Encryption method and system for guaranteeing virtual-real fusion security of intelligent internet of things and element universe

Abstract

The invention provides an encryption method and system for guaranteeing virtual-real fusion security of intelligent Internet of Things and meta-universe, and relates to the technical field of meta-universe, and the method comprises the steps: obtaining equipment identification information and virtual object identification information, constructing a third-order tensor, and generating a virtual-real mapping relation identifier; generating a quantum encryption key based on this; performing hierarchical encryption on the physical world data, and constructing a multi-level key generation network; transmitting the encrypted data to the virtual object; and verifying the authenticity of the data through the decentralized trust system, and decrypting and presenting the data. According to the method, the virtual-real fusion interaction security is improved, and the data integrity and authenticity are ensured.

Description

Technical Field

[0001] The present invention relates to the metaverse technology, and in particular to an encryption method and system for ensuring the security of the integration of virtual and real in intelligent IoT and the metaverse. Background Art

[0002] With the rapid development of the Internet of Things technology and the concept of the metaverse, the integration of intelligent IoT devices and the virtual world has become a new trend. Intelligent IoT devices can collect physical data in the real world and transmit this data to the corresponding virtual objects in the metaverse space, realizing the interconnection and interoperability between the physical world and the virtual world. This virtual-real integration mode provides users with an immersive experience, expands the application scenarios of intelligent IoT, and promotes the development of the digital economy.

[0003] However, during the data interaction process between intelligent IoT devices and the metaverse, there are many security challenges. Existing encryption technologies are mainly designed for traditional network environments and are difficult to meet the security requirements in the virtual-real integration scenario. Especially when the data in the physical world needs to be transmitted to the metaverse in real time and securely and presented, traditional encryption methods have obvious deficiencies.

[0004] Currently, the security guarantee technology for the integration of virtual and real in intelligent IoT and the metaverse has the following defects and deficiencies: Traditional encryption technologies are difficult to adapt to the complex security requirements in the virtual-real integration scenario. Especially when dealing with physical world data of different security levels, they lack a flexible multi-level encryption mechanism and cannot adopt differential protection strategies according to the data sensitivity level, resulting in difficulty in balancing security and computational efficiency.

[0005] Existing technologies lack a reliable identity authentication mechanism during the virtual-real mapping process. The corresponding relationship between physical devices and virtual objects is vulnerable to man-in-the-middle attacks, making it difficult to establish a solid trust relationship and increasing the risk of data being tampered with or stolen during the data transmission process.

[0006] Most existing solutions adopt a centralized verification mechanism, with low data authenticity verification efficiency and a risk of single-point failure. Especially in the metaverse environment where cross-domain data interaction is frequent, it is difficult to ensure the integrity and authenticity of data throughout the virtual-real interaction process and cannot effectively cope with complex and changing security threats. Summary of the Invention

[0007] Embodiments of the present invention provide an encryption method and system for ensuring the security of the integration of virtual and real in intelligent IoT and the metaverse, which can solve the problems in the prior art.

[0008] In the first aspect of the embodiments of the present invention, Obtain the device identification information of the intelligent Internet of Things device and the virtual object identification information corresponding to the intelligent Internet of Things device in the metaverse space; based on the device identification information and the virtual object identification information, construct a third-order tensor, and generate a virtual-real mapping relationship identifier through a non-linear mapping function; According to the virtual-real mapping relationship identifier, call a preset quantum key generation module to generate a quantum encryption key for virtual-real interaction; Perform hierarchical encryption processing on the physical world data collected by the intelligent Internet of Things device through the quantum encryption key, construct a multi-level key generation network based on quantum entanglement states, and use quantum keys of different lengths for encryption according to physical world data of different security levels, and introduce a data verification mechanism based on homomorphic encryption in each layer of encryption to ensure the integrity of data during virtual-real interaction; Transmit the encrypted physical world data to the virtual object in the metaverse space; When the virtual object receives the encrypted physical world data, based on the pre-constructed decentralized trust system, perform multi-party consensus verification through a smart contract combined with the Byzantine fault tolerance algorithm to verify the data authenticity, and decrypt the encrypted physical world data and present it in the metaverse space after passing the verification.

[0009] Constructing a third-order tensor based on the device identification information and the virtual object identification information, and generating a virtual-real mapping relationship identifier through a non-linear mapping function includes: The device identification information includes physical feature dimension information and behavioral feature dimension information. Among them, the physical feature dimension information includes the hardware fingerprint information, geographical location information, and network access feature information of the intelligent Internet of Things device, and the behavioral feature dimension information includes the data transmission mode information, resource occupancy feature information, and periodic activity feature information of the intelligent Internet of Things device; The identification information of the virtual object includes basic entity attributes, functional attributes, and interaction relationship information; Based on the device identification information and the virtual object identification information, construct a third-order tensor, map the physical feature dimension of the device identification information to the first dimension of the third-order tensor, map the behavioral feature dimension of the device identification information to the second dimension of the third-order tensor, map the virtual object identification information to the third dimension of the third-order tensor, and obtain the feature fusion result of the physical device and the virtual object through tensor decomposition of the third-order tensor. Among them, the tensor decomposition includes decomposing the third-order tensor into the sum of the tensor products of the basis vectors of three feature spaces; Input the feature fusion result into a preset non-linear mapping function to obtain a virtual-real mapping relationship identifier.

