A method and system for encrypted transmission of hydrological data

Abstract

This invention provides a method and system for encrypted transmission of hydrological data. The method includes: acquiring hydrological data to be transmitted, including data packets; extracting temporal and spatial correlation features from the data packets to obtain multidimensional hydrological feature vectors; acquiring nodes in the transmission path and constructing connection edges between nodes; assigning dynamic security weights to the connection edges based on the multidimensional hydrological feature vectors; converting the node's own features into updated features according to the dynamic security weights; inputting the updated features into a parameter generation model to dynamically obtain an adaptive encryption parameter set for layered hybrid encryption of the hydrological data, generating and transmitting encrypted data packets. By introducing the spatiotemporal characteristics of the hydrological data itself into the security evaluation chain and aggregating contextual information between nodes, a mechanism is established to generate customized encryption strategies in real time and dynamically based on data content, network status, and topology, thereby changing the statically configured encryption mode.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for encrypted transmission of hydrological data. Background Technology

[0002] Hydrological data, such as water level, flow velocity, flow rate, and water quality parameters, form the basis for decision-making in water resource management, flood warning, water conservancy project scheduling, and environmental protection.

[0003] Hydrological data is typically collected by telemetry stations and sensor nodes (such as IoT nodes) distributed over a wide geographical area and transmitted to data centers via wireless or wired networks.

[0004] However, hydrological data is highly sensitive and valuable, and it needs to be encrypted during transmission. However, most existing encryption transmission schemes are statically configured and cannot dynamically adjust security strategies according to the context of data transmission (such as data type and the spatiotemporal correlation characteristics of the data itself). For example, performing high-strength but time-consuming asymmetric encryption when the channel quality is poor may lead to data packet loss or increased delay. Furthermore, if the security strength is insufficient when transmitting critical early warning data, the risk is extremely high. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for encrypted transmission of hydrological data, which solves the technical problem that most existing encrypted transmission schemes are statically configured and cannot dynamically adjust the transmission security strategy according to the context of data transmission.

[0006] To achieve the above objectives, in a first aspect, embodiments of this application provide a method for encrypted transmission of hydrological data, comprising the following steps:

[0007] The process involves acquiring hydrological data to be transmitted, which includes data packets from several stations and each data packet includes at least one sequence of hydrological parameters. Temporal features are extracted from the data packets, and spatial correlation features are extracted from the data packets. The temporal features and spatial correlation features are then combined into a multidimensional hydrological feature vector.

[0008] The data source node, routing relay node, and target data center node of the site are mapped to nodes, and connection edges between nodes are constructed. Dynamic security weights are assigned to the connection edges based on the multidimensional hydrological feature vector.

[0009] Obtain the self-feature corresponding to the node, select one of its nodes as the calibration node, and select the self-feature corresponding to the calibration node as the calibration feature. Select the nodes that have a connection edge with the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features. Based on the calibration feature, the neighboring features and the dynamic security weight, obtain the updated feature corresponding to the calibration node.

[0010] The updated feature input parameters are used to generate a model to dynamically obtain an adaptive encryption parameter set. Based on the adaptive encryption parameter set, the hydrological data is subjected to layered hybrid encryption to form an encrypted data packet, which is then transmitted.

[0011] Furthermore, the hydrological parameter sequence includes several preprocessed data points under a time series, and the time series features include the sliding window mean, trend slope, and approximate entropy. The step of extracting time series features from the data packet includes:

[0012] Set a sliding window of a preset size, select the preprocessed data points of the current time frame as calibration points, and select several preprocessed data points adjacent to the calibration points as reference points based on the sliding window. Perform mean processing on the calibration points and several reference points to obtain the mean of the sliding window.

[0013] The hydrological parameter sequence is linearly fitted to obtain the trend slope;

[0014] A first comparison length, a second comparison length, and a similarity threshold are set. The difference between the second comparison length and the first comparison length is 1. Based on the first comparison length and the second comparison length, the hydrological parameter sequence is divided into several first vector pairs and several second vector pairs. The first vector pair and the similarity threshold are used to obtain the first vector comparison mean, and the second vector pair and the similarity threshold are used to obtain the second vector comparison mean. The first vector comparison mean and the second vector comparison mean are used to obtain the approximate entropy. The sliding window mean, the trend slope, and the approximate entropy are concatenated to form a time-series feature.

[0015] Furthermore, the formula for obtaining the trend slope is:

[0016] ,

[0017] in, Indicates the slope of the trend. This represents the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of time frames in a hydrological parameter sequence. This represents the preprocessed data point corresponding to the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of the preprocessed data points in the hydrological parameter sequence.

