ensure that the spatiotemporal coverage complementarity between different data packets is fully utilized. Then, the benchmark code is used as the starting point of the genetic algorithm to perform path optimization iterations. Each iteration adjusts the format conversion rules and network topology adaptation parameters according to the effectiveness of the current path, and finally forms an optimized split path set.

[0041] Continuing with the example of the intelligent transportation system above, when integrating traffic data from multiple sources, the system uses the previously mentioned benchmark encoding and genetic algorithms to find the best data integration path to ensure data consistency and accuracy. As new data is added, the system continuously optimizes the path selection to reduce query errors caused by differences in data sources.

[0042] 103. Execute the co-evolution of the split path at the network edge node, drive the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generate a path offset in each iteration and update the network topology adaptation parameters of the gene expression template, and construct an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; In this step, co-evolution refers to the joint improvement process of split paths performed on network edge nodes, aiming to minimize cross-source retrieval conflicts.

[0043] Minimizing cross-source retrieval conflicts means reducing conflicts or inconsistencies that may arise when retrieving data from multiple different sources simultaneously by optimizing the data acquisition path, thereby ensuring data consistency and accuracy.

[0044] The path offset indicates the direction and magnitude of the change in the split path.

[0045] In genetic algorithms, a gene expression template is a template structure that contains the basic building blocks of the solution (such as format conversion rules and network topology adaptation parameters) and is used to guide mutation and crossover operations during the search process.

[0046] Network topology adaptation parameters are used to describe the network structure characteristics and help adjust the algorithm to adapt to the needs of a specific network environment, such as how data packets are routed in the network.

[0047] The adaptability evaluation function is constructed based on the relationship between the dynamic priority parameter and the packet transmission delay, and is used to evaluate the performance of the split path.

[0048] In the embodiment of the present application, the co-evolutionary algorithm is first run on the network edge node to gradually adjust the split path to reduce cross-source retrieval conflicts. This process uses the mutation and crossover operations in the evolutionary algorithm, and generates a path offset and updates the network topology adaptation parameters after each iteration. Then, the adaptability evaluation function calculates the effectiveness of the split path based on the relationship between the dynamic priority parameter and the packet transmission delay, and finally determines the optimal split path.

[0049] In the intelligent transportation system, as new data is continuously added, the system continuously optimizes the data processing path, reduces query errors caused by differences in data sources, and improves overall data processing efficiency. For example, during peak hours, the system automatically adjusts the data processing path to cope with increased data traffic and ensure real-time response.

[0050] 104. Triggering heterogeneous data packet preloading according to the direction and magnitude of the path offset, wherein the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifying the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; In this step, heterogeneous data packet preloading refers to downloading the data packets that may be needed into the cache in advance according to the prediction model for fast access, which is particularly important when processing data packets from different sources and in different formats.

[0051] Preloaded data is a set of data packets that are pre-acquired to speed up queries. These data packets come from neighboring nodes in the network topology and match the spatiotemporal coverage area of ​​the current split path.

[0052] The spatiotemporal coverage area of ​​the current split path refers to the time period and geographical area that the currently selected data acquisition path can cover, and is used to evaluate the effectiveness and applicability of the path.

[0053] Format conversion rules are used to verify the compatibility of data packets under network transmission protocols.

[0054] Network transmission protocols define a set of rules for how data is encapsulated, addressed, transmitted, routed, and received on the network.

[0055] Compatibility verification is the process of ensuring that data packets conform to expected format and protocol standards, and typically involves checking whether the data packets meet specific format requirements and technical specifications.

[0056] In the embodiment of the present application, the preloading operation of heterogeneous data packets is first triggered according to the path offset, and this process involves the analysis and prediction of the network topology. A machine learning algorithm is used to predict the data packets that may be accessed in the future and load them into the cache in advance. At the same time, these data packets are verified using format conversion rules to ensure that they meet the requirements of the network transmission protocol. Ultimately, the above measures speed up data access and improve system performance.

[0057] In the intelligent transportation system, in order to cope with the upcoming peak hours, the system loads the traffic flow data that may be used in advance to ensure real-time response speed. In addition, by verifying the format conversion rules of the data packet, it ensures that all data can be seamlessly integrated into the existing system to provide a consistent service experience.

[0058] 105. Generate feedback parameters through the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and dynamically correct the mapping relationship between the dynamic priority parameter and the gene expression template.

[0059] In this step, the cross-source retrieval conflict index is used to measure the degree of retrieval conflict between different data sources. It is an important indicator for evaluating system performance. A low index means higher data consistency.

[0060] The preloaded data verification result is a check result of the correctness and availability of the preloaded data, ensuring that the data can be used normally in actual applications.

[0061] The feedback parameters are generated by the cross-source retrieval conflict index output by the adaptive evaluation function and the preloaded data verification result, and are used to dynamically correct the mapping relationship between the dynamic priority parameters and the gene expression template to achieve continuous optimization.

[0062] In the embodiment of the present application, firstly, based on the result of the adaptability evaluation function, combined with the cross-source search conflict index and the preloaded data verification situation, the association between the dynamic priority parameter and the gene expression template is dynamically adjusted. This process adopts a closed-loop feedback mechanism to ensure that the system always remains in the optimal state through continuous evaluation and adjustment. Ultimately, through this dynamic correction mechanism, the continuous optimization of the entire data processing process is achieved.

[0063] In an intelligent transportation system, the system continuously monitors its own performance and adjusts its internal mechanisms based on the latest data to ensure that the most accurate and timely traffic information services are always provided. For example, by continuously adjusting the data processing path and optimizing the preloading strategy, the system's response speed and accuracy to emergencies are improved.

[0064] In summary, steps 101 to 105 have greatly improved data processing efficiency and query accuracy, and enhanced user experience, by effectively managing and optimizing multi-source heterogeneous data packets in network data flows, especially in the intelligent transportation system scenario. Each step is closely linked to form a complete optimization cycle, ensuring that the system can maintain efficient operation in a dynamic environment.

[0065] In order to further improve the split path optimization of network edge nodes, the scheme encodes the split path into a multidimensional vector and calculates the topological overlap to identify conflicts, constructs a conflict neighborhood table to generate candidate offsets, uses the selected offsets to adjust the split path parameters, and eliminates low-priority paths. Finally, the spatiotemporal matching changes of the preloaded data packets are evaluated to form an iterative closed loop to continuously reduce cross-source retrieval conflicts. In some embodiments, the co-evolution of the split path is performed at the network edge node in step 103, driving the split path to iterate in the direction of minimizing cross-source retrieval conflicts, and each iteration generates a path offset and updates the network topology adaptation parameters of the gene expression template, including: 201. Encode the split path into a multi-dimensional vector, where each dimension corresponds to an adjustable interval of the network topology adaptation parameter, and calculate the topological overlap based on the spatiotemporal area covered by the current split path; In step 201, a multidimensional vector is a mathematical representation used to describe the various attributes of a split path; network topology adaptation parameters refer to various settings that adjust the network structure to adapt to specific requirements, such as bandwidth, latency, etc. Topology overlap measures the degree of overlap of different data packet transmission paths in time and space, and is used to evaluate the possibility of conflict. This step helps identify potential data conflict areas and provides a basis for subsequent optimization.