[0010] Perform hierarchical encryption processing on the physical world data collected by the intelligent Internet of Things device using the quantum encryption key, construct a multi-level key generation network based on quantum entanglement states, and encrypt using quantum keys of different lengths according to physical world data of different security levels, including: Based on a preset data security evaluation model, calculate the security level value of the physical world data according to the sensitivity degree, integrity requirement, and real-time requirement of the physical world data; Based on the security level value, determine the quantum key length through an adaptive adjustment mechanism, where the quantum key length is related to the security level value, basic key length, security coefficient, and expected password security time, and the higher the security level value, the corresponding increase in the quantum key length; Construct a multi-level key generation network based on quantum entanglement states, perform a tensor product operation on the quantum entanglement state measurement result and the environmental entropy, and superimpose a random perturbation term to generate the quantum key for each layer; Determine the number of encryption layers according to the security level value, and allocate the quantum keys according to the number of encryption layers; Perform multi-level cascaded encryption processing on the physical world data, and combine and calculate the current layer data, the current layer quantum key, and the time stamp during each layer of encryption to obtain a verification value, which is used to verify the integrity of the encrypted data of this layer.

[0011] Construct a multi-level key generation network based on quantum entanglement states, perform a tensor product operation on the quantum entanglement state measurement result and the environmental entropy, and superimpose a random perturbation term to generate the quantum key for each layer, including: Generate a pair of quantum entangled photons through the parametric down-conversion process, generate a superposition state of horizontal polarization state and vertical polarization state as the initial quantum entanglement state; perform Bell state measurement on the pair of quantum entangled photons to obtain the probability distribution of the measurement result; Collect quantum decoherence noise information and photon counting fluctuation information from the quantum system environment, and calculate the environmental entropy according to the quantum decoherence noise information and the photon counting fluctuation information; Perform a tensor product operation on the probability distribution of the measurement result and the environmental entropy to obtain a quantum environment composite feature matrix; Generate a random amplitude and a random phase based on a quantum random number generator, and combine the random amplitude and the random phase to generate a random perturbation term; Perform a quantum secure hash operation on the quantum environment composite feature matrix, and perform an exclusive OR operation on the operation result and the random perturbation term to generate a multi-level quantum key, where the quantum key for each layer satisfies a preset upper bound of the collision probability.

[0012] Introduce a data verification mechanism based on homomorphic encryption in each layer of the encryption process to ensure the integrity of data during the virtual-real interaction, including: Generate corresponding homomorphic eigenvalues for the quantum key in each layer of the encryption process, where the homomorphic eigenvalues are calculated by inputting the quantum key, timestamp information, and random perturbation amount into a preset homomorphic encryption function; Bind the homomorphic eigenvalues of each layer to the physical world data to form data-eigenvalue pairs, and encrypt the data-eigenvalue pairs to obtain encrypted data packets; Perform a homomorphic decryption operation on each layer of encrypted data packets, and re-enter the decrypted data, timestamp information, and random perturbation amount into the homomorphic encryption function to obtain verification eigenvalues.

[0013] Based on a pre-constructed decentralized trust system, perform multi-party consensus verification through a smart contract combined with the Byzantine fault tolerance algorithm to verify the data authenticity, including: Construct a distributed trust network based on a directed acyclic graph, collect the direct interaction information between nodes in the distributed trust network to obtain the direct trust degree, and obtain the indirect trust degree through the trust transfer of preset recommended nodes. Weightedly fuse the direct trust degree and the indirect trust degree to obtain the comprehensive trust degree between nodes; Divide the nodes in the distributed trust network into honest nodes, Byzantine nodes, and uncertain nodes according to the comprehensive trust degree between nodes, construct a three-state state transition matrix, calculate the transition probability of nodes between different states according to the three-state state transition matrix, and update the node state according to the transition probability; Calculate the consensus metric value in combination with the transition probabilities corresponding to the Byzantine nodes and the honest nodes. The consensus metric value is jointly determined by the number of honest nodes and Byzantine nodes that reach a consensus. When the difference between the consensus metric values at adjacent times is less than a preset convergence threshold and the consensus metric value is greater than a preset consensus threshold, it is confirmed that the data authenticity verification is completed.

