Low-altitude economic data management method and system based on block chain

Abstract

The invention discloses a low-altitude economic data management method and system based on a block chain, and relates to the technical field of information management, and the method comprises the steps: inputting a multi-dimensional low-altitude economic data set into a space-time fusion network model, executing topological relation analysis and neighborhood feature aggregation by a space-domain relation analysis layer, carrying out cross-dimensional feature interaction by a space-time interaction layer, and carrying out multi-dimensional low-altitude economic data management. The method comprises the steps of outputting a low-altitude airspace compliance smart contract, executing symbolization analysis and Kriging spatial interpolation on the low-altitude airspace compliance smart contract, obtaining an airspace risk thermodynamic diagram, synchronously executing multi-layer pyramid convolution on the airspace risk thermodynamic diagram, forming an airspace risk value matrix, executing condition mask calculation on the airspace risk value matrix, and obtaining violation probability distribution. According to the method, through the space-time fusion network model and the secure hash algorithm, the security and privacy protection level of the low-altitude economic data are greatly enhanced.

Description

Technical Field

[0001] The present invention relates to the field of information management technology, and in particular to a low-altitude economic data management method and system based on blockchain. Background Art

[0002] With the rapid development of drone technology and low-altitude aircraft, low-altitude economic activities are becoming increasingly frequent, encompassing multiple application scenarios such as logistics and transportation, agricultural plant protection, and emergency rescue. To ensure the safety and compliance of low-altitude flights, there is an urgent need to efficiently integrate and manage low-altitude economic data. Traditional low-altitude economic data management methods primarily rely on centralized database systems and rule-based expert systems, achieving airspace scheduling and flight supervision through pre-set logical judgments and manual intervention. This can, to a certain extent, meet the data processing needs of early low-altitude flight activities.

[0003] However, existing technologies still have some shortcomings. On the one hand, current data management methods lack the ability to effectively adapt to the dynamically changing low-altitude environment and aircraft behavior, making it difficult to update flight safety-related decision-making support information in real time. On the other hand, existing data management methods fail to provide sufficient safeguards for data security and privacy protection, making low-altitude economic data vulnerable to unauthorized access or tampering. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a low-altitude economic data management method based on blockchain to solve the problems of weak adaptability and insufficient security of existing data management methods.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a blockchain-based low-altitude economic data management method, which includes inputting a multi-dimensional low-altitude economic data set into a spatiotemporal fusion network model, performing topological relationship analysis and neighborhood feature aggregation in an airspace relationship analysis layer, and performing cross-dimensional feature interaction in a spatiotemporal interaction layer to output a low-altitude airspace compliance smart contract; Perform symbolic analysis and Kriging spatial interpolation on low-altitude airspace compliance smart contracts to obtain an airspace risk heat map. Simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix. Perform conditional mask calculation on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, perform dynamic key sharding management on the low-altitude airspace compliance smart contract to generate partitioned encrypted data streams. Use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain the blockchain evidence record, perform zero-knowledge audit proof on the blockchain evidence record, and output the data management audit report.

[0007] As a preferred solution of the blockchain-based low-altitude economic data management method described in the present invention, the multidimensional low-altitude economic data set includes aircraft dynamic data, low-altitude airspace environment data and low-altitude airspace compliance rule data.

[0008] As a preferred solution of the blockchain-based low-altitude economic data management method of the present invention, the output of the low-altitude airspace compliance smart contract specifically includes the following steps: Build a spatial relationship analysis layer and a spatiotemporal interaction layer, apply skip connections to perform multi-scale stacking, and construct a spatiotemporal fusion network model; The multi-dimensional low-altitude economic data set is input into the spatiotemporal fusion network model. The spatial relationship analysis layer uses graph convolution to perform topological relationship analysis and neighborhood feature aggregation to generate the spatial node state vector. The spatiotemporal interaction layer applies a multi-head attention mechanism to perform cross-dimensional feature interaction and obtain the spatiotemporal correlation weight matrix; A gating unit is used to perform importance weighting and cross-modal feature fusion on the spatial node state vector and weight matrix to form a spatial regular feature tensor; The contract structure is decoded and parameters are filled in for the airspace rule feature tensor, and a low-altitude airspace compliance smart contract is output.

[0009] As a preferred solution of the blockchain-based low-altitude economic data management method of the present invention, the acquisition of the airspace risk heat map specifically includes the following steps: Decompose the logical structure of the low-altitude airspace compliance smart contract to form a structured rule matrix; The entropy weight method is used to symbolically analyze and redistribute the weights of the structured rule matrix to obtain the symbolic rule parameter distribution; The symbolized rule parameter distribution is subjected to Kriging spatial interpolation and probability surface fitting to generate an airspace risk heat map.