[0066] In the embodiment of the present application, the split paths are first converted into a multidimensional vector form, and each path is represented by a series of numerical values ​​to represent its characteristics in the network. Next, the topological overlap is calculated based on the spatiotemporal area covered by the current path, and the spatiotemporal overlap between the paths is quantified using geographic information system technology and algorithms. This information is then analyzed to determine which paths may cause conflicts, thereby guiding the subsequent optimization work. The final result is a detailed evaluation system for measuring the degree of overlap between the split paths and the network environment.

[0067] 202. Construct a conflicting neighborhood table according to the topology overlap, record the heterogeneous data packet transmission paths and network topology adaptation parameter differences that overlap with the current split path in time and space, and generate candidate offsets; In step 202, the conflict neighbor table is a tool for locating and resolving cross-source search conflicts; heterogeneous data packet transmission paths and network topology adaptation parameter differences refer to the paths used by data packets from different sources during transmission and the differences in the network structure configuration of these paths; candidate offsets are potential optimization solutions generated from these differences to reduce conflicts. In this way, specific paths that need to be optimized can be identified more accurately.

[0068] In the embodiment of the present application, a conflict neighborhood table is first constructed based on the topological overlap, and all relevant transmission paths and parameter differences are recorded. This information is processed using data analysis technology to screen out path combinations that may lead to high conflict rates. Subsequently, candidate offsets are generated based on the characteristics of these paths as the basis for the next step of optimization. Finally, by evaluating the effects of different schemes, the most appropriate offset is selected and applied to the actual path adjustment.

[0069] 203. Select a target offset from the candidate offsets using a dual probability selection mechanism, wherein a first probability distribution is based on a success rate of the candidate offset in reducing conflicts in historical iterations, and a second probability distribution is based on a compatibility constraint between the target offset and a format conversion rule in a gene expression template; In step 203, the dual probability selection mechanism is a decision algorithm that combines two independent probability distributions to determine the best path adjustment solution; the first probability distribution considers the results of previous adjustment attempts and evaluates the effectiveness of each candidate offset in reducing conflicts in the future; the second probability distribution focuses on existing format conversion rules and evaluates whether each candidate offset can work well with other system components. This mechanism improves the efficiency and success rate of the optimization process.

[0070] In the embodiment of the present application, a statistical method is first used to analyze the historical performance of the candidate offsets to determine the first probability distribution. At the same time, the compatibility of each offset is evaluated in combination with the format conversion rules to establish a second probability distribution. Then, the selection probability of each offset is calculated based on these two distributions. Finally, the best offset is selected and applied to the actual path adjustment to ensure that the split path can run efficiently.

[0071] 204. Apply the target offset to the current split path, dynamically adjust the parameter value of the corresponding dimension in the multidimensional vector, and update the topological overlap threshold of the conflicting neighbor table to trigger the elimination of low-priority paths. In step 204, the target offset refers to the selected optimal path adjustment scheme; dynamically adjusting the corresponding dimension parameter value in the multidimensional vector means updating the path attributes according to the new offset, such as increasing or decreasing the delay, changing the bandwidth, etc.; updating the topological overlap threshold of the conflicting neighbor table is to modify the standard value used to determine whether there is a significant spatiotemporal coverage overlap; low-priority path elimination refers to automatically removing those data packet transmission paths that contribute less to the overall performance. This process continuously optimizes the split path configuration and reduces cross-source retrieval conflicts.

[0072] In the embodiment of the present application, the target offset is first applied to update the split path parameters, and the relevant thresholds of the conflicting neighbor table are adjusted synchronously. An automated script is used to execute the elimination process of low-priority paths. The data processing and calculation involved in this process all rely on pre-set algorithms and models. Finally, real-time optimization of the split path configuration is achieved to ensure efficient data transmission.

[0073] 205. Recalculate the time-space coverage matching degree between the updated split path and the preloaded data packet set, and feed back the change in the time-space coverage matching degree to the weight calculation of the dynamic priority parameter to achieve an iterative closed loop.

[0074] In step 205, recalculation refers to re-evaluating the matching between the data packet transmission path and the pre-loaded data packet after applying the new offset; the change in the spatiotemporal coverage matching degree is to use the change in the matching degree as input to adjust the importance weights of the dynamic priority parameters of different paths to optimize future data flow allocation; the weight calculation of the dynamic priority parameters is to use the change information of the spatiotemporal coverage matching degree to adjust the importance weights of different paths to ensure that the system can optimize the allocation strategy of future data flows based on the latest performance; the iterative closed loop refers to continuously improving the selection and configuration of the data packet transmission path by continuously repeating the above process to achieve the best performance state.

[0075] In the embodiment of the present application, the spatiotemporal coverage matching degree after the split path update is first calculated, and its changes are analyzed in detail. Then, the machine learning algorithm is used to analyze this information to adjust the weight of the dynamic priority parameter. The whole process emphasizes data-driven decision making to ensure that each iteration can bring substantial improvements.

[0076] Here is a specific example: In an intelligent traffic management system, heterogeneous data streams generated by multiple sensors need to be integrated and analyzed. First, the transmission paths of each data stream are encoded as multidimensional vectors, and their topological overlap in the network is calculated. Next, a conflict neighborhood table is constructed to identify the packet paths with overlapping spatiotemporal coverage and generate candidate offsets. Then, a dual probability selection mechanism is used to select the optimal offset for path adjustment, while eliminating low-priority paths. Finally, the spatiotemporal matching between the split path and the preloaded packet set is re-evaluated, which significantly reduces cross-source retrieval conflicts and improves the system's response speed and data processing efficiency.

[0077] In summary, steps 201 to 205 achieve effective optimization of the split path on the network edge node, significantly reduce conflicts in cross-source data retrieval, and improve the adaptability and flexibility of the data transmission path. Through continuous feedback and adjustment, the system can more accurately meet real-time needs and enhance overall stability and response speed. This approach not only improves the user experience, but also provides strong support for data management and utilization in complex environments, ensuring efficient data processing and rapid decision-making.

