Blockchain-based cloud printer data security protection method and system

Abstract

This invention discloses a blockchain-based cloud printer data security protection method and system, comprising the following steps: S1, collecting and preprocessing multi-source data; S2, constructing a digital twin topology graph, mapping entity nodes, and updating node state features; S3, executing blockchain data storage, using a hierarchical hash chain to segment and map data, and using elliptic curve cryptography to generate digital signatures, forming an identifier data chain; S4, extracting identity features and node state features and fusing them to generate permission verification results; S5, generating a printing task control strategy, using reinforcement learning to establish a state-action mapping, and distributively executing printing tasks; S6, collecting feedback data, verifying consistency with the identifier data chain, updating node states, and completing data security protection. This invention achieves trusted data storage, dynamic permission control, and intelligent collaborative scheduling for cloud printers, improving data anti-tampering capabilities, anomaly detection accuracy, and task execution security.

Description

Technical Field

[0001] This invention relates to the field of blockchain technology, and in particular to a blockchain-based method and system for protecting the data security of cloud printers. Background Technology

[0002] With the rapid development of cloud computing, mobile internet, and IoT technologies, cloud printing technology is gradually being applied in scenarios such as enterprise offices, education, healthcare, and government services. Users upload print jobs to cloud servers via mobile smart terminals, and the cloud platform handles task scheduling, access control, and printing device control, enabling remote printing across regions and terminals. Compared to traditional local printing methods, cloud printing breaks through device connectivity limitations and improves printing resource utilization and management efficiency. However, at the same time, printed documents, user identity information, and device operation data all need to undergo network transmission, cloud storage, and multi-node interaction, making data security issues increasingly prominent.

[0003] Existing cloud printing systems typically employ user authentication, transmission encryption, and access control to ensure data security. Some solutions introduce blockchain technology to store print logs or critical data, enhancing data credibility and traceability. However, most solutions still utilize a centralized data management model, with blockchain used only for data storage or operation records. This fails to establish a unified and trusted association mechanism with the cloud printing device's operational status, user identity information, and the print job execution process. When print jobs flow between mobile smart terminals, cloud platforms, and cloud printers, the lack of collaborative verification capabilities across these stages makes it difficult to achieve consistency verification between job data, device status, and user identity. If a node's status becomes abnormal or identity information is tampered with, it can easily lead to problems such as unauthorized printing, data leaks, and malicious modification of print jobs.

[0004] On the other hand, existing access control methods mostly employ fixed permissions or role-based static authorization mechanisms. Access policies are typically pre-configured based on user identity, lacking the ability to comprehensively analyze real-time device operating status, node trust status, and historical behavioral characteristics. This makes it impossible to adjust access permissions in real time according to dynamic changes during print job execution. Simultaneously, existing print job scheduling methods primarily allocate resources based on task priority, device idle status, or fixed load balancing strategies. They lack intelligent optimization mechanisms that combine permission verification results, node load status, and task execution feedback, making it difficult to achieve dynamic scheduling and continuous optimization of print jobs. This results in low resource utilization, insufficient task execution efficiency, and impacts the overall data security and operational stability of the cloud printing system.

[0005] Therefore, how to provide a blockchain-based cloud printer data security protection method and system is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] One objective of this invention is to propose a blockchain-based method and system for protecting cloud printer data security. This invention integrates blockchain, digital twins, graph convolutional networks, feature fusion, and reinforcement learning algorithms to achieve trusted storage of cloud printing data, dynamic permission verification, and intelligent task scheduling. It has the advantages of immutable data, dynamic permission control, high scheduling efficiency, and strong security protection capabilities.

[0007] A blockchain-based cloud printer data security protection method according to an embodiment of the present invention includes the following steps: S1. Collect multi-source data during the operation of the cloud printer and perform preprocessing; S2. Construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node association topology structure, and using a graph convolutional network to update node state features. S3. Perform blockchain data storage processing based on the digital twin topology diagram, use a hierarchical hash chain to segment and map the printing task data, use elliptic curve cryptography to generate digital signatures, and write the digital signatures into blockchain nodes to form an identifier data chain. S4. Based on the identifier data chain, extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. S5. Generate a print task control strategy based on the permission verification result, use reinforcement learning algorithm to establish a mapping relationship between task status and control action, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute print tasks through distributed control. S6. Collect feedback data in real time during the execution of printing tasks, verify the consistency of feedback data with the identification data chain, update the node status of the digital twin topology graph, and complete the data security protection of the cloud printer.

[0008] Optionally, the multi-source data specifically includes device operating status data, print task execution data, network communication data collected by sensors deployed inside the cloud printer, as well as user identity data, print file data, and task control instruction data uploaded through mobile smart terminals. The device operating status data includes printer temperature parameters, mechanical motion parameters, power status parameters, and communication link status parameters. The print task execution data includes task number, file summary, print parameters, and execution time information. The user identity data includes user identifier, terminal identifier, and access record information.

[0009] Optionally, the preprocessing specifically includes timestamp calibration of multi-source data, synchronizing sensor-collected data and mobile smart terminal-uploaded data based on a unified time reference, completing missing data using neighborhood interpolation, removing abnormally fluctuating data using sliding window detection, performing normalization encoding on continuous data, converting different types of data into a unified format data feature set, establishing a data index identifier, and writing the mapping relationship between the processed data feature set and the original data into a secure storage area.

[0010] Optionally, S2 specifically includes: S21. Construct a digital twin topology diagram. With the cloud printer device number as the root node, deploy a digital twin in the cloud that corresponds one-to-one with the cloud printer. The status refresh cycle of the twin is consistent with the sensor sampling cycle. Registered mobile smart terminals in the same local area network are used as first-level child nodes, and user accounts authenticated by the terminals are used as second-level child nodes to form a tree-like topology structure. S22. Node attributes include the cloud printer's temperature threshold, power supply voltage, communication packet loss rate, mobile smart terminal's MAC address, operating system version, Bluetooth signal strength, as well as the user account's login time period, historical printed pages, and file type distribution. S23. The edge relationships between nodes are determined based on the data flow direction, including: A device communication edge is established between the cloud printer and the mobile smart terminal, and the edge weight is a weighted value of communication latency and handshake success rate; An identity binding edge is established between the mobile smart terminal and the user account, and the edge weight is the ratio of the number of successful authentications to the number of failed authentications. A task scheduling edge is established between the cloud printer and the print task queue, and the edge weight is the normalized product of the task waiting time and the priority value. S24. A two-layer graph convolutional network is used to iteratively update the node state. The first layer aggregates the original attribute features of the one-hop neighboring nodes, and the second layer aggregates the hidden features of the two-hop neighboring nodes. When updating each layer, the state of the node at the previous time step is introduced as a gating input, and the state feature vector of the node at the current time step is output. The update frequency is synchronized with the sensor sampling period. After each node state update is completed, the result is pushed to the blockchain smart contract storage.