[0018] Furthermore, the step of obtaining the average of the first vector comparisons based on several first vector pairs and the similarity threshold includes:

[0019] Select one of the first vector pairs as the calibration vector pair and select all the first vector pairs as the reference vector pairs to obtain the Chebyshev distance between the calibration vector pair and the reference vector pair.

[0020] The Chebyshev distance is compared with the similarity threshold to obtain the number of similar reference vector pairs whose Chebyshev distance is less than the similarity threshold. The similarity ratio is obtained by comparing the number of similar vector pairs with the total number of the first vector pairs.

[0021] The average value of all the similarity percentages is averaged to obtain the first vector comparison mean.

[0022] Furthermore, the step of extracting spatial correlation features from the data packet includes:

[0023] Obtain the geographical coordinates of the site, obtain the Euclidean distance between the two sites based on the geographical coordinates, and determine the association weight between the two sites based on the Euclidean distance;

[0024] The Pearson correlation coefficient between the two stations was obtained based on the hydrological parameter sequence.

[0025] The spatial correlation features corresponding to the data packet are obtained by using the Pearson correlation coefficient and the correlation weight.

[0026] Furthermore, the formula for obtaining the spatial association features is as follows:

[0027] ,

[0028] in, This represents the spatial association characteristics of data packets from the a-th site. This represents the association weight between data packets from site a and site b, and , This represents the Euclidean distance between station a and station b. This represents the Pearson correlation coefficient between data packets from site a and site b. This indicates the total number of sites.

[0029] Furthermore, the step of assigning dynamic security weights to the connecting edges based on the multidimensional hydrological feature vector includes:

[0030] The two nodes connected to the connection edge are selected as the sending node and the receiving node, respectively. The security score of the sending node and the importance score of the receiving node are obtained, and the network score between the sending node and the receiving node is obtained.

[0031] The data sensitivity score of the multidimensional hydrological feature vector at the sending node is obtained, and a dynamic security weight is constructed by the security score, the importance score, the network score, and the data sensitivity score.

[0032] Furthermore, the formula for obtaining the updated features is:

[0033] ,

[0034] in, This represents the update feature when the p-th node is the calibration node. This represents the calibration feature when the p-th node is a calibration node. This represents the proximity characteristics of the q-th neighboring node when the p-th node is the labeled node. This represents the total number of neighboring nodes when the p-th node is the labeled node. This represents the attention coefficient between the p-th node and its corresponding q-th neighbor when the p-th node is the calibration node. , Both represent weight matrices. This represents a non-linear activation function.

[0035] Furthermore, the adaptive encryption parameter set includes encryption algorithm type, operation mode, key length, and encapsulation mechanism. The step of performing layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet includes:

[0036] The adaptive encryption parameter set is used as a plaintext attribute, and the attribute-based encryption method is used to encrypt the adaptive encryption parameter set into a parameter ciphertext header.

[0037] A session key is randomly generated based on the key length, and the hydrological data is encrypted into ciphertext using the session key, the encryption algorithm type, and the operation mode.

[0038] The session key is encrypted based on the encapsulation mechanism and the public key of the target node to generate an encryption key. The parameter ciphertext header, the data ciphertext, and the encryption key are then encapsulated into an encrypted data packet.

[0039] Secondly, embodiments of this application provide a hydrological data encryption transmission system, applied to the hydrological data encryption transmission method described in the first aspect above, the system comprising:

[0040] The acquisition module is used to acquire hydrological data to be transmitted, the hydrological data including data packets from several stations, the data packets including at least one sequence of hydrological parameters, extracting temporal features from the data packets, extracting spatial correlation features from the data packets, and combining the temporal features and the spatial correlation features into a multidimensional hydrological feature vector.

[0041] The construction module is used to map the data source nodes, routing relay nodes and target data center nodes of the site to nodes, and to construct the connection edges between the nodes, and to assign dynamic security weights to the connection edges based on the multidimensional hydrological feature vector.

[0042] The aggregation module is used to obtain the self-feature corresponding to the node, select one of the nodes as the calibration node, select the self-feature corresponding to the calibration node as the calibration feature, select the nodes with connected edges to the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features, and obtain the updated feature corresponding to the calibration node based on the calibration feature, the neighboring feature and the dynamic security weight.

[0043] The execution module is used to generate a model from the updated feature input parameters to dynamically obtain an adaptive encryption parameter set, perform layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet, and transmit the encrypted data packet.

[0044] Thirdly, embodiments of this application provide a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the hydrological data encryption transmission method as described in the first aspect above.