[0078] In order to further improve the accuracy of candidate offset selection during split path optimization, the scheme combines historical data with real-time conditions to optimize offset selection in path planning. First, the historical success rate of each candidate offset is evaluated, and then its applicability is checked according to existing rules. Then, the weight is dynamically adjusted to adapt to the current traffic hotspots, and finally the optimal offset is determined by roulette screening. This method aims to improve the accuracy and flexibility of decision-making. In some embodiments, the dual probability selection mechanism described in step 203 is used to select the target offset from the candidate offsets, including: 301. Based on the success rate of reducing conflicts of the candidate offsets in historical iterations, the ratio of the number of successes of each candidate offset to the number of path priority enhancements within a preset time window is counted and normalized to an initial weight of a first probability distribution. In step 301, the candidate offsets in the historical iteration refer to all possible position adjustment values ​​that are attempted to reduce conflicts during the path planning process. The success rate is a measure of the probability that a specific offset can successfully reduce conflicts. The preset time window is a fixed time period set for calculating the success rate, and the data collected during this period will be used to evaluate the performance of each candidate offset. The initial weight of the first probability distribution represents the likelihood of each candidate offset based on its past performance.

[0079] In an embodiment of the present application, all data within a preset time window are first analyzed, and the number of successes and the number of path priority increases for each candidate offset are counted. Basic data statistical methods are used here to obtain the number of successes by recording whether the conflict is reduced after each path adjustment, and the number of path priority increases is obtained according to the priority change after path optimization. These statistical data are then normalized to determine the initial weight of the first probability distribution of each candidate offset. The final result is obtained through in-depth mining and quantitative analysis of historical data, which helps to identify the offset that is most likely to solve the current problem.

[0080] 302. Extract the protocol field bound to the format conversion rule in the gene expression template, calculate the overlap ratio between the parameter dimension corresponding to the candidate offset and the preset field value range, and map it to the constraint weight of the second probability distribution; In step 302, the gene expression template is a data structure for describing the path planning parameter settings, which includes various rules and protocol fields. The protocol field bound to the format conversion rule refers to the specific protocol field associated with the rule set defined in the gene expression template for conversion between different formats. The overlap ratio reflects the degree of match between the parameter dimension corresponding to the candidate offset and the preset field value range. The constraint weight of the second probability distribution measures the applicability of the candidate offset based on this degree of match.

[0081] In the embodiment of the present application, the protocol fields specific to the format conversion rules are first extracted from the gene expression template. This process involves data structure analysis and mathematical calculation techniques. Then, the overlap ratio between the parameter dimensions corresponding to the candidate offsets and the value ranges of these fields is calculated. The purpose is to find the best match between the candidate offsets and the existing rules. Finally, the calculated overlap ratio is mapped to the constraint weight of the second probability distribution, thereby providing basic data support for subsequent steps.

[0082] 303. Nonlinearly superimpose the initial weight of the first probability distribution and the constraint weight of the second probability distribution, and dynamically adjust the superposition coefficient according to the proportion of low-priority paths in the conflict neighborhood table, so that the compatibility constraint has a higher weight in the conflict-intensive area; In step 303, nonlinear superposition refers to a weighted summation method combining multiple factors, where the weights are not fixed but dynamically change according to the input. The conflict neighborhood table is a data structure that records all possible conflict locations and their severity, and the proportion of low priority paths refers to the proportion of low priority paths in these conflict areas to the total paths.

[0083] In an embodiment of the present application, a nonlinear superposition method is first used to combine the initial weight of the first probability distribution with the constraint weight of the second probability distribution. This step uses a complex algorithm to dynamically adjust the weight. Then, the superposition coefficient is adjusted according to the proportion of low-priority paths in the conflict neighborhood table to ensure that the compatibility constraint has a higher weight in the conflict-intensive area, and finally a comprehensive selection probability distribution is formed.

[0084] 304. Perform roulette wheel screening based on the comprehensive selection probability to verify the non-conflict between the candidate offset and the core dimension parameters in the gene expression template. If there is a conflict, backtrack to the suboptimal candidate offset until the target offset is selected after the constraints are met.

[0085] In step 304, roulette screening is a random selection method in which the probability of each option being selected is proportional to its weight. The core dimension parameters are the most important parameters in path planning and determine the basic characteristics of the path. This step is to verify whether the candidate offset has no conflict with these key parameters.

[0086] In the embodiment of the present application, a roulette wheel screening is performed based on the comprehensive selection probability generated in step 303 to check whether each candidate offset is compatible with the core dimension parameters in the gene expression template. If a conflict is found, the suboptimal candidate offset is automatically traced back until a target offset that satisfies all constraints is found. This process ensures that the final selected offset is consistent with both historical performance and the current environment.

[0087] Here is a specific example: In a smart city traffic management system, the system evaluates the success rate of different offsets in reducing traffic congestion by analyzing data from the past month. Next, the system checks existing traffic rules and signal control schemes to determine the best match between candidate offsets and these rules. Based on the distribution of congestion hotspots in the current road network, the system dynamically adjusts the strategy for selecting offsets to ensure that compatibility constraints have a higher weight in conflict-intensive areas. Finally, the effectiveness of the selected offset is verified through a roulette screening method, and the next best option is backtracked until the best solution is found. This greatly improves traffic flow and reduces congestion.

[0088] In summary, steps 301 to 304 effectively select the optimal solution by accurately evaluating the historical performance of candidate offsets and their compatibility with existing systems, significantly reducing cross-source retrieval conflicts. The dual probability selection mechanism significantly improves the accuracy and efficiency of offset selection in path planning. By combining historical performance and real-time adaptability, it can not only identify the offset that is most likely to solve the current problem, but also flexibly respond to various emergencies in the complex and ever-changing urban traffic environment. This method uses technical means such as data statistics, overlap ratio calculation and nonlinear superposition to effectively enhance the system's response speed and decision-making quality, making traffic management more scientific and reasonable, and greatly improving the overall operating efficiency and user experience of urban traffic.

[0089] In order to solve the problem of efficiency and accuracy in offset selection in path planning, the scheme determines the weights of the first and second probability distributions according to the success rate of reducing conflicts in historical iterations and the overlap ratio of protocol fields in the process of selecting the target offset, and adjusts these weights through nonlinear superposition, especially increasing the weight of compatibility constraints in conflict-intensive areas. This ensures that the selection of the split path not only reduces conflicts, but also ensures compatibility with network components. In some embodiments, the initial weight of the first probability distribution is nonlinearly superimposed with the constraint weight of the second probability distribution as described in step 303, and the superposition coefficient is dynamically adjusted according to the proportion of low-priority paths in the conflict neighborhood table, including: 401. Based on the comparison between the proportion of low priority paths in the conflict neighborhood table and the historical maximum path capacity, the conflict density index is calculated by a piecewise function; In step 401, the conflict neighborhood table records all possible locations of conflicts and their severity. The low priority path ratio refers to the proportion of low priority paths in these conflict areas to the total paths. The historical maximum path capacity refers to the maximum number of paths that a specific area can carry in a certain period of time in the past. The conflict density index is an indicator calculated by a piecewise function to measure the density of conflicts in the current area.