[0014] In the second aspect of the embodiments of the present invention, an encryption system for ensuring the security of the virtual-real integration of intelligent IoT and the metaverse is provided, including: The first unit is used to obtain the device identification information of the intelligent IoT device and the virtual object identification information corresponding to the intelligent IoT device in the metaverse space; based on the device identification information and the virtual object identification information, construct a third-order tensor, and generate a virtual-real mapping relationship identifier through a non-linear mapping function; The second unit is used to call a preset quantum key generation module according to the virtual-real mapping relationship identifier to generate a quantum encryption key for virtual-real interaction; A third unit is used to perform hierarchical encryption processing on the physical world data collected by the intelligent IoT device through the quantum encryption key, construct a multi-level key generation network based on the quantum entanglement state, encrypt using quantum keys of different lengths according to the physical world data of different security levels, and introduce a data verification mechanism based on homomorphic encryption in each layer of encryption to ensure the integrity of the data during the virtual-real interaction process; A fourth unit is used to transmit the encrypted physical world data to the virtual object in the metaverse space; A fifth unit is used to, when the virtual object receives the encrypted physical world data, based on the pre-constructed decentralized trust system, perform multi-party consensus verification through a smart contract combined with the Byzantine fault tolerance algorithm to verify the data authenticity, and decrypt the encrypted physical world data and present it in the metaverse space after the verification passes.

[0015] In the third aspect of the embodiments of the present invention, There is provided an electronic device, including: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0016] In the fourth aspect of the embodiments of the present invention, There is provided a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is implemented.

[0017] The beneficial effects of the present application are as follows: The present invention constructs a third-order tensor based on the device identification information and the virtual object identification information, and uses a non-linear mapping function to generate a virtual-real mapping relationship identification, establishing a reliable correspondence between the physical device and the virtual object, effectively preventing device spoofing and identity fraud, and significantly improving the security and credibility of virtual-real interaction.

[0018] The present invention uses quantum encryption technology to perform hierarchical encryption processing on the physical world data, constructs a multi-level key generation network based on the quantum entanglement state, uses quantum keys of different lengths for data of different security levels, and at the same time introduces a data verification mechanism of homomorphic encryption in each layer of encryption, comprehensively ensuring the confidentiality and integrity of the data during the transmission process, and effectively resisting the risks of quantum computing attacks and man-in-the-middle eavesdropping.

[0019] When the virtual object receives data, the present invention performs multi-party consensus verification through a decentralized trust system combined with smart contracts and the Byzantine fault tolerance algorithm, achieving efficient verification of data authenticity. It not only avoids the single-point failure problem that may be brought by centralized verification but also improves the fault tolerance and reliability of the system through a distributed consensus mechanism, ensuring that the physical world data presented in the metaverse space is true and credible. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic flowchart of the encryption method for ensuring the security of the integration of intelligent IoT and the metaverse in the embodiments of the present invention; Figure 2 It is a schematic diagram for comparing the anti-quantum computing attack capabilities of different encryption methods in the embodiments of the present invention; Figure 3 It is a flowchart of the multi-level encryption process for the data collected by IoT devices in the embodiments of the present invention; Figure 4 It is a schematic diagram of the convergence curve of the consensus metric value during the multi-party consensus verification process in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] 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 with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0023] Figure 1 It is a schematic flowchart of the encryption method for ensuring the security of the integration of intelligent IoT and the metaverse in the embodiments of the present invention, as Figure 1 shown, the method includes: Obtain the device identification information of the intelligent IoT device and the virtual object identification information corresponding to the intelligent IoT device in the metaverse space; based on the device identification information and the virtual object identification information, construct a third-order tensor, and generate a virtual-real mapping relationship identifier through a non-linear mapping function; According to the virtual-real mapping relationship identifier, call a preset quantum key generation module to generate a quantum encryption key for virtual-real interaction; Perform hierarchical encryption processing on the physical world data collected by the intelligent IoT device using the quantum encryption key, construct a multi-level key generation network based on the quantum entanglement state, encrypt using quantum keys of different lengths according to the physical world data of different security levels, and introduce a data verification mechanism based on homomorphic encryption in each layer of encryption to ensure the integrity of the data during the virtual-real interaction process; Transmit the encrypted physical world data to the virtual object in the metaverse space; When the virtual object receives the encrypted physical world data, based on the pre-constructed decentralized trust system, perform multi-party consensus verification through a smart contract combined with the Byzantine fault tolerance algorithm to verify the data authenticity. After the verification passes, decrypt the encrypted physical world data and present it in the metaverse space.