[0045] Fourthly, embodiments of this application provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the hydrological data encryption transmission method as described in the first aspect above.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: by acquiring a multi-dimensional hydrological feature vector including the temporal features and the spatial correlation features, the attributes of hydrological data in time and space are quantified, and based on this, a subsequent encryption strategy is acquired, making the output of the encryption strategy strongly correlated with the spatiotemporal attributes of the hydrological data; by updating the node features based on the dynamic security weights obtained from the multi-dimensional hydrological feature vectors, the updated features are obtained and applied to the subsequent acquisition of the adaptive encryption parameter set, realizing networked and intelligent context-aware decision-making. The updated features are composite context representations that aggregate different node states, ensuring that the adaptive encryption parameter set is the optimal encryption strategy dynamically acquired after comprehensively considering complete context information, rather than a preset fixed configuration. Attached Figure Description

[0047] Figure 1 This is a flowchart of the hydrological data encryption transmission method in the first embodiment of the present invention;

[0048] Figure 2 This is a structural block diagram of the hydrological data encryption transmission system in the second embodiment of the present invention;

[0049] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0050] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0051] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0053] Please see Figure 1The hydrological data encryption transmission method provided in the first embodiment of the present invention includes the following steps:

[0054] S10: Obtain hydrological data to be transmitted, the hydrological data including data packets from several stations, the data packets including at least one sequence of hydrological parameters, extract temporal features from the data packets, extract spatial correlation features from the data packets, and combine the temporal features and the spatial correlation features into a multidimensional hydrological feature vector.

[0055] The hydrological parameter sequence includes water level, flow velocity, flow rate, water temperature, pH, turbidity, or conductivity values. It should be noted that to ensure data accuracy, the hydrological parameter sequence needs to undergo Kalman filtering and outlier removal. The specific processing procedures are commonly used in data processing and will not be elaborated here. Taking water level as an example, because the sensor continuously collects data, the water level values ​​from several time series within a given time interval constitute a water level sequence. Therefore, the hydrological parameter sequence includes preprocessed data points from several time series.

[0056] Step S10 includes:

[0057] S110: Set a sliding window of a preset size, select the preprocessed data points of the current time frame as calibration points, and select several preprocessed data points adjacent to the calibration points as reference points based on the sliding window, and perform mean processing on the calibration points and several reference points to obtain the mean of the sliding window.

[0058] It should be noted that the sliding window is smaller than the total number of time frames in the hydrological parameter sequence. Assuming there are 12 time frames, the size of the sliding window is less than 12. Taking the 12th time frame as the current time frame, the size of the sliding window is 6. Then, the preprocessed data points under the 7th, 8th, 9th, 10th, and 11th time frames are selected as reference points, and the preprocessed data points under the 7th to 12th time frames are averaged to form the sliding window mean. Taking the water level value as an example, it reflects the local average state of the water level under the 12th time frame, which is used to smooth short-term fluctuations and reflect trends.

[0059] S120: Perform linear fitting on the hydrological parameter sequence to obtain the trend slope;

[0060] The formula for obtaining the trend slope is:

[0061] ,

[0062] in, Indicates the slope of the trend. This represents the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of time frames in a hydrological parameter sequence. This represents the preprocessed data point corresponding to the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of preprocessed data points in a hydrological parameter sequence. Taking water level as an example, it reflects the rate of rise / fall of water level over a time interval.

[0063] S130: Set a first comparison length, a second comparison length, and a similarity threshold. The difference between the second comparison length and the first comparison length is 1. Based on the first comparison length and the second comparison length, the hydrological parameter sequence is divided into several first vector pairs and several second vector pairs. Based on the several first vector pairs and the similarity threshold, a first vector comparison mean is obtained. Based on the several second vector pairs and the similarity threshold, a second vector comparison mean is obtained. Based on the first vector comparison mean and the second vector comparison mean, an approximate entropy is obtained. The sliding window mean, the trend slope, and the approximate entropy are concatenated into a time-series feature.

[0064] In this embodiment, the first comparison length is 1, the second comparison length is 2, and the similarity threshold is 0.2. Assuming the hydrological parameter sequence is [1.0, 1.5, 1.2, 1.7, 1.3], after being separated by the first comparison length, several first vector pairs are respectively: [1.0, 1.5], [1.5, 1.2], [1.2, 1.7], [1.7, 1.3].

[0065] Select one of the first vector pairs as the calibration vector pair and select all the first vector pairs as the reference vector pairs to obtain the Chebyshev distance between the calibration vector pair and the reference vector pair.

[0066] First, [1.0, 1.5] is selected as the calibration vector pair, and [1.0, 1.5], [1.5, 1.2], [1.2, 1.7], and [1.7, 1.3] are all selected as reference vector pairs. The Chebyshev distance between [1.0, 1.5] and [1.5, 1.2] is max(|1.0-1.5|, |1.5-1.2|) = 0.5.