[0090] In the embodiment of the present application, the proportion of low priority paths is first calculated based on the data in the conflict neighborhood table and compared with the historical maximum path capacity. Then a predefined piecewise function is used to calculate the conflict density index. This process involves basic data analysis and mathematical operation techniques, and the final result is a value that reflects the conflict situation in the current area.

[0091] 402. Generate a dynamic superposition coefficient according to the conflict density index, adopt a segmented threshold-triggered adjustment mechanism, and combine the historical sliding mean as a dynamic attenuation factor to suppress the sudden change of the dynamic superposition coefficient, wherein the growth rate of the dynamic superposition coefficient is adjusted as the conflict density index changes within the threshold range; In step 402, the dynamic superposition coefficient is a variable generated according to the conflict density index and used to adjust the relationship between the initial weight of the first probability distribution and the constraint weight of the second probability distribution. The segmented threshold trigger mechanism is a method of automatically adjusting parameters based on preset conditions. The historical sliding mean is used as a dynamic attenuation factor to suppress the sudden change of the dynamic superposition coefficient and ensure its smooth transition. The growth rate describes the speed at which the dynamic superposition coefficient changes with the conflict density index.

[0092] In the embodiment of the present application, firstly, according to the different intervals of the conflict density index, a segmented threshold trigger mechanism is used to adjust the growth rate of the dynamic superposition coefficient, and the historical sliding mean is combined as a dynamic attenuation factor to smooth the transition. This step uses algorithmic logic and statistical methods to ensure that the change of the dynamic superposition coefficient is both sensitive and stable. Through this mechanism, the weight can be effectively adjusted dynamically to adapt to the changing conditions.

[0093] 403. Associating the initial weight of the first probability distribution with the dynamic superposition coefficient, and superimposing the association result with the constraint weight of the second probability distribution, performing logarithmic weighted summation and truncation processing on the superposition coefficient; In step 403, the initial weight of the first probability distribution represents the historical performance probability of each candidate offset. The dynamic superposition coefficient is associated with this weight and superimposed with the constraint weight of the second probability distribution. Logarithmic weighted summation is a mathematical method used to integrate probability distributions from different sources. Truncation is to limit the result to a reasonable range to prevent extreme values ​​from affecting the overall calculation result.

[0094] In the embodiment of the present application, the initial weight of the first probability distribution is first associated with the dynamic superposition coefficient, and then superimposed with the constraint weight of the second probability distribution. This information is integrated using the logarithmic weighted summation method, and the rationality of the result is ensured by truncation. This step uses a combination of mathematical models and algorithms to achieve an effective combination of different weights, and finally forms an optimized comprehensive weight to guide the selection of offsets in path planning.

[0095] 404. Adjust the rate of the dynamic attenuation factor according to the difference between the superposition coefficient and the historical sliding mean. If it exceeds the tolerance interval, increase the rate, and apply the updated rate to the superposition coefficient calculation of the next iteration.

[0096] In step 404, the difference between the superposition coefficient and the historical sliding mean is used to measure whether the current dynamic superposition coefficient has a sudden change. The dynamic attenuation factor is a regulation parameter, and its rate determines the speed and stability of the dynamic superposition coefficient adjustment. The tolerance interval sets an allowable fluctuation range. If it exceeds this range, it is considered that a faster adjustment is needed to restore stability.

[0097] In the embodiment of the present application, the difference between the superposition coefficient and the historical sliding mean is first calculated. If this difference exceeds the preset tolerance interval, it is considered that there is a large fluctuation in the current system state, and the adjustment speed needs to be accelerated to restore the normal state. At this time, the rate of the dynamic attenuation factor is increased, and the updated rate is applied to the calculation of the superposition coefficient of the next iteration. In this way, the drastic changes in the superposition coefficient caused by emergencies can be effectively suppressed to ensure the stability and smooth transition of the entire system.

[0098] Here is a specific example: In a smart city traffic management system, the system first calculates the conflict density index in a certain area by analyzing the data in the conflict neighborhood table. Then, the dynamic superposition coefficient is adjusted according to the index and applied to the calculation of the weight. Then, the adjusted coefficient is applied to the superposition of the initial weight and the constraint weight, the logarithmic weighted summation method is used to integrate the information, and the rationality of the result is ensured by truncation. If the superposition coefficient is found to be out of tolerance, the dynamic attenuation factor is quickly adjusted to maintain stability. In this way, the system can effectively deal with traffic congestion and improve overall operating efficiency.

[0099] In summary, steps 401 to 404 significantly improve the flexibility and accuracy of offset selection in the path planning process. By dynamically adjusting the superposition coefficient and combining historical data for smoothing, not only the response speed of the system is enhanced, but also the decision-making deviation caused by emergencies is effectively avoided. This method greatly improves the level of intelligence in urban traffic management and scheduling, makes traffic management more scientific and reasonable, improves user experience, and improves the overall operation efficiency of the city. At the same time, through the adjustment mechanism of the dynamic attenuation factor, the stability and reliability of the system in the face of emergencies are guaranteed.

[0100] In order to solve the optimization problem between data packet transmission delay and network topology in path planning, the construction of the adaptive evaluation function is based on the correlation between the dynamic priority parameter and the data packet transmission delay, integrating the dynamic weight factor and the spatiotemporal coupling coefficient, and introducing a penalty term related to the difference in network topology adaptation parameters. This process helps to identify the spatiotemporal aggregation in high-latency and low-density areas, and further optimize the data retrieval strategy. In some embodiments, the construction of the adaptive evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay in step 103 includes: 501. Calculate the spatiotemporal coupling coefficient of data packet transmission delay and geographic distribution density based on the spatiotemporal distribution of the dynamic priority parameter; In step 501, the dynamic priority parameter refers to the packet processing order indicator that is dynamically adjusted according to the real-time network status and historical data. The spatiotemporal distribution describes how these parameters change over time and geographic location. The packet transmission delay is the time required for data to travel from the sender to the receiver. The geographic distribution density refers to the number of nodes or devices in a certain area. The spatiotemporal coupling coefficient is a quantitative indicator that measures the degree of mutual influence between the packet transmission delay and the geographic distribution density.

[0101] In the embodiment of the present application, the spatiotemporal distribution of the dynamic priority parameter is first analyzed, and the spatiotemporal coupling coefficient is calculated using a statistical method by collecting the data packet transmission delay and node density information at different geographical locations over a period of time. This process involves data analysis, time series analysis, and mathematical modeling techniques, and ultimately generates a value that reflects the relationship between the two, which serves as the basis for subsequent steps.