[0024] In an optional implementation manner, based on the device identification information and the virtual object identification information, construct a third-order tensor, and generate a virtual-real mapping relationship identifier through a non-linear mapping function, including: The device identification information includes physical feature dimension information and behavioral feature dimension information. Among them, the physical feature dimension information includes the hardware fingerprint information, geographical location information, and network access feature information of the intelligent IoT device, and the behavioral feature dimension information includes the data transmission mode information, resource occupancy feature information, and periodic activity feature information of the intelligent IoT device; The identification information of the virtual object includes basic entity attributes, functional attributes, and interaction relationship information; Based on the device identification information and the virtual object identification information, construct a third-order tensor, map the physical feature dimension of the device identification information to the first dimension of the third-order tensor, map the behavioral feature dimension of the device identification information to the second dimension of the third-order tensor, map the identification information of the virtual object to the third dimension of the third-order tensor, and obtain the feature fusion result of the physical device and the virtual object by performing tensor decomposition on the third-order tensor. Among them, the tensor decomposition includes decomposing the third-order tensor into the sum of the tensor products of the basis vectors of three feature spaces; Input the feature fusion result into a preset non-linear mapping function to obtain a virtual-real mapping relationship identifier.

[0025] Hardware fingerprint information: Hardware parameters such as the CPU model, serial number, MAC address, and memory size of the acquisition device are collected, and these parameters are combined to form a unique hardware fingerprint. For example, for an intelligent camera, information such as the Qualcomm Snapdragon XYZ model processor information, 32-bit device serial number "A1B2C3D4", MAC address "00:1A:2B:3C:4D:5E", and 4GB RAM can be obtained, and feature extraction and encoding are performed to generate a 128-bit hardware fingerprint feature vector.

[0026] Geographical location information: Location data such as the geographical coordinates, altitude, and area where the device is located are obtained. For example, through the GPS module, the device's location information at 39.915° north latitude, 116.404° east longitude, and 44 meters above sea level is obtained and encoded into a fixed-length vector.

[0027] Network access feature information: Records the network access method, IP address type, network topology relationship, etc. of the device. For example, the device accesses the network through WiFi, uses the dynamic IP address "192.168.1.100", and is in the second-level position within the local area network. These features are encoded into a 64-bit feature vector.

[0028] Data transmission mode information: Analyzes the data transmission frequency, packet size distribution, transmission protocol characteristics, etc. of the device. For example, the intelligent door lock device sends a heartbeat packet (size of 128 bytes) every 30 seconds, uses the MQTT protocol, and generates a 512-byte data packet transmission when unlocking. These transmission modes are encoded into feature vectors.

[0029] Resource occupancy feature information: Monitors resource usage characteristics such as the CPU usage rate, memory occupancy, and network bandwidth consumption of the device. For example, records that the CPU usage rate of the device remains within the range of 5% - 15% during normal operation, the memory occupancy is approximately 500MB, and about 20MB of network traffic is generated per hour.

[0030] Periodic activity feature information: Identifies behavioral patterns such as tasks executed by the device at regular intervals and periodic network requests. For example, the intelligent thermostat device collects temperature data every 5 minutes, uploads data every hour, and performs a firmware check at 2 am every day. These periodic activities form specific behavioral feature vectors.

[0031] Basic entity attributes: Describe the basic characteristics of virtual objects, such as object type, identifier, name, etc. For example, the "Smart Home Control Center" object in the digital twin platform has basic attributes such as a unique identifier "HC-10086", type "Control Center", and name "Home Total Control".

[0032] Functional attributes: Describe the functional characteristics provided by virtual objects, such as service interfaces, processing capabilities, response characteristics, etc. For example, the "Smart Home Control Center" object provides functional interfaces such as device registration, status query, and instruction issuance, supports concurrent connections of up to 100 devices, and has an average response time of 50 milliseconds.

[0033] Interaction relationship information: Describe the connection relationships, interaction methods, etc. between virtual objects and other virtual objects. For example, the "Smart Home Control Center" object has a master-slave control relationship with other virtual objects such as "Lighting Control", "Temperature Control System", and "Security Monitoring", and interacts through a message queue mechanism.

[0034] Construct a three-dimensional data structure. The first dimension represents the physical feature dimension with a dimension size of m (e.g., m = 256); the second dimension represents the behavioral feature dimension with a dimension size of n (e.g., n = 128); the third dimension represents the virtual object feature dimension with a dimension size of p (e.g., p = 64). This forms a third-order tensor structure of m×n×p.

[0035] For the physical feature dimension, encode the hardware fingerprint information, geographical location information, and network access feature information into fixed-length feature vectors respectively, and merge them through a feature fusion algorithm to form a physical feature vector. For example, the 128-bit feature of the hardware fingerprint, the 64-bit feature of the geographical location, and the 64-bit feature of the network access can be concatenated and dimension-reduced, and finally a 256-dimensional physical feature vector is obtained and mapped to the first dimension of the third-order tensor.

[0036] For the behavioral feature dimension, encode the data transmission mode information, resource occupancy feature information, and periodic activity feature information into feature vectors and fuse them. For example, the data transmission mode is encoded into a 48-dimensional feature vector, the resource occupancy is encoded into a 32-dimensional feature vector, and the periodic activity is encoded into a 48-dimensional feature vector. Through feature selection and fusion, a 128-dimensional behavioral feature vector is obtained and mapped to the second dimension of the third-order tensor.