[0011] Optionally, S3 specifically includes: S31. Divide the print task data into layers according to file header information, content summary, control instructions and execution log. Perform hash operation on each layer of data to generate the corresponding layer summary. Then, concatenate the layer summaries according to the task execution time order to form a first-level time-series hash chain. S32. Extract four key metadata categories from the data at each level: device identifier, user identifier, timestamp, and task status. Perform association hash calculation on the extracted metadata to generate a secondary association summary that describes the data source and node relationship. S33. Combine the hierarchical digest and the secondary association digest in the first-level time-series hash chain as Merkel leaf nodes to generate the task root hash, and establish a bidirectional mapping relationship between the task data segment, metadata digest and task root hash. S34. A signature mechanism based on the national cryptographic SM2 elliptic curve encryption algorithm is adopted. The task root hash is concatenated with the state feature vector of the corresponding node in the digital twin topology graph to form a signature message body. The node private key is used to generate a digital signature, and random perturbation parameters are generated based on the node's historical signature records to enhance the signature random number generation process. S35. Encapsulate the digital signature, task root hash, hierarchical digest set and Merkel proof path into an identifier data unit and write it into the corresponding data channel of the blockchain network according to the task initiation time order. S36. Based on the blockchain consensus mechanism, the identifier data unit is verified. The hash pointer of the identifier data unit of the previous block, the timestamp of the current block, the consensus node signature, and the set of identifier data units of the current task are written into the block body to form a continuous and related identifier data chain. S37. Establish an index mapping relationship between the identifier data chain and the cloud printer node, mobile smart terminal node and user node number in the digital twin topology diagram, and configure digital signature verification rules in the smart contract. S38. When a new printing task request is received, the smart contract is invoked to read the state feature vector of the corresponding node, and the consistency verification is performed with the signature state in the historical identifier data unit. Based on the verification result, a task data on-chain control instruction is generated, and the identifier data chain of the corresponding node is updated.

[0012] Optionally, S33 specifically includes: S331. Take the four hierarchical digests (file header information digest, content digest, control instruction digest, and execution log digest) in the first-level time-series hash chain, and the four metadata digests (device identifier digest, user identifier digest, timestamp digest, and task status digest) in the second-level association digest, for a total of eight digest values ​​as leaf nodes of the Merkle tree. The leaf node arrangement order is fixed as hierarchical digest first and metadata digest last. The hash calculation of the leaf nodes adopts either left-associative or right-associative methods according to the parity of the task number, forming a differentiated tree structure. S332. Adjacent leaf nodes are paired up and hash operations are performed to generate parent nodes. When the number of parent nodes is odd, the last node is copied to complete the pairing. Aggregation is performed layer by layer until the root node hash value is generated as the task root hash. S333. When establishing a bidirectional mapping relationship, the forward mapping points from the task root hash to the eight leaf node digest values ​​and the original data segment storage address, and the reverse mapping points from each leaf node digest value to the task root hash and other leaf node digest values ​​of the same task. The mapping relationship table is stored in the blockchain state database. The key-value pair structure is the task root hash prefix plus the leaf node index number as the key, and the leaf node digest value plus the data segment offset as the value. S334. When any data segment is tampered with, the task root hash is queried in reverse through the leaf node digest value of the data segment, the hashes of all parent nodes on the root path from the tampered segment are recalculated, and compared with the original Merkel proof path stored in the identifier data chain to locate the tampering location and trigger a data integrity alarm.

[0013] Optionally, S34 specifically includes: S341. Perform a bitwise XOR concatenation between the 256-bit binary string of the task root hash and the state feature vector of the corresponding node in the digital twin topology graph. The state feature vector contains four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. Each dimension is quantized and encoded into a 64-bit binary string, and the concatenation forms a 512-bit signature message body. S342. The national standard SM2 elliptic curve encryption algorithm is adopted, the standard recommended curve parameters are selected, and the signature message body is used as the input message to be signed. S343. During the random number generation process, the previous signature random number in the node's historical signature record, the current timestamp millisecond value, the node's CPU utilization rate, and the remaining memory capacity are used as four-element perturbation sources. The initial random number seed is generated through XOR mixing operation, and then the final signature random number is generated through iterative operation of the national cryptographic SM3 hash algorithm. This enables the same node to generate a unique digital signature for the same task root hash at different times and under different load conditions. S344. Use the node's private key to perform a scalar multiplication operation on the signed message body and a random number to generate a signature component. The signature component, together with the public key coordinates and the message body hash value, constitutes a complete digital signature. S345. Immediately after the signature is generated, the signature random number, signature timestamp, and node load parameters are written to the local secure storage area as input for the historical signature record of the next signature, forming a chain-dependent structure of random number perturbation sources.

