Multi-party supply chain sensing data encryption metering and traceability method based on space-time coupling watermark embedding

Abstract

The invention discloses a multi-party supply chain sensing data encryption measurement and traceability method based on space-time coupling watermark embedding, which comprises the following steps of: constructing a double-domain coupling model of a time sequence and a space distribution characteristic, and embedding dynamic watermark information into a sensing data stream to realize data source identity identification and tamper-proof protection. A self-adaptive key generation and credible chain verification mechanism is introduced in an encryption metering stage, so that the data integrity and confidentiality can be guaranteed on the premise of not influencing the real-time transmission performance of the data; in the tracing stage, a data transmission path is reconstructed by using the watermark tracing matrix, and data responsibility definition and anomaly detection among multiple nodes are realized. The method is suitable for high-security data management and credible tracking in a supply chain multi-node collaborative environment.

Description

Technical Field

[0001] This invention belongs to the fields of Internet of Things and Artificial Intelligence, and specifically relates to a method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding. Background Technology

[0002] With the rapid development of the Internet of Things (IoT) and smart manufacturing, various types of sensors, such as those for temperature, humidity, pressure, energy consumption, and geolocation, are widely deployed across the supply chain to achieve real-time monitoring of raw materials, production, logistics, and sales. However, supply chain sensor data is frequently transmitted and shared among multiple nodes, leading to problems such as heterogeneous data sources, inconsistent sampling frequencies, and significant differences in node reliability. In complex network environments, sensor data is highly susceptible to forgery, tampering, or duplicate reporting, making it difficult to guarantee the integrity and authenticity of data during cross-node transmission.

[0003] Existing supply chain data security methods are mostly based on symmetric encryption, digital signatures, or blockchain notarization technologies. However, most methods only focus on the static encryption of data content, lacking modeling of the correlation between data during temporal changes and spatial flows. Meanwhile, traditional traceability mechanisms rely on a single hash chain or node log verification, which cannot accurately distinguish the responsibilities of multiple nodes and is difficult to achieve efficient data consistency verification in large-scale sensor networks. Especially in dynamic multi-node collaborative scenarios, where data transmission paths are complex and cross-domain, existing solutions exhibit significant performance bottlenecks in terms of tamper resistance and traceability.

[0004] To address the technical challenges of secure and reliable traceability of sensor data transmission in multi-party supply chain environments, a joint modeling method that considers both temporal evolution and spatial distribution characteristics is needed. This invention constructs a spatiotemporally coupled watermark embedding mechanism, combining dynamic identification information with encrypted measurement processes to achieve data source identity binding and transmission trajectory recording. Simultaneously, by combining chained hash verification and a traceability matrix matching model, a traceable, verifiable, and locatable end-to-end data security framework is formed, thereby enhancing the anti-counterfeiting and reliability of data in multi-node collaborative supply chain environments. Summary of the Invention

[0005] This invention proposes a method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding. By constructing a spatiotemporal coupled model that integrates time series and spatial distribution features, dynamic watermark information bound to node identifiers, timestamps, and geographical locations is generated. This watermark is reversibly embedded into the sensor data stream and combined with an adaptive key mechanism to achieve multi-layer encrypted measurement. A trusted transmission chain is established using chained hashing and verification indexes. During the data auditing stage, the data transmission path and node relationships are reconstructed through traceability matrix matching and on-chain record verification, forming a secure measurement system that integrates data encryption, tamper prevention, and traceability. This enables highly reliable management and accurate responsibility identification of multi-node sensor data in the supply chain.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporally coupled watermark embedding includes the following steps:

[0008] S1 extracts time series and spatial distribution features from multi-source sensor data in the supply chain, uses a spatiotemporal coupling analysis model to capture the dynamic correlation and change patterns of data between different nodes, and generates a unique and dynamic spatiotemporal coupling watermark identifier by combining node identifier, timestamp, geographical location and sensor attribute information, and establishes the correspondence between the watermark and the data stream.