[0067] The Chebyshev distance is compared with the similarity threshold to obtain the number of similar reference vector pairs whose Chebyshev distance is less than the similarity threshold. The similarity ratio is obtained by comparing the number of similar vector pairs with the total number of the first vector pairs.

[0068] The average value of all the similarity percentages is averaged to obtain the average value of the first vector comparison.

[0069] Taking the aforementioned five first vector pairs as an example, the number of reference vector pairs with a similarity less than 0.2 is two, meaning the similarity count is two, and the similarity percentage is 50%. Performing the above operation on all first vector pairs yields four similarity percentages, which are then averaged to obtain the first vector comparison mean. The methods for obtaining the second vector pairs and the second vector comparison mean are the same and will not be repeated here. After obtaining the first and second vector comparison mean values, the difference between them forms the approximate entropy.

[0070] S140: Obtain the geographical coordinates of the site, obtain the Euclidean distance between the two sites based on the geographical coordinates, and determine the association weight between the two sites based on the Euclidean distance;

[0071] The calculation of Euclidean distance is commonly used and will not be elaborated here. The formula for obtaining the association weight is:

[0072] ,

[0073] in, This represents the association weight between data packets from site a and site b. This represents the Euclidean distance between station a and station b.

[0074] S150: Obtain the Pearson correlation coefficient between the two stations based on the hydrological parameter sequence;

[0075] The mean values ​​of the hydrological parameter sequences from the two stations are obtained respectively. It should be noted that the number of time frames in the hydrological parameter sequences from both stations is the same, and there is a one-to-one correspondence between the preprocessed data points of the two stations under each time frame. The Pearson correlation coefficient is obtained based on the mean values. The formula for obtaining the Pearson correlation coefficient is as follows:

[0076] ,

[0077] in, This represents the Pearson correlation coefficient between data packets from site a and site b. This represents the preprocessed data point of the a-th station in the x-th time frame. This represents the series mean of the hydrological parameter sequence for station a. This represents the preprocessed data point of the b-th station in the x-th time frame. This represents the series mean of the hydrological parameter sequence at station b. This indicates the total number of time frames.

[0078] S160: Obtain the spatial correlation features corresponding to the data packet using the Pearson correlation coefficient and the correlation weight;

[0079] The formula for obtaining the spatial association features is:

[0080] ,

[0081] in, This represents the spatial association characteristics of data packets from the a-th site. This indicates the total number of sites.

[0082] S20: Map the data source node, routing relay node and target data center node of the site to nodes, and construct the connection edges between the nodes. Based on the multi-dimensional hydrological feature vector, network status information and preset security policy, assign dynamic security weights to the connection edges to form a dynamic security graph.

[0083] Understandably, the data source node of the site is the sensor, and the target data center node is the data center that receives the data. Taking one data source node, two routing relay nodes and one target data center node as an example, the connection edge is constructed as follows: data source node → routing relay node 1 → routing relay node 2 → target data center node.

[0084] Step S20 includes:

[0085] S210: Select the two nodes connected to the connection edge as the sending node and the receiving node respectively, obtain the security score of the sending node and the importance score of the receiving node, and obtain the network score between the sending node and the receiving node.

[0086] In this embodiment, the security score is based on the device's static attributes and dynamic historical data. Specifically, information is extracted from the device's factory-preset security chip level, firmware version number, and security monitoring logs (such as recent login failures and abnormal access records), and the security score is calculated using a predefined scoring rule table. The importance score is a preset value that can be set according to the node's attributes. In this embodiment, the importance score for the target data center node is 1.0, the importance score for the routing relay node is 0.6, and the importance score for the data source node is 0.2. The network score is based on the security and quality status of the link layer. Specifically, the network interface card or communication module driver obtains information such as whether encryption (e.g., AES) is enabled for the current transmission and its encryption strength, the real-time monitoring channel's bit error rate, signal strength, and transmission delay, and the network score is obtained using a predefined scoring rule table.

[0087] S220: Obtain the data sensitivity score of the multidimensional hydrological feature vector at the sending node, and construct a dynamic security weight through the security score, the importance score, the network score and the data sensitivity score;

[0088] In this embodiment, the multidimensional hydrological feature vector is scored based on a preset scoring rule table. Specifically, for features including sliding window mean, trend slope, approximate entropy, and spatial correlation, corresponding scoring segments are set. Taking the trend slope as an example, if the trend slope is greater than 0.8, the score contribution is 0.4; if the trend slope is less than or equal to 0.8 but greater than 0.6, the score contribution is 0.2; and if the trend slope is less than or equal to 0.6, the score contribution is 0.0. Then, all the score contributions are summed to obtain the data sensitivity score.