[0014] Optionally, S4 specifically includes: S41. Extract the signature component, public key coordinates, message body hash value and historical signature sequence of the digital signature from the identifier data chain, and parse out the identity feature vector, which includes four dimensions: signature entropy value, key usage frequency, signature time interval distribution and cross-node signature consistency. S42. Extract the nodes of mobile smart terminals and their associated user accounts from the digital twin topology graph, and parse out the corresponding state feature vectors, which include four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. S43. Construct a dual-channel feature fusion network. The identity feature channel uses one-dimensional convolution to extract the signature time sequence mode, and the state feature channel uses graph attention to extract the topological association mode. The outputs of the two channels are adaptively weighted and fused through a gated cross-attention mechanism. The gate weights are dynamically adjusted by the current network load and the historical verification accuracy. S44. Input the fused joint feature vector into the permission classifier. The permission classifier contains four output categories: normal access, restricted access, temporary authorization, and denied access. Each category corresponds to a different printing permission configuration, including the maximum number of pages that can be printed, the set of accessible file types, the allowed printing time period, and whether secondary approval is required. S45. After the permission verification result is generated, it is written to the smart contract and bound to the most recent signature record of the corresponding terminal in the identifier data chain. When the value of either the terminal credibility or the user risk in the joint feature vector exceeds the dynamic threshold, permission downgrade or revocation is automatically triggered. The downgraded permission configuration is immediately synchronized to the local access control unit of the cloud printer, and printing tasks that exceed the current permission range are refused.

[0015] Optionally, S5 specifically includes: S51. Obtain the status features of the print task that has passed the permission verification, and build a reinforcement learning environment by combining the load status of the cloud printer node in the digital twin topology map. S52. Construct a task scheduling model based on reinforcement learning algorithm, taking the task allocation node, execution order and resource usage ratio as control actions, updating the policy network through task execution feedback, and establishing a mapping relationship between task state and control actions. S53. Evaluate candidate control actions based on task completion delay, node load balancing, equipment operating status, and safety constraints, and generate printing task scheduling parameters. S54. Dynamically adjust the task allocation strategy based on the processing capacity of the cloud printer node, the task queue status and the communication link status, and schedule the printing task to the corresponding cloud printer node for execution through distributed control. S55. Collect printing task execution feedback data, update the state characteristics and control strategies in the task scheduling model, and recalculate the scheduling parameters when the node load status or task execution status changes, so as to realize dynamic scheduling of cloud printing tasks.

[0016] A blockchain-based cloud printer data security protection system according to an embodiment of the present invention includes: The data acquisition module is used to collect multi-source data during the operation of the cloud printer and perform preprocessing. The twin topology module is used to construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node-related topology structure, and using a graph convolutional network to update node state features. The data storage module is used to perform blockchain data storage processing based on the digital twin topology diagram. It uses a hierarchical hash chain to segment and map the printing task data, uses an elliptic curve cryptography algorithm to generate digital signatures, and writes the digital signatures into blockchain nodes to form an identifier data chain. The permission verification module is used to extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. The task control module is used to generate printing task control policies based on the permission verification results, establish a mapping relationship between task status and control actions using reinforcement learning algorithms, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute printing tasks through distributed control. The status update module is used to collect feedback data in real time during the execution of printing tasks, verify the consistency of the feedback data with the identification data chain, update the status of nodes in the digital twin topology graph, and complete the data security protection of the cloud printer.

[0017] The beneficial effects of this invention are: First, this invention establishes a unified mapping relationship between mobile smart terminals, users, cloud printers, and printing tasks by constructing a digital twin topology graph. It also dynamically updates the node status using a graph convolutional network and utilizes hierarchical hash chains, Merkle trees, and elliptic curve cryptography algorithms to achieve trusted storage and digital signature generation of printing task data. This realizes a trusted association between printing task data, device status, and user identity information, effectively improving the integrity, tamper resistance, and full-process traceability of cloud printing data.

[0018] Secondly, this invention extracts identity features based on the identifier data chain and combines them with the node status features in the digital twin topology graph. It completes trusted identity verification through a dual-channel feature fusion network and then uses smart contracts to realize dynamic management of the permission verification results. This allows permission control to be adjusted in real time according to changes in node status, avoiding the problems of permission abuse, identity impersonation and illegal access that exist in traditional static permission management. This improves the access security and dynamic permission control capabilities of the cloud printing system.

[0019] Finally, this invention constructs a printing task control strategy based on reinforcement learning algorithm according to the permission verification results, establishes a mapping relationship between task status and control actions, and performs intelligent scheduling in combination with the load status of cloud printer nodes. At the same time, it uses printing task execution feedback to continuously update the node status of the digital twin topology graph and the blockchain identifier data chain, forming a closed-loop collaborative mechanism for data storage, permission verification, task scheduling and status update. This realizes dynamic optimization scheduling and continuous security protection of cloud printing tasks, and improves system resource utilization, task execution efficiency and overall operational stability. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a blockchain-based cloud printer data security protection method proposed in this invention; Figure 2 This is a schematic diagram of a blockchain-based cloud printer data security protection method proposed in this invention; Figure 3 This is a schematic diagram illustrating the construction of a trusted storage and identification data chain for printing tasks in a blockchain-based cloud printer data security protection method proposed in this invention. Figure 4 This is a module structure diagram of a cloud printer data security protection system based on blockchain proposed in this invention. Detailed Implementation

[0021] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0022] refer to Figures 1-3 A blockchain-based method for protecting cloud printer data security includes the following steps: S1. Collect multi-source data during the operation of the cloud printer and perform preprocessing; S2. Construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node association topology structure, and using a graph convolutional network to update node state features. S3. Perform blockchain data storage processing based on the digital twin topology diagram, use a hierarchical hash chain to segment and map the printing task data, use elliptic curve cryptography to generate digital signatures, and write the digital signatures into blockchain nodes to form an identifier data chain. S4. Based on the identifier data chain, extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. S5. Generate a print task control strategy based on the permission verification result, use reinforcement learning algorithm to establish a mapping relationship between task status and control action, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute print tasks through distributed control. S6. Collect feedback data in real time during the execution of printing tasks, verify the consistency of feedback data with the identification data chain, update the node status of the digital twin topology graph, and complete the data security protection of the cloud printer.

[0023] In this embodiment, the multi-source data specifically includes device operating status data, print task execution data, network communication data collected by sensors deployed inside the cloud printer, as well as user identity data, print file data, and task control instruction data uploaded through mobile smart terminals. Among them, the device operating status data includes printer temperature parameters, mechanical motion parameters, power status parameters, and communication link status parameters; the print task execution data includes task number, file summary, print parameters, and execution time information; and the user identity data includes user identifier, terminal identifier, and access record information.