[0009] S2, during the data transmission phase, embeds the spatiotemporal coupled watermark into the original sensor data stream and uses a reversible embedding algorithm to maintain data parsability; through an adaptive key generation mechanism, the data content and watermark information are encrypted and measured in layers to form an encrypted interaction structure between nodes; a trusted transmission chain is established using chain hashing and verification index to encrypt and record the data flow process between multiple nodes and verify consistency, forming a full-link encrypted measurement system;

[0010] S3, during the data usage, detection, and auditing phases, parses the embedded spatiotemporal watermark information from the data stream and reconstructs the transmission path and node interaction relationship; through traceability matrix matching and on-chain verification records, it verifies and restores the generation, transmission, and reception status of sensor data at each node, forming a complete and traceable spatiotemporal tracking chain.

[0011] The spatiotemporal coupling analysis model in S1 is as follows:

[0012] S11, the supply chain system consists of n nodes, where the i-th node is denoted as n. i Its spatial position vector is p i =(x i ,y i ,z i ), where x i ,yi ,z i These represent the geographic location parameters of the node in the three-dimensional coordinate system; node n i The multidimensional sensing data collected at time t is represented as follows:

[0013]

[0014] Where d is the sensing feature dimension, including temperature, humidity, pressure, and current indicators; the time sampling set of the node is defined as... Where T is the total number of samples;

[0015] S12, to characterize the dynamic evolution of nodes in the time dimension, defines the time window length as L. t Then node n i The time-domain eigenvector at time t is represented as:

[0016] f t (i,t)=g t (s i (t),s i (t-1),…,s i (tL t +1))

[0017] Where g t (·) represents the time correlation extraction function, which can be a mapping operator for time-series neural networks; the time increment feature Δs is introduced. i (t)=s i (t)-s i (t-1) reflects the rate of change of node sensing data within adjacent time intervals;

[0018] S13, at time t, the set of active nodes is Define node n i With n j The spatial adjacency weight is:

[0019] w ij =exp(-α||p i -p j ||2)

[0020] Where ||·||² represents the Euclidean distance operation, α>0 is the spatial attenuation coefficient, used to control the influence of distance on the association strength; node n i Aggregation features in the spatial domain are defined as:

[0021]

[0022] Where g s (·) is a spatial feature mapping function that extracts the collaborative change features of nodes in their spatial neighborhood;

[0023] S14, time-domain features f t (i,t), spatial domain features f s (i,t), node position p i And timestamp t are jointly modeled to form node n i Spatiotemporal coupling eigenvectors:

[0024] h i (t)=Φ(f t (i,t)||f s (i,t)||p i ||t)

[0025] Where || represents vector concatenation operation, Φ(·) is the feature fusion mapping function, defined as Φ(x)=σ(Wx+b), σ(·) is the non-linear activation function, and W and b are the learnable weight matrix and bias term, respectively;

[0026] S15, based on the spatiotemporal coupling feature h i (t) Generate a unique node-time watermark seed identifier:

[0027] z i (t)=H(n i ||t||h i (t))

[0028] Where H(·) is an irreversible hash function used to generate a fixed-length fingerprint representation; z i (t) Input the watermark symbol mapping function Γ(·) to obtain the spatiotemporally coupled watermark sequence w i (t)=Γ(z i (t)), where w i (t) is the sequence of watermark information used for subsequent embedding and tracing.

[0029] The trusted transmission chain established in S2 using chained hashing and verification indexes is as follows:

[0030] S21, There are N nodes n1, n2, ..., n in the supply chain. N The nodes form a data link according to the transmission order; node n i At time t, downstream node n i+1 The transmitted encrypted data block is defined as follows:

[0031] B i (t)={s i (t),w i (t),K i (t),T i ,n i}

[0032] Among them, s i (t) is node n i The original sensor data vector, w i (t) is the embedded spatiotemporally coupled watermark sequence, K i (t) is node n i The temporary key generated at time t, T i It is timestamp information, n i It is a node identifier;

[0033] S22, To achieve chain-like tamper-proof records, calculate the hash value of each data block:

[0034] H i =h(B i (t))

[0035] Where h(·) is a one-way hash function; to establish the transmission association between nodes, each block is linked with the hash value of the previous node to form a chain structure:

[0036]

[0037] in It is the cumulative hash value of the previous node after chaining operations, and || is the concatenation operator. Represents node n i The accumulated hash identifier on the chain after chain verification; the resulting hash sequence. This constitutes a trusted transmission chain arranged sequentially in the supply chain.