[0010] As a preferred solution of the blockchain-based low-altitude economic data management method of the present invention, the formation of the airspace risk value matrix specifically includes the following steps: The airspace risk heat map is subjected to wavelet transformation and multi-scale decomposition to obtain a multi-resolution feature map sequence, and the resolution feature map sequence is subjected to multi-layer pyramid convolution to obtain a multi-level convolution feature map; Channel attention weighting and cross-scale feature fusion are performed on the multi-level convolutional feature map to form a fused feature tensor, which is linearly projected to output the spatial risk value matrix.

[0011] As a preferred solution of the blockchain-based low-altitude economic data management method of the present invention, wherein: the generation of partitioned encrypted data stream specifically includes the following steps: Perform risk quantification and conditional mask calculation on the spatial risk value matrix to generate a binary mask; perform spatial sliding window statistics on the binary mask to obtain the violation probability distribution; According to the probability distribution of violations, the permissions of the low-altitude airspace compliance smart contract are dynamically adjusted to form encryption control parameters, and the encryption control parameters are dynamically key-sharded to generate partitioned encrypted data streams.

[0012] As a preferred solution of the blockchain-based low-altitude economic data management method of the present invention, the output data management audit report specifically includes the following steps: Use the secure hash algorithm to perform coordinate binding and space-time tag encoding on the partitioned encrypted data stream, obtain the space-time tag hash sequence, perform hierarchical aggregation on the space-time tag hash sequence, and generate blockchain evidence records; Extract and reconstruct spatiotemporal features of blockchain evidence records to form an audit evidence set, perform zero-knowledge audit proof on the audit evidence set, and obtain verifiable statements; Perform structured integration of verifiable claims and output data management audit reports.

[0013] In the second aspect, the present invention provides a low-altitude economic data management system based on blockchain, comprising: The contract generation module is used to input the multi-dimensional low-altitude economic data set into the spatiotemporal fusion network model. The airspace relationship analysis layer performs topological relationship analysis and neighborhood feature aggregation. The spatiotemporal interaction layer performs cross-dimensional feature interaction and outputs the low-altitude airspace compliance smart contract. The risk assessment module is used to perform symbolic analysis and Kriging spatial interpolation on the low-altitude airspace compliance smart contract to obtain the airspace risk heat map, and simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix; The security encryption module is used to perform conditional mask calculations on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, it dynamically manages the key sharding of the low-altitude airspace compliance smart contract and generates partitioned encrypted data streams. The audit evidence module is used to use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain blockchain evidence records, perform zero-knowledge audit proof on the blockchain evidence records, and output data management audit reports.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the blockchain-based low-altitude economic data management method as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the blockchain-based low-altitude economic data management method as described in the first aspect of the present invention.

[0016] The present invention achieves the following benefits: It uses a spatiotemporal fusion network model to process and analyze multidimensional low-altitude economic data in real time, significantly improving its adaptability to dynamically changing environments and aircraft behavior. It also utilizes a secure hash algorithm to perform spatiotemporal tag encoding of partitioned encrypted data streams, protecting the data from unauthorized access and tampering, significantly enhancing the security and privacy of low-altitude economic data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 The flowchart of the low-altitude economic data management method based on blockchain.

[0019] Figure 2 Schematic diagram of the low-altitude economic data management system based on blockchain.

[0020] Figure 3 Flowchart of the spatiotemporal fusion network model processing process.

[0021] Figure 4 Flowchart for airspace risk assessment and encryption. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a low-altitude economic data management method based on blockchain, comprising the following steps: S1. Input the multi-dimensional low-altitude economic data set into the spatiotemporal fusion network model. The airspace relationship analysis layer performs topological relationship analysis and neighborhood feature aggregation. The spatiotemporal interaction layer performs cross-dimensional feature interaction and outputs the low-altitude airspace compliance smart contract.

[0026] The specific steps include: S1.1. Collect multi-dimensional low-altitude economic data sets, which include aircraft dynamic data, low-altitude airspace environment data, and low-altitude airspace compliance rule data.

[0027] Aircraft dynamic data includes position information, speed parameters and flight status; position information is collected through the onboard GPS, speed parameters are collected by the inertial measurement unit, and flight status is collected by the flight control unit;

[0028] Low-altitude airspace environmental data includes meteorological conditions, topography, and electromagnetic interference intensity; meteorological conditions are captured by weather radar, topography is collected using lidar scanning equipment, and electromagnetic interference intensity is collected using a spectrum analyzer;

[0029] Low-altitude airspace compliance rule data includes airspace division rules, flight permission conditions and no-fly zone restrictions; airspace division rules are collected from the air traffic control database, flight permission conditions are extracted through aviation regulations documents, and no-fly zone restrictions are collected using geo-fence units.