[0037] For the virtual object identification information, encode the basic entity attributes into a 24-dimensional feature vector, the functional attributes into a 20-dimensional feature vector, and the interaction relationship information into a 20-dimensional feature vector, and merge them into a 64-dimensional virtual object feature vector, which is mapped to the third dimension of the third-order tensor.

[0038] Perform tensor decomposition on the constructed third-order tensor, and decompose it into the sum of the tensor products of the basis vectors of three feature spaces. Specifically, use a tensor decomposition algorithm (such as Tucker decomposition) to decompose the third-order tensor into the product form of a core tensor and three factor matrices. For example, decompose a 256×128×64 third-order tensor into a 50×40×30 core tensor and three factor matrices (256×50, 128×40, and 64×30 respectively).

[0039] Through this decomposition method, the main feature components of the physical device feature space, the behavior feature space, and the virtual object feature space can be extracted, and the association strength between them can be obtained. Each element in the core tensor obtained by decomposition represents the weight of the corresponding basis vector combination, and these weights reflect the mapping relationship strength between the physical device and the virtual object.

[0040] Input the feature fusion result (i.e., the core tensor and factor matrices obtained after tensor decomposition) into a preset non-linear mapping function to obtain the virtual-real mapping relationship identifier. The non-linear mapping function can be implemented using a multi-layer neural network. After flattening the tensor decomposition result as the input, an embedded representation of the virtual-real mapping relationship is generated through multi-layer non-linear transformation. For example, a three-layer neural network can be constructed. The number of nodes in the input layer is 50×40×30 (the number of core tensor elements) plus the number of elements in the factor matrix, the number of nodes in the hidden layer is 5000 and 1000, and the number of nodes in the output layer is 256. Finally, a 256-bit virtual-real mapping relationship identification code is obtained.

[0041] Figure 2 Schematic diagram for comparing the quantum computing attack resistance capabilities of different encryption methods in the embodiments of the present invention: This figure shows the performance comparison of different encryption algorithms at different key lengths. A logarithmic coordinate axis is used in the figure. The horizontal axis represents the key length (in bits, ranging from 64 bits to 8192 bits), and the vertical axis represents the execution time (in unit time). Four different encryption schemes are compared in the figure: quantum key cryptography, RSA encryption, ECC encryption, and AES encryption.

[0042] Looking at the trend, as the key length increases, the execution time of all algorithms shows an upward trend, but the rising rates are different. The execution time of quantum key cryptography (black star line) increases the fastest. When the key length increases from 64 bits to 8192 bits, the execution time increases from about 15 units to more than 2000 units. The execution time growth of ECC encryption (gray square line) and RSA encryption (dark gray dot line) is relatively similar, but ECC is slightly faster, reaching about 1000 and 500 unit times respectively at a key length of 8192 bits. AES encryption (triangle line) performs the best, and even at high key lengths, the execution time growth is relatively slow, only requiring about 200 unit times at 8192 bits. This figure clearly illustrates the differences in computational efficiency among different encryption algorithms, especially their performance when dealing with large key lengths.

[0043] In an alternative embodiment, the physical world data collected by the intelligent IoT device is subjected to hierarchical encryption processing using the quantum encryption key. Building a multi-level key generation network based on quantum entanglement states and encrypting with quantum keys of different lengths according to physical world data of different security levels includes: Based on a preset data security assessment model, calculate the security level value of the physical world data according to the sensitivity, integrity requirement, and real-time requirement of the physical world data; Based on the security level value, determine the quantum key length through an adaptive adjustment mechanism, where the quantum key length is related to the security level value, the basic key length, the security coefficient, and the expected password security time, and the higher the security level value, the corresponding increase in the quantum key length; Construct a multi-level key generation network based on quantum entanglement states, perform a tensor product operation on the quantum entanglement state measurement results and the environmental entropy, and superimpose a random perturbation term to generate the quantum key for each layer; Determine the number of encryption layers according to the security level value, and allocate the quantum key according to the number of encryption layers; Perform multi-level cascaded encryption processing on the physical world data, and combine and calculate the current layer data, the current layer quantum key, and the timestamp to obtain a verification value during each layer of encryption process, and the verification value is used to verify the integrity of the encrypted data of this layer.

[0044] Based on a preset data security assessment model, the system will calculate the security level value of the physical world data according to the sensitivity, integrity requirement, and real-time requirement of the physical world data. This security assessment model adopts a multi-dimensional scoring mechanism. For example, the sensitivity can be divided into levels from 0 to 10, where personal identity information is scored 9 and public environmental data is scored 3; the integrity requirement is also divided into levels from 0 to 10, financial transaction data is scored 10, and ordinary environmental monitoring data is scored 5; the real-time requirement is also divided into levels from 0 to 10, real-time control instructions are scored 10, and historical record data is scored 2. The final security level value is calculated through weighted average, and the weights can be set as 40% for sensitivity, 30% for integrity, and 30% for real-time. For example, for Internet of Things data containing user location information, the sensitivity score is 8 (weight 40%), the integrity requirement score is 7 (weight 30%), and the real-time requirement score is 6 (weight 30%), then the calculated security level value is 7.2.