[0038] S23, Introduce a verification index table Record the transmission status and encrypted metering information of each node:

[0039]

[0040] Where, n i It is a node identifier; T i It is a transmission timestamp; K is the cumulative hash value of this node. i (t) is the encryption key; σ i It is a node signature identifier;

[0041] S24, let node n i The public and private keys are respectively (P) i ,S i Then the data encryption and verification process is defined as follows:

[0042]

[0043] in, Based on temporary key K i A symmetric encryption function for (t); It is the node's private key signing function; Public key-based signature verification function; C i The encrypted ciphertext; S′ i Node digital signature; V i If the verification passes, then V i =1, otherwise V i =0;

[0044] The data transmission chain structure of each node can be represented as follows: The hash value and verification index are updated during each inter-node transmission to achieve dynamic measurement and secure data transmission.

[0045] The records in S3 obtained through source tracing matrix matching and on-chain verification are as follows:

[0046] S31, the encrypted data block received by the receiving node is in For node n i Use temporary key K i The encryption function for (t) is as follows; the corresponding decryption process is: in This is the decryption function; after decryption, a data block containing the original sensor data and watermark information is obtained:

[0047] B i ′(t)={s i ′(t),w i ′(t),K i (t),T i ,n i}

[0048] Where w′ i (t) represents the extracted watermark sequence.

[0049] S32, Construct the watermark matching matrix M between nodes = [m ij ] N×N , where element m ij Defined as node n i With node n j Watermark similarity measurement between:

[0050] m ij =sim(w i ′(t),w j ′(t′))

[0051] sim(·) is the cosine similarity function; when m ij When the threshold is greater than θ, node n is considered to be...i With n j An inheritance relationship exists in the data transmission chain; thus, the tracing matrix is ​​obtained:

[0052]

[0053] S33, the chained hash sequence generated according to step S22. Define the on-chain verification function V chain (·):

[0054]

[0055] Where h(·) is the hash function, and || represents the concatenation operator; if the verification result is 1, it means that the node data is consistent with the chain hash record; if it is 0, it means that the node data is at risk of being tampered with.

[0056] S33, based on the watermark tracing matrix M and the on-chain verification result V chain (n i Define the spatiotemporal path weight matrix:

[0057] R = [r] ij ] N×N ,r ij =m ij ·V chain (n j )

[0058] Where, r ij Reflecting node n i The data is transmitted to node n j The probability or reliability of fidelity during the process; the spatiotemporal tracking path is represented in graph structure as follows:

[0059] ε={(n i ,n j )∣r ij >η}

[0060] in, Let r be the set of nodes; ε be the set of trusted transmission edges; η be the path trust threshold; when r ij When >η, determine node n i With n j There is a valid and reliable transmission relationship;

[0061] S32, calculate the overall credibility score for each node:

[0062]

[0063] Where γ i Represents node n i In the overall transmission chain, the reliability index is γ.i <γ min If the minimum threshold of the system is reached, the node is identified as a potential abnormal node.

[0064] Compared with the prior art, the beneficial effects of this invention are:

[0065] (1) By constructing a spatiotemporal coupled watermark embedding model, dynamic features such as node identifier, timestamp and spatial location are jointly encoded into the watermark, realizing the unique identification and anti-tampering protection of sensor data in the multi-node transmission process, overcoming the problems of data identity separation and easy counterfeiting in the existing technology, and significantly improving data integrity and security credibility.