[0030] S1.2. Preprocess the multidimensional low-altitude economic data set. In the specific operation, for the aircraft dynamic data, apply mean filtering to filter out noise, and use linear interpolation to perform missing value filling to ensure data continuity, and synchronously use the PTP timestamp protocol to perform time alignment to enhance the temporal consistency of the data; for the low-altitude airspace environmental data, apply Min-Max standardization to unify the dimensions and eliminate unit differences, and use Kriging interpolation for spatial sampling to eliminate abnormal data and improve the reliability of the low-altitude airspace environmental data; for the low-altitude airspace compliance rule data, perform structural standardization and semantic normalization to eliminate expression ambiguity, and use principal component analysis to perform dimensionality compression to ensure storage efficiency, and output the preprocessed multidimensional low-altitude economic data set.

[0031] S1.3. Build a spatiotemporal fusion network model and train it. Specifically, in the PyTorch framework, call the graph convolutional network architecture through the nn.Module base class, set the node feature dimension of the graph convolutional network architecture to 128, the number of attention heads to 4, and the activation function to LeakyReLU; connect the graph convolutional network architecture with layer normalization for gradient stabilization to suppress the drift of node feature scales, and retain features through residual connections to complete the construction of the spatial domain relationship parsing layer; call the Transformer architecture through the nn.keras parameter, and add a multi-head attention mechanism to the Transformer architecture to achieve cross-dimensional feature interaction; set the Key dimension of the Transformer architecture to 64, the Value dimension to 64, and the Dropout rate to 0.1; connect the gated linear unit after the Transformer architecture for feature screening to enhance effective information flow, and enhance temporal features through bidirectional time convolution to complete the construction of the spatiotemporal interaction layer;

[0032] Using skip connections, we perform cross-scale feature concatenation on the spatial relationship analysis layer and the spatiotemporal interaction layer to obtain a 256-dimensional fused feature tensor. This 256-dimensional fused feature tensor is then subjected to 1×1 convolution dimensionality reduction to generate a 128-dimensional regular feature vector. We use the Softmax function to weight the importance of the 128-dimensional regular feature vector to obtain the contribution distribution of features at each scale. We dynamically assign weights to the contribution distribution of features at each scale to obtain multi-scale fusion weights. Based on the multi-scale fusion weights, we perform multi-scale stacking on the spatial relationship analysis layer and the spatiotemporal interaction layer, and perform cross-scale feature aggregation through three-dimensional convolution to complete the construction of the spatiotemporal fusion network model.

[0033] Next, the spatiotemporal fusion network model is trained. The historical multidimensional low-altitude economic data set is further divided into a sample set, a training set, and a validation set in a ratio of 7:2:1. On the sample set, random affine transformation is used for data enhancement to form enhanced training samples. On the training set, AdamW is used to perform gradient backpropagation on the enhanced training samples, and the cosine annealing learning rate scheduler is simultaneously applied to dynamically adjust the learning rate to obtain the intermediate parameters of the spatiotemporal fusion network model. On the validation set, the early stopping mechanism is used to monitor the loss of the intermediate parameters of the spatiotemporal fusion network model to obtain the mean absolute error of the validation set. When the mean absolute error of the validation set exceeds the convergence threshold for 15 consecutive rounds, the training is terminated, and ModelSnapshot is simultaneously applied to output the trained spatiotemporal fusion network model.

[0034] It should be noted that the convergence threshold is defined based on three times the sliding standard deviation of the mean absolute error of the validation set, and its value range is [0.03, 0.07].

[0035] S1.4. Generate a low-altitude airspace compliance smart contract using a spatiotemporal fusion network model. Specifically, the multidimensional low-altitude economic dataset is input into the trained spatiotemporal fusion network model through a standardized data interface. The airspace relationship analysis layer uses three-dimensional graph convolution to perform spatiotemporal feature separation on the multidimensional low-altitude economic dataset, obtaining flight location features and environmental association features. The flight location features are used as node attributes, and the environmental association features are used as edge weights to construct a structured airspace topology map.

[0036] A two-layer graph convolutional network architecture is simultaneously applied to perform topological relationship analysis on the structured spatial topology graph. Furthermore, the first layer applies graph convolution to perform local neighborhood feature propagation to obtain primary node features, and performs layer normalization and residual connection on the primary node features to generate stable node representations. The second layer applies high-order graph convolution to perform cross-hop neighborhood aggregation and nonlinear transformation on the stable node representation to obtain high-order topological features. At the same time, feature splicing and linear projection are performed on the high-order topological features to obtain topological relationship representations.

[0037] A gated attention unit is applied to weight the importance of the topological relationship representation to obtain refined topological features; neighborhood feature aggregation is performed on the refined topological features to generate spatial enhancement features, and a multi-layer perceptron is applied to perform nonlinear mapping and dimensionality compression on the spatial enhancement features to obtain a 128-dimensional spatial node state vector.