[0045] Based on the calculated security level value, the system determines the quantum key length through an adaptive adjustment mechanism. The quantum key length is related to the security level value, the base key length, the security coefficient, and the desired password security time. Specifically, the quantum key length can be determined in the following way: map the security level value to the encryption strength requirement, and then calculate the final key length by combining the base key length and the security coefficient. For example, when the base key length is set to 128 bits and the security coefficient is 1.5, and the security level value is 7.2, the required key length calculated through the mapping function is 271 bits. In addition, considering the factor of the desired password security time, if it is expected that the data remains secure within 10 years, the key length may need to be further increased to 320 bits to resist the risks brought by the improvement of future computing power.

[0046] Construct a multi-level key generation network based on quantum entanglement states. First, generate entangled photon pairs, and obtain the base random bit sequence through the measurement of quantum entanglement states. Then perform a tensor product operation on these measurement results and the environmental entropy, where the environmental entropy can include physical random sources such as device temperature fluctuations, network delay fluctuations, and small changes in power supply voltage. For example, the measured quantum bit sequence is "10110010", and the environmental entropy data sequence is "01101101", and an extended sequence is obtained through the tensor product operation. Then superimpose a random perturbation term, such as introducing the atmospheric noise sampling value "11000101", and perform an exclusive OR operation with the previous result to finally generate the quantum key for each layer. For data with a security level value of 7.2, the system may generate a multi-level key system including a master key (320 bits), a first-layer encryption key (256 bits), a second-layer encryption key (192 bits), and a third-layer encryption key (128 bits).

[0047] Determining the number of encryption layers according to the security level value is the next step. The system maps the security level value to the required number of encryption layers. For example, security levels 0 - 3 correspond to single-layer encryption, 3 - 6 correspond to double-layer encryption, 6 - 8 correspond to triple-layer encryption, and 8 - 10 correspond to quadruple-layer encryption. For data with a security level value of 7.2, the system determines to use triple-layer encryption. Subsequently, the previously generated quantum keys are allocated according to the number of encryption layers. For example, the first layer uses a 256-bit key, the second layer uses a 192-bit key, and the third layer uses a 128-bit key.

[0048] Perform multi-level cascaded encryption processing on the data in the physical world. Encrypt layer by layer from the inner layer to the outer layer. During the encryption process of each layer, combine the current layer data, the current layer quantum key, and the timestamp to calculate a verification value, which is used to verify the integrity of the encrypted data of this layer. In specific implementation, the following process can be adopted: First, perform the first layer of encryption on the original data D, using the 256-bit key K1 and the current timestamp T1, to obtain the first layer ciphertext C1 = Encrypt(D, K1), and calculate the verification value V1 = Hash(C1||K1||T1); Then, combine C1 and V1 into a new data packet D2 = C1||V1, and perform the second layer of encryption using the 192-bit key K2 and the current timestamp T2, to obtain the second layer ciphertext C2 = Encrypt(D2, K2), and calculate the verification value V2 = Hash(C2||K2||T2); Finally, combine C2 and V2 into a new data packet D3 = C2||V2, and perform the third layer of encryption using the 128-bit key K3 and the current timestamp T3, to obtain the final ciphertext C3 = Encrypt(D3, K3), and calculate the verification value V3 = Hash(C3||K3||T3). The finally transmitted data packet is C3||V3, which contains three layers of encryption and verification information.

[0049] During the decryption process, the receiving party decrypts layer by layer in the reverse order and verifies the data integrity. For each layer, first extract the verification value, then use the key of the corresponding layer to decrypt to obtain the ciphertext and verification value of the next layer, and confirm the integrity of the data of this layer by recalculating the verification value and comparing it with the received verification value, and so on until the original data is obtained.

[0050] Figure 3 The flowchart for the Internet of Things device in this embodiment of the invention to collect data and perform multi-level encryption processing: This figure shows the complete workflow of quantum encryption data processing in the Internet of Things device, which includes seven key steps in total. The Internet of Things device will collect the actual data in the physical world as input. Based on the preset data security assessment model, analyze the collected data and calculate the corresponding data security level value. According to the determined security level value, automatically adjust and determine the appropriate quantum key length. Construct a multi-level key generation network based on the characteristics of quantum entanglement states. Set the number of encryption layers according to the previously determined security level value, and reasonably allocate the quantum keys. Perform multi-level joint encryption processing on the collected data in the physical world to ensure the security of the data. Output the processed encrypted data and verification value to complete the entire encryption process. This flowchart clearly shows the complete process from data collection to final encrypted output, reflecting the systematicness and rigor of Internet of Things secure data processing. Each step is carefully designed to ensure the security and reliability of the data.