[0089] The formula for obtaining the dynamic security weight is:

[0090] ,

[0091] in, This represents the dynamic safety weight of the edge connecting node e to node f, where node e is the sending node and node f is the receiving node. This represents the security score when the e-th node is the sending node. This represents the network score between node e, which is the sending node, and node f, which is the receiving node. This represents the importance score when the f-th node is the receiving node. This indicates that the e-th node is the sending node, and the multidimensional hydrological feature vector is used to score the data sensitivity of the sending node. , , , Both represent weighting coefficients, and + + + =1. It should be noted that the content of the multidimensional hydrological feature vector may change during the transmission path. A routing relay node may receive hydrological data from multiple data source nodes, aggregate and average them to generate a new aggregated data, and then send it to the next hop. The multidimensional hydrological feature vector of this new aggregated data is different from the original multidimensional hydrological feature vector, so the calculated data sensitivity score will also change.

[0092] S30: Obtain the self-feature corresponding to the node, select one of its nodes as the calibration node, select the self-feature corresponding to the calibration node as the calibration feature, select the nodes with connected edges to the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features, and obtain the updated feature corresponding to the calibration node based on the calibration feature, the neighboring feature and the dynamic security weight;

[0093] The self-characteristics include node type encoding, security score, importance score, node resource status, and data sensitivity score. The security score, importance score, and data sensitivity score can be obtained based on step S20. The node type encoding is a static attribute of the node and can be defined according to the type of the node, such as 0.1 for a data source node, 0.5 for a routing relay node, and 0.9 for a target data center node. The node resource status reflects the current available resources of the node, such as node resource status = 0.7 * power percentage + 0.3 * (1 - CPU utilization).

[0094] The formula for obtaining the updated features is:

[0095] ,

[0096] in, This represents the update feature when the p-th node is the calibration node. This represents the calibration feature when the p-th node is a calibration node. This represents the proximity characteristics of the q-th neighboring node when the p-th node is the labeled node. This represents the total number of neighboring nodes when the p-th node is the labeled node. This represents the attention coefficient between the p-th node and its corresponding q-th neighbor when the p-th node is the calibration node. , Both represent weight matrices. This represents a non-linear activation function. It should be noted that the node's own features are converted into mapped features using a linear transformation matrix. Then, the mapped features of the labeled node and its neighboring nodes are concatenated into vectors. The concatenated vector is then multiplied by the learnable vector to obtain a preliminary association score. Finally, the LeakyReLU activation function is used to convert the preliminary association score into attention coefficients.

[0097] S40: Generate a model by inputting the updated features into parameters to dynamically obtain an adaptive encryption parameter set, perform layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet, and transmit the encrypted data packet;

[0098] In this embodiment, the parameter generation model is a multilayer perceptron, which stores an encrypted database containing several encrypted configurations. Based on the input update features, the multilayer perceptron outputs the probability distribution corresponding to each encrypted configuration, and then selects the encrypted configuration with the highest probability distribution as the adaptive encryption parameter set. The application of multilayer perceptrons for probability prediction is already quite common, and will not be elaborated upon here.

[0099] The adaptive encryption parameter set includes encryption algorithm type, operation mode, key length, and encapsulation mechanism. For example, a certain adaptive encryption parameter set Se can be represented as {Alg (encryption algorithm type): AES-256-GCM, KeyLen (key length): 256, Mode (operation mode): GCM, KEM (encapsulation mechanism): RSA-OAEP-2048}.

[0100] Step S40 includes:

[0101] S410: Using the adaptive encryption parameter set as a plaintext attribute, encrypt the adaptive encryption parameter set into a parameter ciphertext header using an attribute-based encryption method;

[0102] Using attribute-based encryption methods ensures that only recipients who meet specific attribute policies can decrypt the data, such as those whose role is "data center". The purpose is to securely and verifiably transmit the encryption policy itself to authorized recipients. This ensures that the recipient knows exactly what encryption configuration the sender used to protect the data, which is a prerequisite for decryption.

[0103] S420: Randomly generate a session key based on the key length, and encrypt the hydrological data into ciphertext using the session key, the encryption algorithm type, and the operation mode;

[0104] The session key is dynamically generated each time encryption is performed. After the session key is generated, encryption is performed according to the specified encryption algorithm type and operation mode to ensure that the encryption strategy of the adaptive encryption parameter set is executed.