[0038] The spatiotemporal interaction layer separates the spatiotemporal features of the multi-dimensional low-altitude economic data set to obtain spatiotemporal decoupling features; performs dimension expansion and nonlinear activation on the spatiotemporal decoupling features to obtain enhanced trajectory representation; adopts a multi-head attention mechanism (8 heads) to perform cross-dimensional feature interaction on the enhanced trajectory representation, and further, performs channel splitting and reorganization on the enhanced trajectory representation to generate a multi-head attention subspace; performs parallel attention calculation on the multi-head attention subspace to obtain the trajectory vector, environment vector and rule vector; defines the trajectory vector as the query vector, the environment vector as the key vector, and the rule vector as the value vector; performs scaled dot product similarity operation on the query vector, key vector and value vector to obtain the attention weight values ​​of different dimensions. The specific mathematical formula is as follows, ;

[0039] in, represents the attention weight value, represents the query vector, represents the key vector, represents the transpose operation, represents the scaling factor, represents a value vector,

[0040] It should be noted that the scaling factor is defined based on the feature dimension of the key vector and its value range is [32-512].

[0041] Softmax normalization is performed on the attention weight values ​​of different dimensions to obtain the attention distribution matrix; the attention distribution matrix is ​​weighted and multi-head splicing is performed to obtain the fusion feature matrix; the fusion feature matrix is ​​batch normalized (BatchNorm) to obtain spatiotemporal enhancement features; the spatiotemporal enhancement features are linearly projected and dimensionally compressed to output a 256-dimensional spatiotemporal correlation weight matrix.

[0042] Feature concatenation is performed on the spatial node state vector and weight matrix to obtain preliminary fusion features; channel dimension separation and GELU nonlinear activation are performed on the preliminary fusion features to obtain multi-channel feature representation; bidirectional gating units are used to perform directional feature processing on the multi-channel feature representation. Furthermore, the feature screening gate of the forward gating unit performs importance weighting on the multi-channel feature representation to obtain forward refined features, and the feature projection gate performs dimension transformation and feature scaling on the forward refined features to generate forward projection features; the context fusion gate of the backward gating unit performs environmental feature injection on the generated forward projection features to form bidirectional interactive features, and the feature normalization gate performs scale adjustment and distribution alignment on the bidirectional interactive features to obtain stable feature output; spatiotemporal feature fusion is performed on the stable feature output to obtain a unified feature representation, and the unified feature representation is tensor reorganized to obtain a fourth-order tensor structure containing spatial grid, time step, feature channel and regular dimension; the fourth-order tensor structure is coupled between channels to obtain a structured feature tensor, and separable convolution is applied to perform multi-scale projection on the structured feature tensor to generate a spatial regular feature tensor.

[0043] Next, in the contract structure decoding stage, Tucker decomposition is used to perform hierarchical decomposition on the spatial rule feature tensor. Furthermore, tensor expansion and matrix reorganization are performed on the spatial rule feature tensor to obtain an expanded multidimensional matrix. Least squares projection is performed on the expanded multidimensional matrix to obtain a core tensor and a factor matrix. The core tensor retains the potential structure of the spatial rules, and the factor matrix encodes the association weights of each dimension (space / time / rules). The core tensor and the factor matrix are subjected to feature reweighting through the attention mechanism to obtain a weighted feature representation. The weighted feature representation is then subjected to cross-dimensional feature fusion to obtain the intermediate representation of the contract decoding.

[0044] During the parameter filling phase, dynamic feature mapping is performed on the contract decoding intermediate representation, the contract decoding intermediate representation is converted to the parameter space, parameterized contract fields are generated, field alignment and parameter filling are performed on the parameterized contract fields, and the initial contract terms are obtained; the initial contract terms are structured and field optimization and reorganization are performed to generate a structured contract representation, the structured contract representation is serialized, and a low-altitude airspace compliant smart contract is output.

[0045] S2. Perform symbolic analysis and Kriging spatial interpolation on the low-altitude airspace compliance smart contract to obtain the airspace risk heat map. Simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix.

[0046] The specific steps include: S2.1. Use non-negative matrix decomposition to decompose the logical structure of the low-altitude airspace compliance smart contract to form a structured rule matrix. In the specific operation, the low-altitude airspace compliance smart contract is feature decoupled to obtain the core semantic features, and logical relationship mapping and multi-layer stacking are performed on the core semantic features to generate logical feature units; next, the logical feature units are dynamically projected and nonlinearly activated to generate a logical structure representation, and the logical structure representation is simultaneously iteratively optimized and non-negative decomposition is performed to extract the initial basic matrix and weight coefficient matrix; the initial basic matrix is ​​sparsely processed and orthogonally projected to form a rule base matrix, and the weight coefficient matrix is ​​subjected to outlier removal and normalization scaling to obtain a standardized weight matrix; semantic clustering and rule merging are performed on the standardized weight matrix and the rule base matrix to output a structured rule matrix.