[0024] In this embodiment, the preprocessing specifically includes timestamp calibration of multi-source data, synchronizing sensor-collected data and mobile smart terminal-uploaded data based on a unified time reference, completing missing data using neighborhood interpolation, removing abnormal fluctuation data using sliding window detection, performing normalization encoding on continuous data, converting different types of data into a unified format data feature set, establishing data index identifiers, and writing the mapping relationship between the processed data feature set and the original data into a secure storage area.

[0025] In this embodiment, S2 specifically includes: S21. Construct a digital twin topology diagram. With the cloud printer device number as the root node, deploy a digital twin in the cloud that corresponds one-to-one with the cloud printer. The status refresh cycle of the twin is consistent with the sensor sampling cycle. Registered mobile smart terminals in the same local area network are used as first-level child nodes, and user accounts authenticated by the terminals are used as second-level child nodes to form a tree-like topology structure. S22. Node attributes include the cloud printer's temperature threshold, power supply voltage, communication packet loss rate, mobile smart terminal's MAC address, operating system version, Bluetooth signal strength, as well as the user account's login time period, historical printed pages, and file type distribution. S23. The edge relationships between nodes are determined based on the data flow direction, including: A device communication edge is established between the cloud printer and the mobile smart terminal, and the edge weight is a weighted value of communication latency and handshake success rate; An identity binding edge is established between the mobile smart terminal and the user account, and the edge weight is the ratio of the number of successful authentications to the number of failed authentications. A task scheduling edge is established between the cloud printer and the print task queue, and the edge weight is the normalized product of the task waiting time and the priority value. S24. A two-layer graph convolutional network is used to iteratively update the node state. The first layer aggregates the original attribute features of the one-hop neighboring nodes, and the second layer aggregates the hidden features of the two-hop neighboring nodes. When updating each layer, the state of the node at the previous time step is introduced as a gating input, and the state feature vector of the node at the current time step is output. The update frequency is synchronized with the sensor sampling period. After each node state update is completed, the result is pushed to the blockchain smart contract storage.

[0026] In this embodiment, S3 specifically includes: S31. Divide the print task data into layers according to file header information, content summary, control instructions and execution log. Perform hash operation on each layer of data to generate the corresponding layer summary. Then, concatenate the layer summaries according to the task execution time order to form a first-level time-series hash chain. S32. Extract four key metadata categories from the data at each level: device identifier, user identifier, timestamp, and task status. Perform association hash calculation on the extracted metadata to generate a secondary association summary that describes the data source and node relationship. S33. Combine the hierarchical digest and the secondary association digest in the first-level time-series hash chain as Merkel leaf nodes to generate the task root hash, and establish a bidirectional mapping relationship between the task data segment, metadata digest and task root hash. S34. A signature mechanism based on the national cryptographic SM2 elliptic curve encryption algorithm is adopted. The task root hash is concatenated with the state feature vector of the corresponding node in the digital twin topology graph to form a signature message body. The node private key is used to generate a digital signature, and random perturbation parameters are generated based on the node's historical signature records to enhance the signature random number generation process. S35. Encapsulate the digital signature, task root hash, hierarchical digest set and Merkel proof path into an identifier data unit and write it into the corresponding data channel of the blockchain network according to the task initiation time order. S36. Based on the blockchain consensus mechanism, the identifier data unit is verified. The hash pointer of the identifier data unit of the previous block, the timestamp of the current block, the consensus node signature, and the set of identifier data units of the current task are written into the block body to form a continuous and related identifier data chain. S37. Establish an index mapping relationship between the identifier data chain and the cloud printer node, mobile smart terminal node and user node number in the digital twin topology diagram, and configure digital signature verification rules in the smart contract. S38. When a new printing task request is received, the smart contract is invoked to read the state feature vector of the corresponding node, and the consistency verification is performed with the signature state in the historical identifier data unit. Based on the verification result, a task data on-chain control instruction is generated, and the identifier data chain of the corresponding node is updated.

[0027] In this embodiment, S33 specifically includes: S331. Take the four hierarchical digests (file header information digest, content digest, control instruction digest, and execution log digest) in the first-level time-series hash chain, and the four metadata digests (device identifier digest, user identifier digest, timestamp digest, and task status digest) in the second-level association digest, for a total of eight digest values ​​as leaf nodes of the Merkle tree. The leaf node arrangement order is fixed as hierarchical digest first and metadata digest last. The hash calculation of the leaf nodes adopts either left-associative or right-associative methods according to the parity of the task number, forming a differentiated tree structure. S332. Adjacent leaf nodes are paired up and hash operations are performed to generate parent nodes. When the number of parent nodes is odd, the last node is copied to complete the pairing. Aggregation is performed layer by layer until the root node hash value is generated as the task root hash. S333. When establishing a bidirectional mapping relationship, the forward mapping points from the task root hash to the eight leaf node digest values ​​and the original data segment storage address, and the reverse mapping points from each leaf node digest value to the task root hash and other leaf node digest values ​​of the same task. The mapping relationship table is stored in the blockchain state database. The key-value pair structure is the task root hash prefix plus the leaf node index number as the key, and the leaf node digest value plus the data segment offset as the value. S334. When any data segment is tampered with, the task root hash is queried in reverse through the leaf node digest value of the data segment, the hashes of all parent nodes on the root path from the tampered segment are recalculated, and compared with the original Merkel proof path stored in the identifier data chain to locate the tampering location and trigger a data integrity alarm.

[0028] In this embodiment, S34 specifically includes: S341. Perform a bitwise XOR concatenation between the 256-bit binary string of the task root hash and the state feature vector of the corresponding node in the digital twin topology graph. The state feature vector contains four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. Each dimension is quantized and encoded into a 64-bit binary string, and the concatenation forms a 512-bit signature message body. S342. The national standard SM2 elliptic curve encryption algorithm is adopted, the standard recommended curve parameters are selected, and the signature message body is used as the input message to be signed. S343. During the random number generation process, the previous signature random number in the node's historical signature record, the current timestamp millisecond value, the node's CPU utilization rate, and the remaining memory capacity are used as four-element perturbation sources. The initial random number seed is generated through XOR mixing operation, and then the final signature random number is generated through iterative operation of the national cryptographic SM3 hash algorithm. This enables the same node to generate a unique digital signature for the same task root hash at different times and under different load conditions. S344. Use the node's private key to perform a scalar multiplication operation on the signed message body and a random number to generate a signature component. The signature component, together with the public key coordinates and the message body hash value, constitutes a complete digital signature. S345. Immediately after the signature is generated, the signature random number, signature timestamp, and node load parameters are written to the local secure storage area as input for the historical signature record of the next signature, forming a chain-dependent structure of random number perturbation sources.