[0105] S430: Encrypt the session key based on the encapsulation mechanism and the public key of the target node to generate an encryption key, and encapsulate the parameter ciphertext header, the data ciphertext, and the encryption key into an encrypted data packet;

[0106] The target node refers to the data recipient in this data transmission. Taking the data source node → routing relay node as an example, the target node is the routing relay node. Attribute-based encryption is used to generate the parameter ciphertext header, symmetric encryption is used to protect the data body (generating ciphertext), and asymmetric encryption is used to protect the session key. This achieves a balance between security, flexibility, and efficiency. The attribute-based encryption method ensures the confidentiality of the encryption policy and fine-grained access control; symmetric encryption is highly efficient for processing large volumes of hydrological data; and asymmetric encryption securely distributes the session key, decoupling the secure transmission of dynamic policies, efficient data encryption, and secure key management. This allows the system to maintain overall security while focusing the main computation on efficient symmetric encryption when executing dynamically changing encryption policies.

[0107] By acquiring a multidimensional hydrological feature vector that includes the temporal and spatial correlation features, the temporal and spatial attributes of hydrological data are quantified. Based on this, subsequent encryption strategies are acquired, ensuring a strong correlation between the output of the encryption strategy and the spatiotemporal attributes of the hydrological data. Dynamic security weights obtained from the multidimensional hydrological feature vectors are used to update node features, resulting in updated features that are then applied to the acquisition of subsequent adaptive encryption parameter sets. This achieves networked and intelligent context-aware decision-making. The updated features are composite context representations aggregating different node states, ensuring that the adaptive encryption parameter set is the optimal encryption strategy dynamically acquired after comprehensively considering complete context information, rather than a preset fixed configuration. In other words, by introducing the spatiotemporal characteristics of the hydrological data itself into the security evaluation chain and aggregating contextual information between nodes, a mechanism is established to generate customized encryption strategies in real-time and dynamically based on data content, network status, and topology, thereby changing the statically configured encryption mode.

[0108] Please see Figure 2 The second embodiment of the present invention provides a hydrological data encryption transmission system, which is applied to the hydrological data encryption transmission method described in the above embodiments, and will not be repeated hereafter. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0109] The system includes:

[0110] The acquisition module 10 is used to acquire hydrological data to be transmitted. The hydrological data includes data packets from several stations. The data packets include at least one sequence of hydrological parameters. Temporal features are extracted from the data packets, and spatial correlation features are extracted from the data packets. The temporal features and the spatial correlation features are combined into a multidimensional hydrological feature vector.

[0111] The acquisition module 10 includes:

[0112] The first unit is used to set a sliding window of a preset size, select the preprocessed data points of the current time frame as calibration points, and select several preprocessed data points adjacent to the calibration points as reference points based on the sliding window, and perform mean processing on the calibration points and several reference points to obtain the mean of the sliding window.

[0113] The second unit is used to perform linear fitting on the hydrological parameter sequence to obtain the trend slope.

[0114] The third unit is used to set a first comparison length, a second comparison length, and a similarity threshold. The difference between the second comparison length and the first comparison length is 1. Based on the first comparison length and the second comparison length, the hydrological parameter sequence is divided into several first vector pairs and several second vector pairs. Based on the several first vector pairs and the similarity threshold, a first vector comparison mean is obtained, and based on the several second vector pairs and the similarity threshold, a second vector comparison mean is obtained. Based on the first vector comparison mean and the second vector comparison mean, an approximate entropy is obtained. The sliding window mean, the trend slope, and the approximate entropy are concatenated into a time-series feature.

[0115] The third unit is further configured to select one of the first vector pairs as calibration vector pairs and select all the first vector pairs as reference vector pairs, obtain the Chebyshev distance between the calibration vector pairs and the reference vector pairs; compare the Chebyshev distance with the similarity threshold to obtain the number of similar reference vector pairs whose Chebyshev distance is less than the similarity threshold, obtain the similarity ratio by comparing the number of similarities with the total number of the first vector pairs; and perform mean processing on all the similarity ratios to obtain the first vector comparison mean.

[0116] The fourth unit is used to obtain the geographical coordinates of the site, obtain the Euclidean distance between the two sites based on the geographical coordinates, and determine the association weight between the two sites based on the Euclidean distance.

[0117] The fifth unit is used to obtain the Pearson correlation coefficient between the two stations based on the hydrological parameter sequence;

[0118] The sixth unit is used to obtain the spatial correlation features corresponding to the data packet through the Pearson correlation coefficient and the correlation weight;

[0119] The construction module 20 is used to map the data source node, routing relay node and target data center node of the site to nodes, and to construct the connection edge between the nodes, and to assign dynamic security weights to the connection edge based on the multidimensional hydrological feature vector.