[0047] S2.2. Use the entropy weight method to perform symbolic analysis and weight redistribution on the structured rule matrix to obtain the symbolic rule parameter distribution. In the specific operation, in the symbolic analysis stage, the dimensional features in the structured rule matrix are extracted, and the range calculation is performed on the dimensional features to generate the eigenvalue dispersion. The structured rule matrix is ​​sorted by importance according to the eigenvalue dispersion to obtain the feature sorting sequence. At the same time, the information entropy calculation is performed on the feature sorting sequence to obtain the information entropy value. The specific mathematical formula is as follows: ;

[0048] in, represents the information entropy value, The total number of dimensions representing the dimensional features, Represents the dimension index, Indicated in dimension The probability weight of

[0049] Min-Max normalization is used to dynamically scale the information entropy value to obtain the preliminary allocation weights, and semantic mapping is performed on the preliminary allocation weights to obtain symbolic weights; in the weight redistribution stage, the symbolic weights are hierarchically aggregated to obtain an optimized weight set. At the same time, based on the optimized weight set, weight redistribution and dimensionality compression are performed on the structured rule matrix to form symbolic rule parameters, and multidimensional space projection and density clustering are performed on the symbolic rule parameters to obtain the distribution of symbolic rule parameters.

[0050] S2.3. Perform kriging spatial interpolation and probability surface fitting on the symbolized rule parameter distribution to generate an airspace risk heat map. Specifically, in the spatial interpolation stage, perform feature decoupling on the symbolized rule parameter distribution, extract spatially related features, perform principal component decomposition on the spatially related features, generate the first principal component vector and the second principal component vector, and use the first principal component vector as the x-axis in the three-dimensional space and the second principal component vector as the y-axis in the three-dimensional space to construct a three-dimensional spatial interpolation network. Next, perform kriging spatial interpolation and neighborhood aggregation on the three-dimensional spatial interpolation network to obtain an initial interpolation result. Perform filtering smoothing and outlier removal on the initial interpolation result to obtain an optimized interpolation result.

[0051] In the surface fitting stage, kernel density estimation is used to perform probability density fitting on the optimized interpolation results. Furthermore, neighborhood weighted averaging is performed on the optimized interpolation results to generate an initial probability distribution, and the initial probability distribution is subjected to median filtering and surface fitting to obtain a probability density surface. At the same time, the probability density surface is scaled and coordinates transformed to form a standard thermal layer. Layer overlay is performed on the standard thermal layer to obtain an airspace risk heat map.

[0052] S2.4. Perform wavelet transform and multi-scale decomposition on the airspace risk heat map to obtain a multi-resolution feature map sequence, and perform multi-layer pyramid convolution on the resolution feature map sequence to obtain a multi-level convolution feature map. In the specific operation, perform wavelet transform on the airspace risk heat map, further perform time-frequency decomposition on the airspace risk heat map to obtain a wavelet coefficient matrix, perform coefficient normalization and feature reorganization on the wavelet coefficient matrix to obtain a reorganized feature representation; then perform multi-scale decomposition on the reorganized feature representation, further perform smoothing filtering and edge enhancement through low-pass and high-pass filtering respectively, and extract the approximate component and detail component in the reorganized feature representation; perform feature weighted fusion on the approximate component and detail component to obtain a fused feature map, perform bilinear interpolation and resolution alignment on the fused feature map, and generate a multi-resolution feature map sequence;

[0053] Subsequently, a three-layer pyramid structure is applied to perform multi-layer pyramid convolution on the multi-resolution feature map sequence. Furthermore, the first-layer pyramid structure performs 5×5 separable convolution on the multi-resolution feature map sequence, and gradually expands the receptive field through three layers of downsampling to output the primary convolution feature map; the second-layer pyramid structure uses void convolution to further expand the receptive field of the primary convolution feature map, obtains the enhanced feature map, and performs feature normalization on the enhanced feature map to obtain the intermediate convolution feature map; the third-layer pyramid structure performs 1×1 point convolution on the intermediate convolution feature map to obtain the advanced convolution feature map, and performs cross-level feature fusion on the advanced convolution feature map to output the multi-level convolution feature map.

[0054] S2.5. Perform channel attention weighting and cross-scale feature fusion on the multi-level convolution feature map to form a fused feature tensor, perform linear projection on the fused feature tensor, and output the spatial risk value matrix. In the specific operation, in the channel attention weighting stage, use global average pooling to perform channel feature compression and linear transformation on the multi-level convolution feature map to obtain the channel attention weight; according to the channel attention weight, perform Sigmoid activation and channel weighted multiplication on the multi-level convolution feature map to generate a channel enhanced feature map; then perform bilinear upsampling and resolution alignment on the channel enhanced feature map to obtain a standardized feature map, and perform cross-level feature splicing on the standardized feature map to generate a weighted multi-scale feature tensor;

[0055] In the cross-scale feature fusion stage, cross-scale fusion and residual connection are performed on the weighted multi-scale feature tensor to obtain preliminary fusion features. At the same time, batch normalization and LeakyReLU activation are performed on the preliminary fusion features to generate optimized fusion features. The optimized fusion features are reorganized in channel dimensions and feature dimensionality reduction to obtain the final fusion feature tensor.