[0029] In this embodiment, S4 specifically includes: S41. Extract the signature component, public key coordinates, message body hash value and historical signature sequence of the digital signature from the identifier data chain, and parse out the identity feature vector, which includes four dimensions: signature entropy value, key usage frequency, signature time interval distribution and cross-node signature consistency. S42. Extract the nodes of mobile smart terminals and their associated user accounts from the digital twin topology graph, and parse out the corresponding state feature vectors, which include four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. S43. Construct a dual-channel feature fusion network. The identity feature channel uses one-dimensional convolution to extract the signature time sequence mode, and the state feature channel uses graph attention to extract the topological association mode. The outputs of the two channels are adaptively weighted and fused through a gated cross-attention mechanism. The gate weights are dynamically adjusted by the current network load and the historical verification accuracy. S44. Input the fused joint feature vector into the permission classifier. The permission classifier contains four output categories: normal access, restricted access, temporary authorization, and denied access. Each category corresponds to a different printing permission configuration, including the maximum number of pages that can be printed, the set of accessible file types, the allowed printing time period, and whether secondary approval is required. S45. After the permission verification result is generated, it is written to the smart contract and bound to the most recent signature record of the corresponding terminal in the identifier data chain. When the value of either the terminal credibility or the user risk in the joint feature vector exceeds the dynamic threshold, permission downgrade or revocation is automatically triggered. The downgraded permission configuration is immediately synchronized to the local access control unit of the cloud printer, and printing tasks that exceed the current permission range are refused.

[0030] In this embodiment, S5 specifically includes: S51. Obtain the status features of the print task that has passed the permission verification, and build a reinforcement learning environment by combining the load status of the cloud printer node in the digital twin topology map. S52. Construct a task scheduling model based on reinforcement learning algorithm, introduce a dual-policy network to evaluate the action value of the current task state, correct the policy deviation through the target network, and use task execution feedback to update the action selection probability to generate a printing task control policy. S53. Calculate the comprehensive evaluation value of candidate scheduling actions based on the task control strategy, and use task completion delay, node load difference, resource consumption and security constraint status as evaluation parameters to screen target control actions and generate and print task scheduling parameters. S54. Based on the scheduling parameters, the cloud printer nodes are assigned tasks, the status information of each node is synchronized through the distributed control mechanism, and the task execution path is dynamically adjusted based on the node load change results to achieve collaborative scheduling of printing tasks. S55. Collect feedback data during the execution of the printing task, input the feedback data into the task scheduling model for strategy iteration and update, and regenerate the printing task control strategy when the task status or node load changes.

[0031] In this embodiment, S52 specifically includes: S521. The task scheduling model adopts the Actor-Critic architecture. The Actor network includes an online policy network and a target policy network, and the Critic network includes an online value network and a target value network. All four networks are three-layer fully connected structures, with the number of hidden layer neurons being 256, 128, and 64 respectively. The activation function is the linear rectified function LeakyReLU. S522: The online policy network receives the current task status vector as input. The task status vector is composed of ten dimensions: the access permission category code in the permission verification result, the queue length, temperature, power voltage fluctuation, and communication packet loss rate in the cloud printer node load status, and the number of pages, color mode, paper size, priority value, and deadline of the print task. It outputs the probability distribution of five control actions, namely, task admission, task suspension, task migration, task splitting, and task degradation. S523. The online value network receives the concatenated vector of the current task state and the action output by the online policy network as input, and outputs the expected reward value of the state-action pair. S524. The target policy network and target value network replicate the parameters of the online policy network and online value network, respectively. The parameter update adopts a soft update mechanism, and the online network parameters are migrated to the target network in a fixed proportion in each iteration. S525. During training, the experience replay buffer stores quadruple samples, including the current task state, executed action, immediate reward, and next task state. The buffer capacity is set to 100,000 samples. When sampling, samples are prioritized according to the absolute value of the sample reward value. The probability of high-reward samples being sampled is increased by three times. S526. Instant reporting is calculated by weighting four sub-indicators: print job waiting time, number of times printer temperature exceeds limits, job deadline default rate, and success rate of blocking permission violations. The weights are dynamically adjusted based on the current system operation phase. Increase the weighting coefficients of waiting time and deadline default rate during periods of high system load; Increase the weighting factor for temperature over-limit and stroke conflict during low system load phases; S527. When updating policy network parameters, the permission verification result is introduced as a policy constraint mask. The task status corresponding to the denied access category directly blocks all action outputs. The restricted access category blocks task migration and task splitting actions. The temporary authorization category blocks task splitting actions. The normal access category opens all action space. S528. For every thousand policy network parameter updates completed, the deviation between the actual completion delay, actual resource consumption, and actual number of security events and the predicted values ​​in the task execution feedback is used to calculate the policy deviation correction amount, and the target policy network parameters are further corrected. The correction amount is proportional to the deviation amount. S529. The policy network outputs the action probability distribution, which is sampled after temperature parameter adjustment to determine the final control action. The initial value of the temperature parameter is 1.0, which decays exponentially to 0.1 with the increase of training rounds. In the early stage, it encourages the exploration of diverse actions, and in the later stage, it converges to the optimal action selection to generate the printing task control strategy.

[0032] In this embodiment, S6 specifically includes: S61. Real-time collection of feedback data during the execution of printing tasks. The feedback data includes task execution status, device operation status, node communication status and access log information. The data is then used to verify the consistency of the task identifier, digital signature and timestamp in the identifier data chain and a verification result is generated. S62. The feedback data is fused with the node state vector of the digital twin topology graph in a time sequence, and a historical state memory unit is introduced to correct the node state drift and generate updated node state features. S63. Calculate the node credibility based on the verification results and node status characteristics, map abnormal nodes and related nodes to the digital twin topology graph, update the credibility status of adjacent nodes through the graph neighborhood propagation mechanism, and form a global credibility status distribution. S64. Based on the updated node trust status, dynamically adjust the access permissions and data synchronization strategies of the corresponding nodes in the identifier data chain, write the update results into the blockchain smart contract, and synchronously update the node status of the digital twin topology graph to complete the closed loop of cloud printer data security protection.