[0102] 502. Generate a dynamic weight factor according to the spatiotemporal coupling coefficient, associate the spatiotemporal aggregation of the high-delay region with the low-density region, and constrain the dynamic weight factor based on the network topology adaptation parameter of the gene expression template; In step 502, the dynamic weight factor is a variable used to adjust the priority of packet processing, and its value is generated based on the spatiotemporal coupling coefficient. The spatiotemporal aggregation of high-latency areas and low-density areas represents the concentration of these two factors in a specific time and space. The network topology adaptation parameters in the gene expression template are a set of rules for guiding the optimal configuration of the network structure. The dynamic weight factor is constrained on this basis to ensure that efficient packet processing can be maintained when the network topology changes.

[0103] In an embodiment of the present application, a dynamic weight factor is first generated according to the spatiotemporal coupling coefficient calculated in step 501. This step adopts a regression analysis method based on machine learning to predict the optimal weight factor under different conditions by training the model. The model is trained using historical data, including information such as the spatiotemporal coupling coefficient, the geographical distribution density, and the corresponding transmission delay, to find the best weight factor allocation strategy. Next, the spatiotemporal aggregation of high-latency areas and low-density areas is associated. Cluster analysis technology is used here to identify spatial areas with similar characteristics (such as high latency and low density) and evaluate the spatiotemporal aggregation of these areas. This analysis helps determine which areas require higher weight factors for priority processing, thereby reducing delays in the overall network. Finally, all of the above information is integrated to form a comprehensive dynamic weight factor adjustment mechanism. This mechanism can not only respond to real-time changes in network conditions, but also improve its accuracy through continuous learning and optimization.

[0104] 503. The dynamic priority parameter, the dynamic weight factor and the spatiotemporal coupling coefficient are integrated to construct an adaptability evaluation function, and a penalty term related to the difference in network topology adaptation parameters is introduced, wherein the dynamic priority parameter is used as the main decision variable and the dynamic weight factor is used as the conflict sensitivity adjustment coefficient.

[0105] In step 503, the adaptability evaluation function is a mathematical model that comprehensively considers multiple factors and is intended to evaluate system performance under different strategies. The dynamic priority parameter, as the main decision variable, determines the basic processing order of data packets, while the dynamic weight factor, as the conflict sensitivity adjustment coefficient, is used to fine-tune the priority to cope with emergencies. The penalty term related to the difference in network topology adaptation parameters is introduced to prevent performance degradation caused by network structure adjustment.

[0106] In the embodiment of the present application, the dynamic priority parameter, the dynamic weight factor and the spatiotemporal coupling coefficient are first fused. Specifically, a multivariate regression analysis method is used to quantify the relationship between these factors, and the combination of these parameters is optimized by a machine learning algorithm. Next, when constructing an adaptability evaluation function, a penalty term related to the difference in network topology adaptation parameters is introduced. This process involves an in-depth analysis of the current network status and historical data to determine the optimal parameter configuration. In order to ensure that the evaluation function can reflect the actual operation, simulation technology is also used to simulate the performance under different network conditions. The final result is an evaluation function that can comprehensively consider various influencing factors and provide the best solution.

[0107] Here is a specific example: In a smart city traffic management system, the system first analyzes the spatiotemporal distribution of dynamic priority parameters in a certain area and calculates the spatiotemporal coupling coefficient of packet transmission delay and geographical distribution density. Then, a dynamic weight factor is generated based on the spatiotemporal coupling coefficient and applied to network topology optimization. Then, the dynamic priority parameters, dynamic weight factors and spatiotemporal coupling coefficients are integrated to construct an adaptive evaluation function, and a penalty term is introduced to cope with the challenges brought by changes in network topology. In this way, the system can manage data flows more effectively and improve overall operating efficiency.

[0108] In summary, steps 501 to 503 significantly improve the flexibility and accuracy of data packet processing in path planning. By integrating multiple factors to construct an adaptive evaluation function and continuously optimizing it using a verification and feedback adjustment mechanism, not only the response speed and stability of the system are enhanced, but also the performance fluctuations caused by network structure adjustments are effectively avoided. This method greatly improves the intelligence level of data management and scheduling, makes traffic management more scientific and reasonable, improves user experience, and significantly improves the overall operation efficiency of the city. At the same time, the stability and reliability of the system in the face of emergencies are guaranteed through a dynamic adjustment mechanism.

[0109] In order to solve the optimization problem between data packet transmission delay and network topology in path planning, the solution proposes a method for calculating dynamic weight factors based on spatiotemporal coupling coefficients to address the spatiotemporal aggregation correlation problem between high-latency areas and low-density areas. This method accurately quantifies the spatiotemporal aggregation difference index of high-latency and low-density areas by combining the inverse correlation of geographic distribution gradient changes, and adjusts it through network topology adaptation parameters to achieve effective control of the dynamic weight factor. In some embodiments, the generation of a dynamic weight factor according to the spatiotemporal coupling coefficient in step 502 to associate the spatiotemporal aggregation of high-latency areas with low-density areas includes: 601. Calculate the spatiotemporal aggregation difference index between the high-latency area and the low-density area based on the spatiotemporal coupling coefficient and the inverse correlation between the transmission path fluctuation and the geographical distribution gradient change; In step 601, the spatiotemporal coupling coefficient is a quantitative indicator for measuring the degree of mutual influence between data packet transmission delay and geographical distribution density. Transmission path fluctuation refers to the path changes encountered by data packets during transmission in the network. Geographic distribution gradient changes describe the density changes of nodes or devices in different geographical locations. The spatiotemporal aggregation difference index is an indicator for measuring the difference in spatiotemporal aggregation between high-latency areas and low-density areas.

[0110] In an embodiment of the present application, firstly, based on the spatiotemporal coupling coefficient and the inverse correlation between the transmission path fluctuation and the geographical distribution gradient change, this step adopts statistical analysis and mathematical modeling techniques, and collects data on the transmission path fluctuation, geographical distribution density and its changing trend in different geographical locations over a period of time to calculate the spatiotemporal aggregation difference index between the high-latency area and the low-density area. The final result is a numerical value reflecting the difference in spatiotemporal aggregation between the high-latency area and the low-density area, which serves as the basis for subsequent steps.