[0056] In the linear projection stage, 1×1 convolution is applied to perform channel dimensionality reduction and probability mapping on the final fused feature tensor, outputting the normalized risk probability, performing feature space projection on the normalized risk probability, and generating an initial risk matrix; edge smoothing and noise suppression are performed on the initial risk matrix to obtain the optimized risk matrix, and probability normalization and physical dimension conversion are performed on the optimized risk matrix to form a spatial risk value matrix.

[0057] S3. Perform conditional mask calculation on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, perform dynamic key sharding management on the low-altitude airspace compliance smart contract to generate partitioned encrypted data streams.

[0058] The specific steps include: S3.1. Perform risk quantification and conditional mask calculation on the spatial risk value matrix to generate a binary mask; perform spatial sliding window statistics on the binary mask to obtain the violation probability distribution. In the specific operation, in the risk quantification stage, the spatial risk value matrix is ​​numerically normalized to obtain the risk value; the spatial risk value matrix is ​​divided into regions using the segmentation threshold. For example, when the risk value is lower than the segmentation threshold, it is divided into a safe area (set to 0); when the risk value is higher than the dynamic segmentation threshold, it is divided into a violation area (set to 1). The result of the division is subjected to matrix filling and edge mirroring extension to output a preliminary binary mask; then the preliminary binary mask is subjected to conditional mask calculation, and further, a 3×3 circular structure element matrix is ​​used to perform a closed operation on the preliminary binary mask, and the closed operation result value is output; then an open operation is performed on the closed operation result value through a 5×5 rectangular kernel matrix to eliminate isolated noise points and obtain a binary mask. The specific mathematical formula is as follows, ;

[0059] in, represents a binary mask, represents the preliminary binary mask, represents the circular structure element matrix, represents a rectangular kernel matrix, represents the closing operation, Indicates open operation;

[0060] It should be noted that the circular structure element matrix refers to the 3×3 discretized disk kernel matrix used to eliminate holes filled by closing operations; the rectangular kernel matrix refers to the 5×5 all-one matrix used for opening operations to eliminate noise; the segmentation threshold is defined based on the statistical distribution characteristics of the risk value and has a value range of [0.15, 0.85].

[0061] Then, spatial sliding window statistics are performed on the binary mask. Furthermore, a 50×50 pixel square window is set, and the binary mask is slid with a step size of 25 pixels to output the window statistical record of the binary mask. Mean filtering is used to spatially smooth and aggregate the window statistical record of the binary mask to obtain the initial probability distribution. Probability normalization and regional clustering are performed on the initial probability distribution to generate a regional probability map. At the same time, bilinear interpolation is performed on the regional probability map to obtain the violation probability distribution.

[0062] S3.2. Dynamically adjust the permissions of the low-altitude airspace compliance smart contract based on the violation probability distribution to form encryption control parameters, perform dynamic key sharding management on the encryption control parameters, and generate partitioned encrypted data streams. In the specific operation, during the permission adjustment stage, perform numerical mapping on the violation probability distribution and map the violation probability distribution to three numerical intervals: [0, 0.3), [0.3, 0.6), and [0.6, 1.0]. Among them, the interval [0, 0.3) corresponds to level 0 permission (fully open), the interval [0.3, 0.6) corresponds to level 1 permission (approval required), and the interval [0.6, 1.0] corresponds to level 1 permission (approval required). The .0] interval corresponds to level 2 authority (full prohibition); according to the authority level, the authority of the low-altitude airspace compliance smart contract is dynamically adjusted. For example, at level 0 authority, the low-altitude airspace compliance smart contract is adjusted to authorize passage; at level 1 authority, the low-altitude airspace compliance smart contract is adjusted to manually review and approve; at level 2 authority, the low-altitude airspace compliance smart contract is adjusted to intercept and prohibit. During the adjustment process, the contract state variables, timestamps, and spatial coordinates are recorded as dynamic control parameters, and asymmetric encryption is performed on the dynamic control parameters to obtain encrypted control parameters;

[0063] In the key sharding management stage, the encryption control parameters are randomly sampled and parameter aggregation is performed to obtain the key encryption sharding set; the key encryption sharding set is distributedly stored to obtain the sharding storage certificate; the sharding storage certificate is stored on the blockchain and broadcast synchronously to the terminal to obtain a distributed encryption record; the distributed encryption record is encapsulated in streaming data to generate a partitioned encrypted data stream.

[0064] S4. Use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain the blockchain evidence record, perform zero-knowledge audit proof on the blockchain evidence record, and output the data management audit report.