[0033] refer to Figure 4 A blockchain-based cloud printer data security protection system includes: The data acquisition module is used to collect multi-source data during the operation of the cloud printer and perform preprocessing. The twin topology module is used to construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node-related topology structure, and using a graph convolutional network to update node state features. The data storage module is used to perform blockchain data storage processing based on the digital twin topology diagram. It uses a hierarchical hash chain to segment and map the printing task data, uses an elliptic curve cryptography algorithm to generate digital signatures, and writes the digital signatures into blockchain nodes to form an identifier data chain. The permission verification module is used to extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. The task control module is used to generate printing task control policies based on the permission verification results, establish a mapping relationship between task status and control actions using reinforcement learning algorithms, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute printing tasks through distributed control. The status update module is used to collect feedback data in real time during the execution of printing tasks, verify the consistency of the feedback data with the identification data chain, update the status of nodes in the digital twin topology graph, and complete the data security protection of the cloud printer.

[0034] Example 1: To verify the feasibility of this invention in practice, it was applied to a large-scale cloud printing service scenario. This scenario deployed 120 cloud printers, 580 mobile smart terminals, and had 2136 registered users. On average, it processed approximately 18,640 print jobs daily, printing approximately 238,500 pages per day. The system adopted a unified cloud management model. Mobile smart terminals sent print jobs to the cloud printers via the network. Print file types included text files, image files, spreadsheet files, and mixed documents. Due to the simultaneous access of a large number of terminals, problems such as identity theft, job tampering, device malfunctions, unauthorized access, and uneven node load arose during the transmission, storage, and execution of print jobs. Statistics show that under the traditional cloud printing management method, there are an average of 129 identity authentication anomalies, 74 duplicate task submissions, 58 task execution failures, 46 abnormal access behaviors, and 31 print data integrity verification failures per day. The average task response time is 3.84 seconds, the average task waiting time is 2.61 seconds, and the device load standard deviation is 0.37. Some printing devices are in a high-load state for a long time, while the resource utilization of some devices is insufficient, resulting in low overall operating efficiency. At the same time, since the print logs are only stored on the central server, there is a lack of reliable traceability of the entire task execution process. When abnormal access or print data is tampered with, it is impossible to quickly locate the abnormal node and the source of abnormal data.

[0035] By employing this invention, a data acquisition module is deployed within all cloud printers to collect real-time data on device operating status, print job execution, network communication, and identity information and print file data uploaded by mobile smart terminals. Multi-source data synchronization is achieved through a unified time benchmark. Subsequently, a digital twin topology is constructed, mapping cloud printers, mobile smart terminals, and users as digital entity nodes. A graph convolutional network is used to continuously update node status features, ensuring that each print job is associated with its corresponding device, terminal, and user status. Upon entering the system, the print job's header information, content digest, control commands, and execution logs undergo hierarchical hashing. A Merkle tree is used to generate the task root hash, and the national cryptographic SM2 elliptic curve algorithm is employed to generate a digital signature. The digital signature and task identifier are written to the blockchain to form an identifier data chain, achieving trusted evidence storage throughout the entire print job process. When a mobile smart terminal initiates another print request, the system automatically extracts the identity features from the identifier data chain and combines them with the node status features in the digital twin topology. A dual-channel feature fusion network is used to complete permission verification, dynamically generating the corresponding print permission configuration. For print tasks that have passed authorization verification, the reinforcement learning task scheduling model generates print task control strategies based on task status, node load, and device operating status. Through a dual-strategy network, it continuously optimizes the mapping relationship between task status and control actions, dynamically allocating print tasks to cloud printer nodes with lower loads. During the printing process, feedback data is continuously collected and its consistency is verified against the identifier data chain in the blockchain. When abnormal node status, communication, or data consistency is detected, the node status in the digital twin topology is immediately updated, and the node trustworthiness and print task control strategy are recalculated, achieving dynamic security protection and closed-loop control of the printing process.

[0036] Table 1. Statistical Table of the Implementation Effect of Cloud Printer Data Security Protection Methods

[0037] As shown in Table 1, after continuous verification through a large number of printing tasks, a total of 223,680 blockchain identification data units were generated during the testing process. The digital signature verification success rate reached 99.996%, and the identification data chain consistency verification success rate reached 99.992%. The system identified 1,427 abnormal access requests, all of which were successfully blocked. Among them, the identity impersonation interception rate reached 99.31%, the abnormal node identification accuracy rate reached 98.94%, the print data tampering detection rate reached 99.47%, and the false alarm rate was only 0.52%. After policy iteration, the reinforcement learning scheduling model reduced the average task response time from 3.84 seconds to 1.39 seconds, the average task waiting time from 2.61 seconds to 0.83 seconds, the node load standard deviation from 0.37 to 0.11, the average device resource utilization rate from 63.8% to 91.6%, the on-time task completion rate from 91.8% to 99.2%, and the print task migration success rate to 98.8%. No data leakage events due to abnormal permissions occurred during continuous operation. Simultaneously, the digital twin topology map can reflect the real-time operating status of each node. When node trustworthiness decreases, the system can complete permission adjustments and task rescheduling within an average of 0.42 seconds, preventing abnormal nodes from participating in print task execution and ensuring the stable operation of the entire cloud printing network.

[0038] As can be seen from the above implementation results, the present invention not only ensures the authenticity, integrity and traceability of printed data, but also significantly improves the utilization rate of printing resources and the efficiency of task execution. At the same time, it effectively reduces the impact of identity impersonation, data tampering, abnormal access and node failure on the cloud printing system, verifying that the present invention has good engineering application value and stable data security protection capabilities.