[0111] 602. Generate an initial dynamic weight factor according to the spatiotemporal aggregation difference index, associate the path conflict sensitivity of the high-delay area with the spatiotemporal coverage sparsity of the low-density area, and constrain the growth rate of the initial dynamic weight factor through the network topology adaptation parameter of the gene expression template; In step 602, the initial dynamic weight factor is a variable generated based on the time-space aggregation difference index, which is used to adjust the packet processing priority. Path conflict sensitivity refers to the degree of increase in packet transmission delay due to path conflicts in high-latency areas. Sparsity of time-space coverage describes the sparseness of node or device distribution in low-density areas. Network topology adaptation parameters are a set of rules used to guide the optimal configuration of network structure. The growth rate controls the speed at which the initial dynamic weight factor changes over time.

[0112] In the embodiment of the present application, the initial dynamic weight factor is generated according to the spatiotemporal aggregation difference index calculated in step 601. A machine learning algorithm, such as a regression tree or a neural network, is used here to train historical data to find the optimal weight factor allocation strategy. Next, the path conflict sensitivity in the high-latency area is associated with the spatiotemporal coverage sparsity in the low-density area to form a comprehensive evaluation model. Then, the growth rate of the initial dynamic weight factor is constrained based on the network topology adaptation parameters in the gene expression template to ensure that it can remain efficient under different network states. This process utilizes optimization algorithms and simulation techniques to ensure the effectiveness and stability of the weight factor.

[0113] 603. Based on the historical correction record of the network topology adaptation parameter, a dynamic attenuation factor is applied to the initial dynamic weight factor to suppress the asymmetric correlation overload between the high-latency area and the low-density area, and generate a dynamic weight factor.

[0114] In step 603, the historical modification record of the network topology adaptation parameter contains relevant information about the past adjustments to the network topology. The dynamic attenuation factor is an adjustment parameter used to suppress the asymmetric correlation overload phenomenon of the initial dynamic weight factor and ensure its smooth transition in practical applications. The final generated dynamic weight factor is a weight factor adjusted by the dynamic attenuation factor, which can better adapt to the real-time network conditions.

[0115] In an embodiment of the present application, a dynamic attenuation factor is first applied to the initial dynamic weight factor based on the historical correction record of the network topology adaptation parameters. Specifically, the adaptive control theory is used to predict the optimal attenuation factor value under the current network state based on the data in the historical correction record. Then, the asymmetric correlation overload between the high-latency area and the low-density area is suppressed by the dynamic attenuation factor to ensure that the weight factor remains stable under different conditions. Finally, all information is integrated to generate the final dynamic weight factor. This step combines data analysis, adaptive control, and optimization algorithms to ensure that the system can maintain efficient operation in a complex and changeable actual environment.

[0116] Here is a specific example: In a smart city traffic management system, the system first calculates the spatiotemporal aggregation difference index between high-latency areas and low-density areas based on the spatiotemporal coupling coefficient. Then, the initial dynamic weight factor is generated based on the index and applied to the network topology optimization, taking into account path conflict sensitivity and spatiotemporal coverage sparsity. Then, based on the historical correction record of the network topology adaptation parameters, a dynamic attenuation factor is applied to the initial dynamic weight factor to ensure its stability in practical applications. For example, when a high latency and low density are detected in a certain area, the system will automatically adjust the weight factor to prioritize the data packets in this area, thereby alleviating congestion and improving overall traffic fluency.

[0117] In summary, steps 601 to 603 significantly improve the flexibility and accuracy of data packet processing. By dynamically adjusting the weight factor and performing real-time optimization in combination with changes in network topology, not only the response speed and stability of the system are enhanced, but also the performance fluctuations caused by network structure adjustments are effectively avoided. This method greatly improves the intelligence level of data management and scheduling, makes traffic management more scientific and reasonable, improves user experience, and significantly improves the overall operation efficiency of the city. At the same time, by introducing a dynamic attenuation factor mechanism, the stability and reliability of the system in the face of emergencies are guaranteed, ensuring efficient path planning and data management.

[0118] In order to solve the cross-source retrieval conflict and protocol compatibility problems of data packet transmission in path planning, the solution generates feedback parameters based on the cross-source retrieval conflict index output by the adaptability evaluation function and the preloaded data verification result. This method constructs a multidimensional feedback model by calculating the conflict and compatibility coupling coefficient, combining the cross-source retrieval conflict index and the protocol compatibility ratio, and effectively promotes the system's ability to self-adjust and optimize. In some embodiments, the feedback parameters generated by the cross-source retrieval conflict index output by the adaptability evaluation function and the preloaded data verification result in step 105 include: 701. Calculate the conflict and compatibility coupling coefficient based on the cross-source search conflict index output by the adaptability evaluation function and the protocol compatibility ratio in the preloaded data verification result; In step 701, the cross-source search conflict index is a quantitative indicator for measuring the degree of search conflicts between different data sources. The protocol compatibility ratio in the preloaded data verification result refers to the degree of compatibility between different protocols found in the data verification process. The conflict and compatibility coupling coefficient is an indicator for measuring the degree of mutual influence between conflict and compatibility.

[0119] In the embodiment of the present application, firstly, based on the cross-source retrieval conflict index output by the adaptability evaluation function and the protocol compatibility ratio in the preloaded data verification result, a statistical analysis method is used, combined with historical data and the current network status, to calculate the conflict and compatibility coupling coefficient through data analysis technology. Specifically, the conflict situation between different data sources and the verification results of protocol compatibility over a period of time are collected, and a mathematical model is constructed using these data to calculate the conflict and compatibility coupling coefficient. The final result is a numerical value that reflects the intensity of the interaction between conflict and compatibility, which serves as the basis for subsequent steps.

[0120] 702. Generate an initial feedback parameter according to the conflict and compatibility coupling coefficient, associate the protocol compatibility loss in the high-conflict area with the path offset, and impose a dynamic attenuation constraint on the initial feedback parameter based on the historical correction record of the gene expression template; In step 702, the initial feedback parameter is a variable generated based on the conflict and compatibility coupling coefficient, which is used to adjust the packet processing priority. The protocol compatibility loss in the high-conflict area refers to the data transmission problem caused by protocol incompatibility in the high-conflict area. The path offset is a strategy for adjusting the path to reduce conflicts. The dynamic attenuation constraint is a mechanism for suppressing the change speed of the initial feedback parameter to ensure its smooth transition.

[0121] In the embodiment of the present application, the initial feedback parameters are generated according to the conflict and compatibility coupling coefficient calculated in step 701. A machine learning algorithm (such as a support vector machine or a neural network) is used here to find the optimal feedback parameter allocation strategy by training historical data. Next, the lack of protocol compatibility in high-conflict areas is associated with the path offset to form a comprehensive evaluation model. Then, dynamic attenuation constraints are imposed on the initial feedback parameters based on the historical correction records of the gene expression template to ensure that they remain efficient and stable under different network conditions. This process utilizes optimization algorithms and simulation techniques to ensure the effectiveness and stability of the feedback parameters.