[0065] The specific steps include: S4.1. Use the secure hash algorithm to perform coordinate binding and spatiotemporal tag encoding on the partitioned encrypted data stream, obtain the spatiotemporal tag hash sequence, perform hierarchical aggregation on the spatiotemporal tag hash sequence, and generate blockchain evidence records. In the specific operation, in the coordinate binding stage, perform spatial feature extraction on the partitioned encrypted data stream to obtain the spatial feature vector; apply the PTP timestamp synchronization protocol to perform spatiotemporal alignment on the spatial feature vector to obtain spatiotemporal alignment data, synchronously perform coordinate projection transformation on the spatiotemporal alignment data to obtain high-precision coordinate data, and embed the high-precision coordinate data into the partitioned encrypted data stream to obtain coordinate binding data; in the spatiotemporal tag encoding stage, perform coordinate binding on the coordinate binding. The data is divided into time and space dimensions to obtain time and space shard data; hash field mapping and sequence normalization are performed on the time and space shard data to generate a primary hash sequence; timestamps are embedded in the primary hash sequence to output a time and space tag hash sequence; in the blockchain evidence stage, hierarchical aggregation processing is performed on the time and space tag hash sequence: the first layer performs period bucketing and hash aggregation on the time and space tag hash sequence according to the time dimension to obtain a daily time aggregate root; the second layer performs block division and spatial fusion on the daily time aggregate root according to the space dimension to obtain a time and space aggregation certificate; the third layer writes the time and space aggregation certificate and the daily time aggregate root into the Ethereum blockchain to generate a package blockchain evidence record.

[0066] S4.2. Perform spatiotemporal feature extraction and reorganization on blockchain evidence records to form an audit evidence set, perform zero-knowledge audit proof on the audit evidence set, and obtain a verifiable statement. Specifically, in the feature extraction and reorganization stage, perform random stratified sampling on the blockchain evidence records to obtain the original evidence data, perform timestamp extraction and coordinate decoding on the original evidence data to obtain spatiotemporal data tuples; perform multidimensional decomposition on the spatiotemporal data tuples to obtain spatiotemporal feature vectors; perform feature reorganization on the spatiotemporal feature vectors to generate a standardized feature matrix; perform feature weight assignment and weighted aggregation on the standardized feature matrix to output the audit evidence set; in the zero-knowledge proof stage, use the zk-SNARKs protocol to perform key mapping on the audit evidence set to obtain the proof key, perform key pair separation on the proof key, and generate the verification key; perform zero-knowledge proof and statement encapsulation on the proof key and verification key to output a verifiable statement;

[0067] Perform data desensitization and format conversion on verifiable statements to output standard audit statements; securely encapsulate standard audit statements to generate audit evidence packages; structure and integrate audit evidence packages to output data management audit reports.

[0068] This embodiment also provides a blockchain-based low-altitude economic data management system, including: a contract generation module for inputting a multi-dimensional low-altitude economic data set into a spatiotemporal fusion network model; an airspace relationship analysis layer for performing topological relationship analysis and neighborhood feature aggregation; a spatiotemporal interaction layer for performing cross-dimensional feature interaction and outputting a low-altitude airspace compliance smart contract; The risk assessment module is used to perform symbolic analysis and Kriging spatial interpolation on the low-altitude airspace compliance smart contract to obtain the airspace risk heat map, and simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix; The security encryption module is used to perform conditional mask calculations on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, it dynamically manages the key sharding of the low-altitude airspace compliance smart contract and generates partitioned encrypted data streams. The audit evidence module is used to use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain blockchain evidence records, perform zero-knowledge audit proof on the blockchain evidence records, and output data management audit reports.

[0069] This embodiment also provides a computer device, which is suitable for the low-altitude economic data management method based on blockchain, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the low-altitude economic data management method based on blockchain proposed in the above embodiment.

[0070] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0071] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the low-altitude economic data management method based on blockchain as proposed in the above embodiment; the storage medium 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.

[0072] In summary, this invention significantly improves its adaptability to dynamically changing environments and aircraft behaviors by using a spatiotemporal fusion network model to process and analyze multidimensional low-altitude economic data in real time. It also utilizes a secure hash algorithm to perform spatiotemporal tag encoding of partitioned encrypted data streams, protecting the data from unauthorized access and tampering, significantly enhancing the security and privacy of low-altitude economic data.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A low-altitude economic data management method based on blockchain, characterized by: include, The multi-dimensional low-altitude economic data set is input into the spatiotemporal fusion network model. The airspace relationship analysis layer performs topological relationship analysis and neighborhood feature aggregation. The spatiotemporal interaction layer performs cross-dimensional feature interaction and outputs a low-altitude airspace compliance smart contract. Perform symbolic analysis and Kriging spatial interpolation on low-altitude airspace compliance smart contracts to obtain an airspace risk heat map. Simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix. Perform conditional mask calculation on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, perform dynamic key sharding management on the low-altitude airspace compliance smart contract to generate partitioned encrypted data streams. Use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain the blockchain evidence record, perform zero-knowledge audit proof on the blockchain evidence record, and output the data management audit report.