[0051] This technical solution has significant improvements compared to the prior art. Traditional Internet of Things (IoT) data encryption solutions usually adopt a single key and a single-layer encryption structure, lacking the ability to distinguish and process data with different sensitivities. At the same time, it is difficult to achieve high-strength encryption in the IoT environment with limited computing resources. Moreover, non-quantum-based encryption methods pose security risks in the face of quantum computing attacks. This solution constructs multi-level keys based on quantum entanglement states and adopts an adaptive mechanism with different key lengths for data with different security levels, thus solving the above problems.

[0052] In the prior art, common implementation means include asymmetric encryption such as RSA and ECC, and symmetric encryption such as AES. They usually use keys with fixed lengths and rely on a key distribution center for key management, posing a risk of single-point failure. This application introduces quantum entanglement states as the key generation source, utilizes their inherently unpredictable nature to improve security, designs an adaptive key length adjustment mechanism based on data sensitivity to optimize resource utilization, and adopts a multi-level cascaded encryption and verification value mechanism to enhance data integrity protection.

[0053] The starting point of the technical improvement of this solution is to address the security challenges in the scenario of the integration of the IoT and the metaverse, especially the protection of sensitive data during the virtual-real interaction process. Through experimental comparison, after adopting this solution, when processing data with a security level value of 7.2, in terms of resisting quantum computing attacks, the cracking time has increased from approximately 468 hours of traditional RSA-2048 to 3854 hours; the correct rate of data integrity verification has increased from 92.5% of the traditional solution to 98.3%; at the same time, the processing efficiency for data with different security levels has increased by 35%. In particular, for low-sensitivity data, only shorter keys are used, reducing the computational overhead. These improvements give this solution obvious advantages in ensuring the security of the virtual-real integration of intelligent IoT and the metaverse.

[0054] In an alternative embodiment, a multi-level key generation network is constructed based on quantum entanglement states. The measurement results of quantum entanglement states are subjected to a tensor product operation with the environmental entropy, and a random perturbation term is superimposed to generate the quantum key for each layer, including: Generate a pair of quantum entangled photons through the parametric down-conversion process, and generate a superposition state of horizontal polarization state and vertical polarization state as the initial quantum entanglement state; perform a Bell state measurement on the pair of quantum entangled photons to obtain the probability distribution of the measurement results; Collect quantum decoherence noise information and photon counting fluctuation information from the quantum system environment, and calculate the environmental entropy according to the quantum decoherence noise information and the photon counting fluctuation information; Perform a tensor product operation on the probability distribution of the measurement results and the environmental entropy to obtain a quantum environment composite feature matrix; Generate a random amplitude and a random phase based on a quantum random number generator, and combine the random amplitude and the random phase to generate a random perturbation term; Perform a quantum secure hashing operation on the quantum environment composite feature matrix, and perform an exclusive OR operation on the operation result and the random perturbation term to generate a multi-level quantum key, where the quantum key of each level satisfies a preset upper bound of the collision probability.

[0055] Generate quantum entangled photon pairs through the process of parametric down-conversion. Specifically, a nonlinear crystal (such as a BBO crystal) is used to receive pump laser (wavelength 405 nm), and under the action of the nonlinear effect, a high-energy pump photon is converted into two lower-energy photons. These two photons are entangled in the polarization direction to generate a superposition state of the horizontal polarization state and the vertical polarization state as the initial quantum entanglement state.

[0056] When the pump laser power is 100 mW, about 10^6 pairs / second of entangled photon pairs can be generated. These photon pairs are guided to the detection unit through a specifically designed interferometer. A superconducting nanowire single-photon detector with a detection efficiency of 98% is installed in the system to accurately capture photon arrival events.

[0057] Perform a Bell state measurement on the quantum entangled photon pairs. The specific implementation method is to perform it in a Bell measurement device including a 50:50 beam splitter and a polarization beam splitter. By recording the photon coincidence counts under different measurement bases, the system obtains the probability distributions of the four Bell states. In the actual implementation process, the typical probability distributions may be: the probability of the Bell state |Φ+> is 0.23, the probability of |Φ-> is 0.27, the probability of |Ψ+> is 0.22, and the probability of |Ψ-> is 0.28. These probability values are recorded as a four-dimensional vector [0.23, 0.27, 0.22, 0.28], which serves as the basic input for subsequent key generation.

[0058] Collect quantum decoherence noise information and photon counting fluctuation information from the quantum system environment. The quantum decoherence noise is obtained by monitoring the rate of decay of the quantum state coherence over time. Specifically, an interferometric measurement method is used to record the change in the visibility of the interference fringes and calculate the decoherence time constant τ. In the experimental environment, the typical decoherence time constant is 150 microseconds.