[0122] 703. Based on the initial feedback parameters, the conflict and compatibility coupling coefficient, the cross-source retrieval conflict index and the protocol compatibility ratio are integrated to construct a multidimensional feedback model, and a penalty term related to the dynamic priority parameter difference is introduced to generate feedback parameters, wherein the cross-source retrieval conflict index is used as the main regulating variable of the multidimensional feedback model.

[0123] In step 703, the multi-dimensional feedback model is a mathematical model that comprehensively considers multiple factors and is intended to evaluate system performance under different strategies. The penalty term related to the dynamic priority parameter difference is to prevent performance degradation caused by the change of dynamic priority parameters. The cross-source retrieval conflict index, as the main adjustment variable, determines the basic processing order of the data packet.

[0124] In the embodiment of the present application, regression analysis and optimization algorithms are first used to integrate the conflict and compatibility coupling coefficient, the cross-source retrieval conflict index and the protocol compatibility ratio to form a multi-dimensional feedback model. Then, a penalty term related to the difference in dynamic priority parameters is introduced to prevent performance fluctuations caused by changes in dynamic priority parameters. By simulating the performance under different network conditions, the model parameters are adjusted until the best performance is achieved. The final result is a feedback model that can comprehensively consider various factors and provide the best solution to guide actual operations.

[0125] Here is a specific example: In a smart city traffic management system, the system first calculates the conflict and compatibility coupling coefficient. Then, the initial feedback parameters are generated based on the coefficient and applied to the network topology optimization, taking into account the lack of protocol compatibility and path offset in high-conflict areas. Then, the conflict and compatibility coupling coefficient, cross-source retrieval conflict index, and protocol compatibility ratio are integrated to construct a multi-dimensional feedback model. For example, when a high conflict and low protocol compatibility are detected in a certain area, the system will automatically adjust the feedback parameters to optimize the path selection, thereby improving the overall traffic flow.

[0126] In summary, steps 701 to 703 significantly improve the flexibility and accuracy of data packet processing. By dynamically adjusting the feedback parameters and performing real-time optimization in combination with changes in network topology, not only the response speed and stability of the system are enhanced, but also the performance fluctuations caused by protocol compatibility and cross-source retrieval conflicts are effectively avoided. This method greatly improves the intelligence level of data management and scheduling, makes traffic management more scientific and reasonable, improves user experience, and significantly improves the overall operation efficiency of the city. At the same time, by introducing multidimensional feedback models and penalty terms, the stability and reliability of the system in the face of complex and changeable actual environments are guaranteed, ensuring efficient path planning and data management.

[0127] Figure 2 The present application provides a schematic diagram of a multi-source heterogeneous spatial data retrieval system based on an improved R-tree. Figure 2 As shown, the system includes: The parsing module 21 parses multi-source heterogeneous data packets carrying geographic tags in the network data stream, each data packet containing spatial information of at least two different coordinate systems, sampling frequencies or data formats, and generates dynamic priority parameters constraining the R-tree splitting dimension according to the spatiotemporal aggregation of each data packet on the transmission path; A matching module 22 extracts a reference code of a split path corresponding to a cross-source query pattern in a historical record of a network data flow, wherein the reference code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and uses the reference code as a gene expression template of the split path in an initial population of a genetic algorithm; An updating module 23 performs the co-evolution of the split path at the network edge node, drives the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generates a path offset in each iteration and updates the network topology adaptation parameters of the gene expression template, and constructs an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; A trigger module 24 triggers the preloading of heterogeneous data packets according to the direction and magnitude of the path offset, wherein the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifies the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; The generating module 25 generates feedback parameters through the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and dynamically corrects the mapping relationship between the dynamic priority parameter and the gene expression template.

[0128] Figure 2 The multi-source heterogeneous spatial data retrieval system based on R-tree improvement can be executed Figure 1 The implementation principle and technical effect of the multi-source heterogeneous spatial data retrieval method based on R-tree improvement described in the embodiment shown are not repeated here. The specific way in which each module and unit performs operations in the multi-source heterogeneous spatial data retrieval system based on R-tree improvement in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0129] In one possible design, Figure 2 The multi-source heterogeneous spatial data retrieval system based on the improved R-tree of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0130] The processing component 32 is used for the above Figure 1 The embodiment provides a multi-source heterogeneous spatial data retrieval method based on an improved R-tree.

[0131] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0132] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0133] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0134] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0135] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0136] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0137] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a multi-source heterogeneous spatial data retrieval method based on an improved R-tree.

[0138] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0139] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0140] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-source heterogeneous spatial data retrieval method based on an improved R-tree, characterized in that: include: Parse multi-source heterogeneous data packets carrying geographic tags in network data streams. Each data packet contains spatial information of at least two different coordinate systems, sampling frequencies or data formats. Generate dynamic priority parameters that constrain the R-tree splitting dimension based on the spatiotemporal aggregation of each data packet on the transmission path. Extracting a benchmark code of a split path corresponding to a cross-source query pattern in a historical record of a network data flow, wherein the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and using the benchmark code as a gene expression template of the split path in an initial population of a genetic algorithm; Executing the co-evolution of the split path at the network edge node, driving the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generating a path offset and updating the network topology adaptation parameters of the gene expression template in each iteration, and constructing an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; Triggering heterogeneous data packet preloading according to the direction and magnitude of the path offset, the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifying the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; Feedback parameters are generated by the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and the mapping relationship between the dynamic priority parameter and the gene expression template is dynamically corrected.

2. The method according to claim 1, characterized in that The method of executing the co-evolution of the split path at the network edge node, driving the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generating a path offset in each iteration and updating the network topology adaptation parameters of the gene expression template, includes: The split path is encoded into a multi-dimensional vector, where each dimension corresponds to the adjustable interval of the network topology adaptation parameter. The topological overlap is calculated based on the spatiotemporal area covered by the current split path. Constructing a conflicting neighborhood table according to the topological overlap, recording the heterogeneous data packet transmission paths and network topology adaptation parameter differences that overlap with the current split path in time and space, and generating candidate offsets; A dual probability selection mechanism is used to select a target offset from the candidate offsets, wherein a first probability distribution is based on a success rate of the candidate offset in reducing conflicts in historical iterations, and a second probability distribution is based on a compatibility constraint between the target offset and a format conversion rule in a gene expression template; Applying the target offset to the current split path, dynamically adjusting the parameter value of the corresponding dimension in the multidimensional vector, and updating the topological overlap threshold of the conflicting neighbor table to trigger the elimination of low-priority paths; The time-space coverage matching degree between the updated split path and the preloaded data packet set is recalculated, and the change in the time-space coverage matching degree is fed back to the weight calculation of the dynamic priority parameter to achieve an iterative closed loop.