[0039] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A blockchain-based method for protecting the data security of cloud printers, characterized in that, Includes the following steps: S1. Collect multi-source data during the operation of the cloud printer and perform preprocessing; S2. Construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node association topology structure, and using a graph convolutional network to update node state features. S3. Perform blockchain data storage processing based on the digital twin topology diagram, use a hierarchical hash chain to segment and map the printing task data, use elliptic curve cryptography to generate digital signatures, and write the digital signatures into blockchain nodes to form an identifier data chain. S4. Based on the identifier data chain, extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. S5. Generate a print task control strategy based on the permission verification result, use reinforcement learning algorithm to establish a mapping relationship between task status and control action, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute print tasks through distributed control. S6. Collect feedback data in real time during the execution of printing tasks, verify the consistency of feedback data with the identification data chain, update the node status of the digital twin topology graph, and complete the data security protection of the cloud printer.

2. The method for data security protection of cloud printers based on blockchain according to claim 1, characterized in that, The multi-source data specifically includes device operating status data, print task execution data, network communication data collected by sensors deployed inside the cloud printer, as well as user identity data, print file data, and task control command data uploaded through mobile smart terminals. Among them, the device operating status data includes printer temperature parameters, mechanical motion parameters, power status parameters, and communication link status parameters; the print task execution data includes task number, file summary, print parameters, and execution time information; and the user identity data includes user identifier, terminal identifier, and access record information.

3. The method for data security protection of cloud printers based on blockchain according to claim 1, characterized in that, The preprocessing specifically includes timestamp calibration of multi-source data, synchronizing sensor-collected data and mobile smart terminal-uploaded data based on a unified time reference, completing missing data using neighborhood interpolation, removing abnormally fluctuating data using sliding window detection, performing normalization encoding on continuous data, converting different types of data into a unified format data feature set, establishing data index identifiers, and writing the mapping relationship between the processed data feature set and the original data into a secure storage area.

4. The method for data security protection of cloud printers based on blockchain according to claim 1, characterized in that, S2 specifically includes: S21. Construct a digital twin topology diagram. With the cloud printer device number as the root node, deploy a digital twin in the cloud that corresponds one-to-one with the cloud printer. The status refresh cycle of the twin is consistent with the sensor sampling cycle. Registered mobile smart terminals in the same local area network are used as first-level child nodes, and user accounts authenticated by the terminals are used as second-level child nodes to form a tree-like topology structure. S22. Node attributes include the cloud printer's temperature threshold, power supply voltage, communication packet loss rate, mobile smart terminal's MAC address, operating system version, Bluetooth signal strength, as well as the user account's login time period, historical printed pages, and file type distribution. S23. The edge relationships between nodes are determined based on the data flow direction, including: A device communication edge is established between the cloud printer and the mobile smart terminal, and the edge weight is a weighted value of communication latency and handshake success rate; An identity binding edge is established between the mobile smart terminal and the user account, and the edge weight is the ratio of the number of successful authentications to the number of failed authentications. A task scheduling edge is established between the cloud printer and the print task queue, and the edge weight is the normalized product of the task waiting time and the priority value. S24. A two-layer graph convolutional network is used to iteratively update the node state. The first layer aggregates the original attribute features of the one-hop neighboring nodes, and the second layer aggregates the hidden features of the two-hop neighboring nodes. When updating each layer, the state of the node at the previous time step is introduced as a gating input, and the state feature vector of the node at the current time step is output. The update frequency is synchronized with the sensor sampling period. After each node state update is completed, the result is pushed to the blockchain smart contract storage.

5. A method for protecting cloud printer data security based on blockchain according to claim 1, characterized in that, S3 specifically includes: S31. Divide the print task data into layers according to file header information, content summary, control instructions and execution log. Perform hash operation on each layer of data to generate the corresponding layer summary. Then, concatenate the layer summaries according to the task execution time order to form a first-level time-series hash chain. S32. Extract four key metadata categories from the data at each level: device identifier, user identifier, timestamp, and task status. Perform association hash calculation on the extracted metadata to generate a secondary association summary that describes the data source and node relationship. S33. Combine the hierarchical digest and the secondary association digest in the first-level time-series hash chain as Merkel leaf nodes to generate the task root hash, and establish a bidirectional mapping relationship between the task data segment, metadata digest and task root hash. S34. A signature mechanism based on the national cryptographic SM2 elliptic curve encryption algorithm is adopted. The task root hash is concatenated with the state feature vector of the corresponding node in the digital twin topology graph to form a signature message body. The node private key is used to generate a digital signature, and random perturbation parameters are generated based on the node's historical signature records to enhance the signature random number generation process. S35. Encapsulate the digital signature, task root hash, hierarchical digest set and Merkel proof path into an identifier data unit and write it into the corresponding data channel of the blockchain network according to the task initiation time order. S36. Based on the blockchain consensus mechanism, the identifier data unit is verified. The hash pointer of the identifier data unit of the previous block, the timestamp of the current block, the consensus node signature, and the set of identifier data units of the current task are written into the block body to form a continuous and related identifier data chain. S37. Establish an index mapping relationship between the identifier data chain and the cloud printer node, mobile smart terminal node and user node number in the digital twin topology diagram, and configure digital signature verification rules in the smart contract. S38. When a new printing task request is received, the smart contract is invoked to read the state feature vector of the corresponding node, and the consistency verification is performed with the signature state in the historical identifier data unit. Based on the verification result, a task data on-chain control instruction is generated, and the identifier data chain of the corresponding node is updated.