[0120] The construction module 20 includes:

[0121] The seventh unit is used to select the two nodes connected to the connection edge as the sending node and the receiving node, respectively, obtain the security score of the sending node and the importance score of the receiving node, and obtain the network score between the sending node and the receiving node.

[0122] The eighth unit is used to obtain the data sensitivity score of the multidimensional hydrological feature vector at the sending node, and construct a dynamic security weight through the security score, the importance score, the network score and the data sensitivity score;

[0123] The aggregation module 30 is used to obtain the self-feature corresponding to the node, select one of the nodes as the calibration node, select the self-feature corresponding to the calibration node as the calibration feature, select the nodes with connected edges to the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features, and obtain the updated feature corresponding to the calibration node based on the calibration feature, the neighboring feature and the dynamic security weight.

[0124] Execution module 40 is used to generate a model from the updated feature input parameters to dynamically obtain an adaptive encryption parameter set, perform layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet, and transmit the encrypted data packet.

[0125] The execution module 40 includes:

[0126] The ninth unit is used to encrypt the adaptive encryption parameter set as a plaintext attribute using an attribute-based encryption method to encrypt the adaptive encryption parameter set into a parameter ciphertext header.

[0127] The tenth unit is used to randomly generate a session key based on the key length, and encrypt the hydrological data into ciphertext using the session key, the encryption algorithm type, and the operation mode;

[0128] The eleventh unit is used to encrypt the session key based on the encapsulation mechanism and the public key of the target node to generate an encryption key, and to encapsulate the parameter ciphertext header, the data ciphertext, and the encryption key into an encrypted data packet.

[0129] The present invention also provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the hydrological data encryption transmission method as described in the above technical solutions.

[0130] The present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the hydrological data encryption transmission method as described in the above technical solution.

[0131] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0132] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for encrypted transmission of hydrological data, characterized in that, Includes the following steps: The process involves acquiring hydrological data to be transmitted, which includes data packets from several stations and at least one sequence of hydrological parameters. Temporal features are extracted from the data packets, including sliding window mean, trend slope, and approximate entropy. Spatial correlation features are also extracted from the data packets. The temporal features and spatial correlation features are then combined into a multidimensional hydrological feature vector. The data source node, routing relay node, and target data center node of the site are mapped to nodes, and connection edges between nodes are constructed. Dynamic security weights are assigned to the connection edges based on the multidimensional hydrological feature vector. The step of assigning dynamic security weights to the connecting edges based on the multidimensional hydrological feature vector includes: The two nodes connected to the connection edge are selected as the sending node and the receiving node, respectively. The security score of the sending node and the importance score of the receiving node are obtained, and the network score between the sending node and the receiving node is obtained. The data sensitivity score of the multidimensional hydrological feature vector at the sending node is obtained, and a dynamic security weight is constructed by the security score, the importance score, the network score, and the data sensitivity score. The steps for obtaining the data sensitivity score include: For the sliding window mean, trend slope, approximate entropy, and spatial correlation features included in the multidimensional hydrological feature vector, corresponding scoring segments are set. Sum all the score contributions to obtain the data sensitivity score; Obtain the self-feature corresponding to the node, select one of its nodes as the calibration node, and select the self-feature corresponding to the calibration node as the calibration feature. Select the nodes that have a connection edge with the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features. Based on the calibration feature, the neighboring features and the dynamic security weight, obtain the updated feature corresponding to the calibration node. The updated feature input parameters are used to generate a model to dynamically obtain an adaptive encryption parameter set. Based on the adaptive encryption parameter set, the hydrological data is subjected to layered hybrid encryption to form an encrypted data packet, which is then transmitted.

2. The method for encrypted transmission of hydrological data according to claim 1, characterized in that, The hydrological parameter sequence includes several preprocessed data points in time series. The steps for extracting time series features from the data packet include: Set a sliding window of a preset size, select the preprocessed data points of the current time frame as calibration points, and select several preprocessed data points adjacent to the calibration points as reference points based on the sliding window. Perform mean processing on the calibration points and several reference points to obtain the mean of the sliding window. The hydrological parameter sequence is linearly fitted to obtain the trend slope; A first comparison length, a second comparison length, and a similarity threshold are set. The difference between the second comparison length and the first comparison length is 1. Based on the first comparison length and the second comparison length, the hydrological parameter sequence is divided into several first vector pairs and several second vector pairs. The first vector pair and the similarity threshold are used to obtain the first vector comparison mean, and the second vector pair and the similarity threshold are used to obtain the second vector comparison mean. The first vector comparison mean and the second vector comparison mean are used to obtain the approximate entropy. The sliding window mean, the trend slope, and the approximate entropy are concatenated to form a time-series feature.