[0066] (2) By combining chain hash verification and traceability matrix matching mechanism, a trusted transmission chain between multiple nodes is established to achieve accurate tracking of data flow path and identification of node responsibility. This solves the defects of coarse traceability granularity, broken verification chain and unclear node responsibility in existing traceability methods, and improves the traceability and verifiability of supply chain sensor data. Attached Figure Description

[0067] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0068] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0069] Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. These descriptions are only for further illustrating the features and advantages of the present invention, and not for limiting the claims of the present invention. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0071] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0072] like Figure 1 As shown, the framework of this invention mainly consists of the following three steps, which are connected layer by layer and ultimately merged. The process mainly includes the following steps:

[0073] S1 extracts time series and spatial distribution features from multi-source sensor data in the supply chain, uses a spatiotemporal coupling analysis model to capture the dynamic correlation and change patterns of data between different nodes, and generates a unique and dynamic spatiotemporal coupling watermark identifier by combining node identifier, timestamp, geographical location and sensor attribute information, and establishes the correspondence between the watermark and the data stream.

[0074] The spatiotemporal coupling analysis model in S1 is as follows:

[0075] S11, the supply chain system consists of N nodes, where the i-th node is denoted as n. i Its spatial position vector is p i =(x i ,y i ,z i ), where x i ,y i ,z i These represent the geographic location parameters of the node in the three-dimensional coordinate system; node n i The multidimensional sensing data collected at time t is represented as follows:

[0076]

[0077] Where d is the sensing feature dimension, including temperature, humidity, pressure, and current indicators; the time sampling set of the node is defined as... Where T is the total number of samples;

[0078] S12, to characterize the dynamic evolution of nodes in the time dimension, defines the time window length as L. t Then node n i The time-domain eigenvector at time t is represented as:

[0079] f t (i,t)=g t (s i (t),s i (t-1),…,s i (tL t +1))

[0080] Where g t(·) represents the time correlation extraction function, which can be a mapping operator for time-series neural networks; the time increment feature Δs is introduced. i (t)=s i (t)-s i (t-1) reflects the rate of change of node sensing data within adjacent time intervals;

[0081] S13, at time t, the set of active nodes is Define node n i With n j The spatial adjacency weight is:

[0082] w ij =exp(-α||p i -p j ||2)

[0083] Where ||·||² represents the Euclidean distance operation, α>0 is the spatial attenuation coefficient, used to control the influence of distance on the association strength; node n i Aggregation features in the spatial domain are defined as:

[0084]

[0085] Where g s (·) is a spatial feature mapping function that extracts the collaborative change features of nodes in their spatial neighborhood;

[0086] S14, time-domain features f t (i,t), spatial domain features f s (i,t), node position p i And timestamp t are jointly modeled to form node n i Spatiotemporal coupling eigenvectors:

[0087] h i (t)=Φ(f t (i,t)||f s (i,t)||p i ||t)

[0088] Where || represents vector concatenation operation, Φ(·) is the feature fusion mapping function, defined as Φ(x)=σ(Wx+b), σ(·) is the non-linear activation function, and W and b are the learnable weight matrix and bias term, respectively;

[0089] S15, based on the spatiotemporal coupling feature h i (t) Generate a unique node-time watermark seed identifier:

[0090] z i (t)=H(n i ||t||h i(t))

[0091] Where H(·) is an irreversible hash function used to generate a fixed-length fingerprint representation; z i (t) Input the watermark symbol mapping function Γ(·) to obtain the spatiotemporally coupled watermark sequence w i (t)=Γ(z i (t)), where w i (t) is the sequence of watermark information used for subsequent embedding and tracing.

[0092] S2, during the data transmission phase, embeds the spatiotemporal coupled watermark into the original sensor data stream and uses a reversible embedding algorithm to maintain data parsability; through an adaptive key generation mechanism, the data content and watermark information are encrypted and measured in layers to form an encrypted interaction structure between nodes; a trusted transmission chain is established using chain hashing and verification index to encrypt and record the data flow process between multiple nodes and verify consistency, forming a full-link encrypted measurement system;

[0093] The trusted transmission chain established in S2 using chained hashing and verification indexes is as follows:

[0094] S21, There are N nodes n1, n2, ..., n in the supply chain. N The nodes form a data link according to the transmission order; node n i At time t, downstream node n i+1 The transmitted encrypted data block is defined as follows:

[0095] B i (t)={s i (t),w i (t),K i (t),T i ,n i}

[0096] Among them, s i (t) is node n i The original sensor data vector, w i (t) is the embedded spatiotemporally coupled watermark sequence, K i (t) is node n i The temporary key generated at time t, T i It is timestamp information, n i It is a node identifier;

[0097] S22, To achieve chain-like tamper-proof records, calculate the hash value of each data block:

[0098] H i =h(B i (t))

[0099] Where h(·) is a one-way hash function; to establish the transmission association between nodes, each block is linked with the hash value of the previous node to form a chain structure:

[0100]

[0101] in It is the cumulative hash value of the previous node after chaining operations, and || is the concatenation operator. Represents node n i The accumulated hash identifier on the chain after chain verification; the resulting hash sequence. This constitutes a trusted transmission chain arranged sequentially in the supply chain.

[0102] S23, Introduce a verification index table Record the transmission status and encrypted metering information of each node:

[0103]

[0104] Where, n i It is a node identifier; T i It is a transmission timestamp; K is the cumulative hash value of this node. i (t) is the encryption key; σ i It is a node signature identifier;

[0105] S24, let node n i The public and private keys are respectively (P) i ,S i Then the data encryption and verification process is defined as follows:

[0106]

[0107] in, Based on temporary key K i A symmetric encryption function for (t); It is the node's private key signing function; Public key-based signature verification function; C i The encrypted ciphertext; S′ i Node digital signature; V i If the verification passes, then V i =1, otherwise V i =0;

[0108] The data transmission chain structure of each node can be represented as follows: The hash value and verification index are updated during each inter-node transmission to achieve dynamic measurement and secure data transmission.

[0109] S3, during the data usage, detection, and auditing phases, parses the embedded spatiotemporal watermark information from the data stream and reconstructs the transmission path and node interaction relationship; through traceability matrix matching and on-chain verification records, it verifies and restores the generation, transmission, and reception status of sensor data at each node, forming a complete and traceable spatiotemporal tracking chain.

[0110] The records in S3 obtained through source tracing matrix matching and on-chain verification are as follows:

[0111] S31, the encrypted data block received by the receiving node is in For node n i Use temporary key K i The encryption function for (t) is as follows; the corresponding decryption process is: in This is the decryption function; after decryption, a data block containing the original sensor data and watermark information is obtained:

[0112] B i ′(t)={s i ′(t),w i ′(t),K i (t),T i ,n i}

[0113] Where W′ i (t) represents the extracted watermark sequence.

[0114] S32, Construct the watermark matching matrix M between nodes = [m ij ] N×N , where element m ij Defined as node n i With node n j Watermark similarity measurement between:

[0115] m ij =sim(w i ′(t),w j ′(t′))

[0116] sim(·) is the cosine similarity function; when m ij When the threshold is greater than θ, node n is considered to be... i With n j An inheritance relationship exists in the data transmission chain; thus, the tracing matrix is ​​obtained:

[0117]

[0118] S33, the chained hash sequence generated according to step S22. Define the on-chain verification function V chain (·):

[0119]

[0120] Where h(·) is the hash function, and || represents the concatenation operator; if the verification result is 1, it means that the node data is consistent with the chain hash record; if it is 0, it means that the node data is at risk of being tampered with.

[0121] S33, based on the watermark tracing matrix M and the on-chain verification result V chain (n i Define the spatiotemporal path weight matrix:

[0122] R = [r] ij ] N×N ,r ij =m ij ·V chain (n j )

[0123] Where, r ij Reflecting node n i The data is transmitted to node n j The probability or reliability of fidelity during the process; the spatiotemporal tracking path is represented in graph structure as follows:

[0124] ε={(n i ,n j )∣r ij >η}

[0125] in, Let r be the set of nodes; ε be the set of trusted transmission edges; η be the path trust threshold; when r ij When >η, determine node n i With n j There is a valid and reliable transmission relationship;

[0126] S32, calculate the overall credibility score for each node:

[0127]

[0128] Where γ i Represents node n i In the overall transmission chain, the reliability index is γ. i <γ min If the minimum threshold of the system is reached, the node is identified as a potential abnormal node.