6. A method for protecting cloud printer data security based on blockchain according to claim 5, characterized in that, Specifically, S33 includes: S331. Take the four hierarchical digests (file header information digest, content digest, control instruction digest, and execution log digest) in the first-level time-series hash chain, and the four metadata digests (device identifier digest, user identifier digest, timestamp digest, and task status digest) in the second-level association digest, for a total of eight digest values ​​as leaf nodes of the Merkle tree. The leaf node arrangement order is fixed as hierarchical digest first and metadata digest last. The hash calculation of the leaf nodes adopts either left-associative or right-associative methods according to the parity of the task number, forming a differentiated tree structure. S332. Adjacent leaf nodes are paired up and hash operations are performed to generate parent nodes. When the number of parent nodes is odd, the last node is copied to complete the pairing. Aggregation is performed layer by layer until the root node hash value is generated as the task root hash. S333. When establishing a bidirectional mapping relationship, the forward mapping points from the task root hash to the eight leaf node digest values ​​and the original data segment storage address, and the reverse mapping points from each leaf node digest value to the task root hash and other leaf node digest values ​​of the same task. The mapping relationship table is stored in the blockchain state database. The key-value pair structure is the task root hash prefix plus the leaf node index number as the key, and the leaf node digest value plus the data segment offset as the value. S334. When any data segment is tampered with, the task root hash is queried in reverse through the leaf node digest value of the data segment, the hashes of all parent nodes on the root path from the tampered segment are recalculated, and compared with the original Merkel proof path stored in the identifier data chain to locate the tampering location and trigger a data integrity alarm.

7. A method for protecting cloud printer data security based on blockchain according to claim 5, characterized in that, S34 specifically includes: S341. Perform a bitwise XOR concatenation between the 256-bit binary string of the task root hash and the state feature vector of the corresponding node in the digital twin topology graph. The state feature vector contains four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. Each dimension is quantized and encoded into a 64-bit binary string, and the concatenation forms a 512-bit signature message body. S342. Use the national cryptographic SM2 elliptic curve encryption algorithm, select the standard recommended curve parameters, and use the signature message body as the input message to be signed; S343. During the random number generation process, the previous signature random number in the node's historical signature record, the current timestamp millisecond value, the node's CPU utilization rate, and the remaining memory capacity are used as four-element perturbation sources. The initial random number seed is generated through XOR mixing operation, and then the final signature random number is generated through iterative operation of the national cryptographic SM3 hash algorithm. This enables the same node to generate a unique digital signature for the same task root hash at different times and under different load conditions. S344. Use the node's private key to perform a scalar multiplication operation on the signed message body and a random number to generate a signature component. The signature component, together with the public key coordinates and the message body hash value, constitutes a complete digital signature. S345. Immediately after the signature is generated, the signature random number, signature timestamp, and node load parameters are written to the local secure storage area as input for the historical signature record of the next signature, forming a chain-dependent structure of random number perturbation sources.

8. A method for protecting cloud printer data security based on blockchain according to claim 1, characterized in that, S4 specifically includes: S41. Extract the signature component, public key coordinates, message body hash value and historical signature sequence of the digital signature from the identifier data chain, and parse out the identity feature vector, which includes four dimensions: signature entropy value, key usage frequency, signature time interval distribution and cross-node signature consistency. S42. Extract the nodes of mobile smart terminals and their associated user accounts from the digital twin topology graph, and parse out the corresponding state feature vectors, which include four dimensions: device health, terminal trustworthiness, user risk, and neighborhood aggregation anomaly index. S43. Construct a dual-channel feature fusion network. The identity feature channel uses one-dimensional convolution to extract the signature time sequence mode, and the state feature channel uses graph attention to extract the topological association mode. The outputs of the two channels are adaptively weighted and fused through a gated cross-attention mechanism. The gate weights are dynamically adjusted by the current network load and the historical verification accuracy. S44. Input the fused joint feature vector into the permission classifier. The permission classifier contains four output categories: normal access, restricted access, temporary authorization, and denied access. Each category corresponds to a different printing permission configuration, including the maximum number of pages that can be printed, the set of accessible file types, the allowed printing time period, and whether secondary approval is required. S45. After the permission verification result is generated, it is written to the smart contract and bound to the most recent signature record of the corresponding terminal in the identifier data chain. When the value of either the terminal credibility or the user risk in the joint feature vector exceeds the dynamic threshold, permission downgrade or revocation is automatically triggered. The downgraded permission configuration is immediately synchronized to the local access control unit of the cloud printer, and printing tasks that exceed the current permission range are refused.

9. A method for data security protection of cloud printers based on blockchain according to claim 1, characterized in that, S5 specifically includes: S51. Obtain the status features of the print task that has passed the permission verification, and build a reinforcement learning environment by combining the load status of the cloud printer node in the digital twin topology map. S52. Construct a task scheduling model based on reinforcement learning algorithm, taking the task allocation node, execution order and resource usage ratio as control actions, updating the policy network through task execution feedback, and establishing a mapping relationship between task state and control actions. S53. Evaluate candidate control actions based on task completion delay, node load balancing, equipment operating status, and safety constraints, and generate printing task scheduling parameters. S54. Dynamically adjust the task allocation strategy based on the processing capacity of the cloud printer node, the task queue status and the communication link status, and schedule the printing task to the corresponding cloud printer node for execution through distributed control. S55. Collect printing task execution feedback data, update the state characteristics and control strategies in the task scheduling model, and recalculate the scheduling parameters when the node load status or task execution status changes, so as to realize dynamic scheduling of cloud printing tasks.

10. A blockchain-based cloud printer data security protection system, implementing the blockchain-based cloud printer data security protection method according to any one of claims 1 to 9, characterized in that, include: The data acquisition module is used to collect multi-source data during the operation of the cloud printer and perform preprocessing. The twin topology module is used to construct a digital twin topology graph, mapping mobile smart terminals, users, and cloud printers as digital entity nodes, establishing a node-related topology structure, and using a graph convolutional network to update node state features. The data storage module is used to perform blockchain data storage processing based on the digital twin topology diagram. It uses a hierarchical hash chain to segment and map the printing task data, uses an elliptic curve cryptography algorithm to generate digital signatures, and writes the digital signatures into blockchain nodes to form an identifier data chain. The permission verification module is used to extract the identity features from the digital signature and the node state features from the digital twin topology graph, perform feature fusion calculation, and generate the permission verification result corresponding to the mobile smart terminal. The task control module is used to generate printing task control policies based on the permission verification results, establish a mapping relationship between task status and control actions using reinforcement learning algorithms, calculate scheduling parameters in combination with the load status of cloud printer nodes, and execute printing tasks through distributed control. The status update module is used to collect feedback data in real time during the execution of printing tasks, verify the consistency of the feedback data with the identification data chain, update the status of nodes in the digital twin topology graph, and complete the data security protection of the cloud printer.

Images