3. The method according to claim 2, characterized in that The adopting a dual probability selection mechanism to select a target offset from the candidate offsets includes: Based on the success rate of reducing conflicts by the candidate offset in historical iterations, the ratio of the number of successes of each candidate offset to the number of path priority enhancements within a preset time window is counted and normalized to an initial weight of the first probability distribution; Extracting the protocol field bound to the format conversion rule in the gene expression template, calculating the overlap ratio between the parameter dimension corresponding to the candidate offset and the preset field value range, and mapping it to the constraint weight of the second probability distribution; Nonlinearly superimposing the initial weight of the first probability distribution and the constraint weight of the second probability distribution, and dynamically adjusting the superposition coefficient according to the proportion of low-priority paths in the conflict neighborhood table, so that the compatibility constraint has a higher weight in the conflict-intensive area; Roulette screening is performed based on the comprehensive selection probability to verify the non-conflict between the candidate offset and the core dimension parameters in the gene expression template. If there is a conflict, the second-best candidate offset is traced back until the target offset is selected after the constraints are met.

4. The method according to claim 3, characterized in that: The nonlinear superposition of the initial weight of the first probability distribution and the constraint weight of the second probability distribution, wherein the superposition coefficient is dynamically adjusted according to the proportion of the low priority path in the conflict neighborhood table, comprises: Based on the comparison between the proportion of low-priority paths in the conflict neighborhood table and the historical maximum path capacity, the conflict density index is calculated through a piecewise function; A dynamic superposition coefficient is generated according to the conflict density index, and an adjustment mechanism triggered by a segmented threshold is adopted, and a historical sliding mean is combined as a dynamic attenuation factor to suppress the sudden change of the dynamic superposition coefficient, wherein the growth rate of the dynamic superposition coefficient is adjusted as the conflict density index changes within the threshold interval; Associating the initial weight of the first probability distribution with the dynamic superposition coefficient, and superimposing the association result with the constraint weight of the second probability distribution, performing logarithmic weighted summation and truncation processing on the superposition coefficient; The rate of the dynamic attenuation factor is adjusted according to the difference between the superposition coefficient and the historical sliding mean. If it exceeds the tolerance interval, the rate is increased, and the updated rate is applied to the superposition coefficient calculation of the next iteration.

5. The method according to claim 1, characterized in that The constructing of an adaptability evaluation function based on the association between the dynamic priority parameter and the data packet transmission delay comprises: Based on the spatiotemporal distribution of the dynamic priority parameters, calculating the spatiotemporal coupling coefficient of data packet transmission delay and geographic distribution density; Generating a dynamic weight factor according to the spatiotemporal coupling coefficient, associating the spatiotemporal aggregation of high-delay regions with low-density regions, and constraining the dynamic weight factor based on a network topology adaptation parameter of a gene expression template; The dynamic priority parameter, dynamic weight factor and spatiotemporal coupling coefficient are integrated to construct an adaptability evaluation function, and a penalty term related to the difference in network topology adaptation parameters is introduced, wherein the dynamic priority parameter is used as the main decision variable and the dynamic weight factor is used as the conflict sensitivity adjustment coefficient.

6. The method according to claim 5, characterized in that Generating a dynamic weight factor according to the spatiotemporal coupling coefficient to associate the spatiotemporal aggregation of the high-delay area with the low-density area includes: Based on the spatiotemporal coupling coefficient and the inverse correlation between the transmission path fluctuation and the change in the geographical distribution gradient, the spatiotemporal aggregation difference index between the high-latency area and the low-density area is calculated; Generate an initial dynamic weight factor according to the spatiotemporal aggregation difference index, associate the path conflict sensitivity of the high-delay area with the spatiotemporal coverage sparsity of the low-density area, and constrain the growth rate of the initial dynamic weight factor through the network topology adaptation parameter of the gene expression template; Based on the historical correction record of the network topology adaptation parameter, a dynamic attenuation factor is applied to the initial dynamic weight factor to suppress the asymmetric associated overload between the high-latency area and the low-density area, thereby generating a dynamic weight factor.

7. The method according to claim 1, characterized in that The generating of feedback parameters by using the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result includes: Calculate the conflict and compatibility coupling coefficient based on the cross-source retrieval conflict index output by the adaptability evaluation function and the protocol compatibility ratio in the preloaded data verification result; generating an initial feedback parameter according to the conflict and compatibility coupling coefficient, associating the protocol compatibility loss in the high-conflict region with the path offset, and applying a dynamic attenuation constraint to the initial feedback parameter based on the historical correction record of the gene expression template; Based on the initial feedback parameters, the conflict and compatibility coupling coefficient, the cross-source retrieval conflict index and the protocol compatibility ratio are integrated to construct a multidimensional feedback model, and a penalty term related to the dynamic priority parameter difference is introduced to generate feedback parameters, wherein the cross-source retrieval conflict index serves as the main regulating variable of the multidimensional feedback model.

8. A multi-source heterogeneous spatial data retrieval system based on R-tree improvement, characterized in that: include: A parsing module is used to parse multi-source heterogeneous data packets carrying geographic tags in network data streams. Each data packet contains spatial information of at least two different coordinate systems, sampling frequencies or data formats. The dynamic priority parameters constraining the R-tree splitting dimension are generated according to the spatiotemporal aggregation of each data packet on the transmission path. A matching module extracts a benchmark code of a split path corresponding to a cross-source query pattern in a historical record of a network data flow, wherein the benchmark code is generated by matching the spatiotemporal coverage complementarity of different data packet transmission paths, and uses the benchmark code as a gene expression template of the split path in an initial population of a genetic algorithm; An update module executes the co-evolution of the split path at the network edge node, drives the split path to iterate in the direction of minimizing cross-source retrieval conflicts, generates a path offset in each iteration and updates the network topology adaptation parameters of the gene expression template, and constructs an adaptability evaluation function based on the correlation between the dynamic priority parameter and the data packet transmission delay; A trigger module triggers the preloading of heterogeneous data packets according to the direction and magnitude of the path offset, wherein the preloaded data is a set of data packets in adjacent nodes of the network topology that match the spatiotemporal coverage area of ​​the current split path, and verifies the compatibility of the data packets with the network transmission protocol through the format conversion rules in the gene expression template; A generation module generates feedback parameters through the cross-source search conflict index output by the adaptability evaluation function and the preloaded data verification result, and dynamically corrects the mapping relationship between the dynamic priority parameter and the gene expression template.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a multi-source heterogeneous spatial data retrieval method based on an R-tree improvement as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, the multi-source heterogeneous spatial data retrieval method based on the improved R-tree as described in any one of claims 1 to 7 is implemented.

Images