[0129] This invention proposes a method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporally coupled watermark embedding. By constructing a spatiotemporally coupled model that integrates time series and spatial distribution features, dynamic watermark information bound to node identifiers, timestamps, and spatial coordinates is generated. This watermark is embedded into multi-source sensor data streams and employs an adaptive key mechanism for layered encryption and chained hash measurement, forming a trusted data transmission chain. During the traceability phase, data path reconstruction and node responsibility identification are achieved through source matrix matching and on-chain verification records. This method can be widely applied to scenarios such as supply chain sensing, intelligent manufacturing, cold chain logistics, energy metering, and IoT security management, possessing comprehensive capabilities including data anti-counterfeiting, encrypted measurement, full-process tracking, and trusted node verification.

[0130] The above description only illustrates the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention, and all such changes should be included within the protection scope of the present invention.

Claims

1. A method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding, characterized in that, Includes the following steps: S1 extracts time series and spatial distribution features from multi-source sensor data in the supply chain, uses a spatiotemporal coupling analysis model to capture the dynamic correlation and change patterns of data between different nodes, and generates a unique and dynamic spatiotemporal coupling watermark identifier by combining node identifier, timestamp, geographical location and sensor attribute information, and establishes the correspondence between the watermark and the data stream. S2, during the data transmission phase, embeds the spatiotemporal coupled watermark into the original sensor data stream and uses a reversible embedding algorithm to maintain data parsability; through an adaptive key generation mechanism, the data content and watermark information are encrypted and measured in layers to form an encrypted interaction structure between nodes; a trusted transmission chain is established using chain hashing and verification index to encrypt and record the data flow process between multiple nodes and verify consistency, forming a full-link encrypted measurement system; S3, during the data usage, detection, and auditing phases, parses the embedded spatiotemporal watermark information from the data stream and reconstructs the transmission path and node interaction relationship; through traceability matrix matching and on-chain verification records, it verifies and restores the generation, transmission, and reception status of sensor data at each node, forming a complete and traceable spatiotemporal tracking chain.

2. The method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding according to claim 1, characterized in that, The spatiotemporal coupling analysis model in S1 is as follows: S11, the supply chain system consists of N nodes, where the i-th node is denoted as n. i Its spatial position vector is p i =(x i ,y i ,z i ), where x i ,y i ,z i These represent the geographic location parameters of the node in the three-dimensional coordinate system; node n i The multidimensional sensing data collected at time t is represented as follows: Where d is the sensing feature dimension, including temperature, humidity, pressure, and current indicators; the time sampling set of the node is defined as... Where T is the total number of samples; S12, to characterize the dynamic evolution of nodes in the time dimension, defines the time window length as L. t Then node n i The time-domain eigenvector at time t is represented as: f t (i,t)=g t (s i (t),s i (t-1),…,s i (t-L t +1)) Where g t (·) represents the time correlation extraction function, which is a time-series neural network mapping operator; a time increment feature Δs is introduced. i (t)=s i (t)-s i (t-1) reflects the rate of change of node sensing data within adjacent time intervals; S13, at time t, the set of active nodes is Define node n i With n j The spatial adjacency weight is: w ij =exp(-α||p i -p j ||2) Where ||·||² represents the Euclidean distance operation, α>0 is the spatial attenuation coefficient, used to control the influence of distance on the association strength; node n i Aggregation features in the spatial domain are defined as: Where g s (·) is a spatial feature mapping function that extracts the collaborative change features of nodes in their spatial neighborhood; S14, time-domain features f t (i,t), spatial domain features f s (i,t), node position p i And timestamp t are jointly modeled to form node n i Spatiotemporal coupling eigenvectors: h i (t)=Φ(f t (i,t)||f s (i,t)||p i ||t) Where || represents vector concatenation operation, Φ(·) is the feature fusion mapping function, defined as Φ(x)=σ(Wx+b), σ(·) is the non-linear activation function, and W and b are the learnable weight matrix and bias term, respectively; S15, based on the spatiotemporal coupling feature h i (t) Generate a unique node-time watermark seed identifier: z i (t)=H(n i ||t||h i (t)) Where H(·) is an irreversible hash function used to generate a fixed-length fingerprint representation; z i (t) Input the watermark symbol mapping function Γ(·) to obtain the spatiotemporally coupled watermark sequence w i (t)=Γ(z i (t)), where w i (t) is the sequence of watermark information used for subsequent embedding and tracing.