2. The low-altitude economic data management method based on blockchain according to claim 1, characterized in that: The multi-dimensional low-altitude economic data set includes aircraft dynamic data, low-altitude airspace environment data and low-altitude airspace compliance rule data.

3. The low-altitude economic data management method based on blockchain according to claim 2, characterized in that: The output of the low-altitude airspace compliance smart contract specifically includes the following steps: Build a spatial relationship analysis layer and a spatiotemporal interaction layer, apply skip connections to perform multi-scale stacking, and construct a spatiotemporal fusion network model; The multi-dimensional low-altitude economic data set is input into the spatiotemporal fusion network model. The spatial relationship analysis layer uses graph convolution to perform topological relationship analysis and neighborhood feature aggregation to generate the spatial node state vector. The spatiotemporal interaction layer applies a multi-head attention mechanism to perform cross-dimensional feature interaction and obtain the spatiotemporal correlation weight matrix; A gating unit is used to perform importance weighting and cross-modal feature fusion on the spatial node state vector and weight matrix to form a spatial regular feature tensor; The contract structure is decoded and parameters are filled in for the airspace rule feature tensor, and a low-altitude airspace compliance smart contract is output.

4. The low-altitude economic data management method based on blockchain as claimed in claim 3 is characterized by: The acquisition of the airspace risk heat map specifically includes the following steps: Decompose the logical structure of the low-altitude airspace compliance smart contract to form a structured rule matrix; The entropy weight method is used to symbolically analyze and redistribute the weights of the structured rule matrix to obtain the symbolic rule parameter distribution; The symbolized rule parameter distribution is subjected to Kriging spatial interpolation and probability surface fitting to generate an airspace risk heat map.

5. The low-altitude economic data management method based on blockchain according to claim 4 is characterized in that: The formation of the airspace risk value matrix specifically includes the following steps: The airspace risk heat map is subjected to wavelet transformation and multi-scale decomposition to obtain a multi-resolution feature map sequence, and the resolution feature map sequence is subjected to multi-layer pyramid convolution to obtain a multi-level convolution feature map; Channel attention weighting and cross-scale feature fusion are performed on the multi-level convolutional feature map to form a fused feature tensor, which is linearly projected to output the spatial risk value matrix.

6. The low-altitude economic data management method based on blockchain according to claim 5, characterized in that: The generating of partition encrypted data stream specifically includes the following steps: Perform risk quantification and conditional mask calculation on the spatial risk value matrix to generate a binary mask; Perform spatial sliding window statistics on the binary mask to obtain the violation probability distribution; According to the probability distribution of violations, the permissions of the low-altitude airspace compliance smart contract are dynamically adjusted to form encryption control parameters, and the encryption control parameters are dynamically key-sharded to generate partitioned encrypted data streams.

7. The low-altitude economic data management method based on blockchain according to claim 6, characterized in that: The output data management audit report specifically includes the following steps: Use the secure hash algorithm to perform coordinate binding and space-time tag encoding on the partitioned encrypted data stream, obtain the space-time tag hash sequence, perform hierarchical aggregation on the space-time tag hash sequence, and generate blockchain evidence records; Extract and reconstruct spatiotemporal features of blockchain evidence records to form an audit evidence set, perform zero-knowledge audit proof on the audit evidence set, and obtain verifiable statements; Perform structured integration of verifiable claims and output data management audit reports.

8. A blockchain-based low-altitude economic data management system, based on the blockchain-based low-altitude economic data management method according to any one of claims 1 to 7, characterized in that: include, The contract generation module is used to input the multi-dimensional low-altitude economic data set into the spatiotemporal fusion network model. The airspace relationship analysis layer performs topological relationship analysis and neighborhood feature aggregation. The spatiotemporal interaction layer performs cross-dimensional feature interaction and outputs the low-altitude airspace compliance smart contract. The risk assessment module is used to perform symbolic analysis and Kriging spatial interpolation on the low-altitude airspace compliance smart contract to obtain the airspace risk heat map, and simultaneously perform multi-layer pyramid convolution on the airspace risk heat map to form an airspace risk value matrix; The security encryption module is used to perform conditional mask calculations on the airspace risk value matrix to obtain the violation probability distribution. Based on the violation probability distribution, it dynamically manages the key sharding of the low-altitude airspace compliance smart contract and generates partitioned encrypted data streams. The audit evidence module is used to use the secure hash algorithm to perform time-space tag encoding on the partitioned encrypted data stream, obtain blockchain evidence records, perform zero-knowledge audit proof on the blockchain evidence records, and output data management audit reports.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the blockchain-based low-altitude economic data management method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the blockchain-based low-altitude economic data management method according to any one of claims 1 to 7 are implemented.

Images