3. The method for encrypted transmission of hydrological data according to claim 2, characterized in that, The formula for obtaining the trend slope is: ， in, Indicates the slope of the trend. This represents the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of time frames in a hydrological parameter sequence. This represents the preprocessed data point corresponding to the i-th time frame in the hydrological parameter sequence. This represents the arithmetic mean of the preprocessed data points in the hydrological parameter sequence.

4. The method for encrypted transmission of hydrological data according to claim 2, characterized in that, The step of obtaining the average of the first vector comparisons based on several pairs of the first vectors and the similarity threshold includes: Select one of the first vector pairs as the calibration vector pair and select all the first vector pairs as the reference vector pairs to obtain the Chebyshev distance between the calibration vector pair and the reference vector pair. The Chebyshev distance is compared with the similarity threshold to obtain the number of similar reference vector pairs whose Chebyshev distance is less than the similarity threshold. The similarity ratio is obtained by comparing the number of similar vector pairs with the total number of the first vector pairs. The average value of all the similarity percentages is averaged to obtain the first vector comparison mean.

5. The method for encrypted transmission of hydrological data according to claim 1, characterized in that, The step of extracting spatial correlation features from the data packet includes: Obtain the geographical coordinates of the site, obtain the Euclidean distance between the two sites based on the geographical coordinates, and determine the association weight between the two sites based on the Euclidean distance; The Pearson correlation coefficient between the two stations was obtained based on the hydrological parameter sequence. The spatial correlation features corresponding to the data packet are obtained by using the Pearson correlation coefficient and the correlation weight.

6. The method for encrypted transmission of hydrological data according to claim 5, characterized in that, The formula for obtaining the spatial association features is as follows: ， in, This represents the spatial association characteristics of data packets from the a-th site. This represents the association weight between data packets from site a and site b, and , This represents the Euclidean distance between station a and station b. This represents the Pearson correlation coefficient between data packets from site a and site b. This indicates the total number of sites.

7. The method for encrypted transmission of hydrological data according to claim 1, characterized in that, The formula for obtaining the updated features is: ， in, This represents the update feature when the p-th node is the calibration node. This represents the calibration feature when the p-th node is a calibration node. This represents the proximity characteristics of the q-th neighboring node when the p-th node is the labeled node. This represents the total number of neighboring nodes when the p-th node is the labeled node. This represents the attention coefficient between the p-th node and its corresponding q-th neighbor when the p-th node is the calibration node. , Both represent weight matrices. This represents a non-linear activation function.

8. The method for encrypted transmission of hydrological data according to claim 1, characterized in that, The adaptive encryption parameter set includes encryption algorithm type, operation mode, key length, and encapsulation mechanism. The step of performing layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet includes: The adaptive encryption parameter set is used as a plaintext attribute, and the attribute-based encryption method is used to encrypt the adaptive encryption parameter set into a parameter ciphertext header. A session key is randomly generated based on the key length, and the hydrological data is encrypted into ciphertext using the session key, the encryption algorithm type, and the operation mode. The session key is encrypted based on the encapsulation mechanism and the public key of the target node to generate an encryption key. The parameter ciphertext header, the data ciphertext, and the encryption key are then encapsulated into an encrypted data packet.

9. A hydrological data encryption transmission system, applied to the hydrological data encryption transmission method as described in any one of claims 1 to 8, characterized in that, The system includes: The acquisition module is used to acquire hydrological data to be transmitted, the hydrological data including data packets from several stations, the data packets including at least one sequence of hydrological parameters, extracting temporal features from the data packets, extracting spatial correlation features from the data packets, and combining the temporal features and the spatial correlation features into a multidimensional hydrological feature vector. The construction module is used to map the data source nodes, routing relay nodes and target data center nodes of the site to nodes, and to construct the connection edges between the nodes, and to assign dynamic security weights to the connection edges based on the multidimensional hydrological feature vector. The aggregation module is used to obtain the self-feature corresponding to the node, select one of the nodes as the calibration node, select the self-feature corresponding to the calibration node as the calibration feature, select the nodes with connected edges to the calibration node as neighboring nodes, and select the self-feature corresponding to the neighboring nodes as neighboring features, and obtain the updated features corresponding to the calibration node based on the calibration feature, the neighboring features and the dynamic security weight. The execution module is used to generate a model from the updated feature input parameters to dynamically obtain an adaptive encryption parameter set, perform layered hybrid encryption on the hydrological data based on the adaptive encryption parameter set to form an encrypted data packet, and transmit the encrypted data packet.

Images