3. The method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding according to claim 2, characterized in that, The trusted transmission chain established in S2 using chained hashing and verification indexes is as follows: S21, There are N nodes n1, n2, ..., n in the supply chain. N The nodes form a data link according to the transmission order; node n i At time t, downstream node n i+1 The transmitted encrypted data block is defined as follows: B i (t)={s i (t),w i (t),K i (t),T i ,n i } Among them, s i (t) is node n i The original sensor data vector, w i (t) is the embedded spatiotemporally coupled watermark sequence, K i (t) is node n i The temporary key generated at time t, T i It is timestamp information, n i It is a node identifier; S22, To achieve chain-like tamper-proof records, calculate the hash value of each data block: H i =h(B i (t)) Where h(·) is a one-way hash function; to establish the transmission association between nodes, each block is linked with the hash value of the previous node to form a chain structure: in It is the cumulative hash value of the previous node after chaining operations, and || is the concatenation operator. Represents node n i The accumulated hash identifier on the chain after chain verification; the resulting hash sequence. This constitutes a trusted transmission chain arranged sequentially in the supply chain. S23, Introduce a verification index table Record the transmission status and encrypted metering information of each node: Where, n i It is a node identifier; T i It is a transmission timestamp; K is the cumulative hash value of this node. i (t) is the encryption key; σ i It is a node signature identifier; S24, let node n i The public and private keys are (P) i ,S i Then the data encryption and verification process is defined as follows: in, Based on temporary key K i A symmetric encryption function for (t); It is the node's private key signing function; Public key-based signature verification function; C i The encrypted ciphertext; S′ i Node digital signature; V i If the verification passes, then V i =1, otherwise V i =0; The data transmission chain structure of each node can be represented as follows: The hash value and verification index are updated during each inter-node transmission to achieve dynamic measurement and secure data transmission.

4. The method for encrypted measurement and traceability of multi-party supply chain sensor data based on spatiotemporal coupled watermark embedding according to claim 3, characterized in that, The records in S3 obtained through source tracing matrix matching and on-chain verification are as follows: S31, the encrypted data block received by the receiving node is in For node n i Use temporary key K i The encryption function for (t) is as follows; the corresponding decryption process is: in This is the decryption function; after decryption, a data block containing the original sensor data and watermark information is obtained: B i ′(t)={s i ′(t),w i ′(t),K i (t),T i ,n i } Where w′ i (t) represents the extracted watermark sequence. S32, Construct the watermark matching matrix M between nodes = [m ij ] N×N , where element m ij Defined as node n i With node n j Watermark similarity measurement between: m ij =sim(w i ′(t),w j ′(t′)) sim(·) is the cosine similarity function; when m ij When the threshold is greater than θ, node n is considered to be... i With n j An inheritance relationship exists in the data transmission chain; thus, the tracing matrix is ​​obtained: S33, the chained hash sequence generated according to step S22. Define the on-chain verification function V chain (·): Where h(·) is the hash function, and || represents the concatenation operator; if the verification result is 1, it means that the node data is consistent with the chain hash record; if it is 0, it means that the node data is at risk of being tampered with. S33, based on the watermark tracing matrix M and the on-chain verification result V chain (n i Define the spatiotemporal path weight matrix: R=[r ij ] N×N ,r ij =m ij ·V chain (n j ) Where, r ij Reflecting node n i The data is transmitted to node n j The probability or reliability of fidelity during the process; the spatiotemporal tracking path is represented in graph structure as follows: in, Let r be the set of nodes; ε be the set of trusted transmission edges; η be the path trust threshold; when r ij When >η, determine node n i With n j There is a valid and reliable transmission relationship; S32, calculate the overall credibility score for each node: Where γ i Represents node n i In the overall transmission chain, the reliability index is γ. i <γ min If the minimum threshold of the system is reached, the node is identified as a potential abnormal node.

Images