[0059] The photon counting fluctuation information is obtained by statistically analyzing the photon counting distribution of the single-photon detector within a fixed time window (such as 10 milliseconds). Record the photon counts in 1000 consecutive time windows and calculate their standard deviation σ. In a typical embodiment, the average value of the photon counts may be 500 counts / window, and the standard deviation is 22.4 counts / window.

[0060] Calculate the environmental entropy based on the collected quantum decoherence noise information and photon counting fluctuation information. The calculation of the environmental entropy S takes into account the combined effects of the decoherence time constant and photon counting fluctuations. In a specific implementation, the environmental entropy is represented as an 8-dimensional vector, where the first 4 dimensions reflect the decoherence characteristics and the last 4 dimensions reflect the photon counting fluctuation characteristics. For example, the environmental entropy vector obtained from a certain measurement may be [0.32, 0.18, 0.25, 0.25, 0.15, 0.30, 0.27, 0.28].

[0061] Perform a tensor product operation on the probability distribution of the measurement result and the environmental entropy to obtain a quantum environment composite characteristic matrix. In this embodiment, a four-dimensional Bell state probability distribution vector and an eight-dimensional environmental entropy vector perform a tensor product operation to generate a 4×8-dimensional composite characteristic matrix.

[0062] Multiply each element in the Bell state probability distribution vector by each element in the environmental entropy vector to form the corresponding element in the matrix. For example, if the Bell state probability distribution is [0.23, 0.27, 0.22, 0.28] and the environmental entropy vector is [0.32, 0.18, 0.25, 0.25, 0.15, 0.30, 0.27, 0.28], then the first row elements of the composite characteristic matrix are [0.23×0.32, 0.23×0.18, 0.23×0.25,...], resulting in [0.0736, 0.0414, 0.0575, 0.0575, 0.0345, 0.0690, 0.0621, 0.0644]. Build a complete 4×8 matrix in this way.

[0063] Generate a random amplitude and a random phase based on a quantum random number generator. The quantum random number generator uses the quantum random behavior of single photons on a 50:50 beam splitter to generate true random numbers. In a specific implementation, after the photons emitted by a single photon source (such as a attenuated laser or a parametric down-conversion source) pass through the beam splitter, a binary random number stream is generated according to the detection results of the photons at the two output ports.

[0064] The random amplitude has an 8-bit precision and its value range is between 0 and 255; the random phase has an 8-bit precision and represents the phase angle in the range of 0 to 2π. For example, a randomly generated amplitude may be 137 and the random phase may be 93 (corresponding to an angle of approximately 2.28 radians).

[0065] A random perturbation term is generated by combining a random amplitude and a random phase. In actual implementation, the random perturbation term is constructed as a matrix with the same dimension as the composite feature matrix, and each element is calculated by combining the random amplitude and the random phase. Specifically, the perturbation term at position (i,j) in the matrix can be expressed as multiplying the random amplitude by the value of the corresponding element in the composite feature matrix at that position, and then determining the positive or negative sign according to the random phase.

[0066] Perform a quantum-secure hashing operation on the composite feature matrix of the quantum environment. This step uses the SHA-3 algorithm to convert the 4×8-dimensional composite feature matrix into a 256-bit hash value. The improvement lies in converting the matrix elements into fixed-precision integers (such as rounding after multiplying by 10^6) before input, and adding a permutation operation based on quantum characteristics inside the hashing function.

[0067] Perform an exclusive OR operation on the result of the hashing operation and the random perturbation term. The exclusive OR operation is performed bit by bit, that is, each bit of the hash value is exclusive ORed with the corresponding bit of the random perturbation term. For example, if a certain bit of the hash value is 1 and the corresponding bit of the perturbation term is 0, the result is 1; if both are 1, the result is 0.

[0068] The generated result is used as the first-layer quantum key. In a multi-level structure, the output of each layer will be used as one of the inputs for the next layer. In specific implementation, the first-layer key is combined with the results of a new round of quantum measurements to generate the second-layer key; the second-layer key is then combined with the results of a new round of quantum measurements to generate the third-layer key, and so on.

[0069] The system ensures that the quantum key of each layer meets the preset upper bound of the collision probability by adjusting parameters. In practical applications, the typical upper bound of the collision probability is set to 2^(-128), which means that the probability of finding two different inputs generating the same key does not exceed 2^(-128). By increasing the entropy source of quantum randomness, optimizing the security parameters of the hashing function, and adjusting the dependencies between layers, the system can be guaranteed to reach the expected security level.

[0070] The generated three-layer keys are binary strings with lengths of 256 bits, 384 bits, and 512 bits respectively. The key of each layer can be directly used for encryption applications with different security levels, or used as key material for other cryptographic protocols.

[0071] In an alternative implementation, a data verification mechanism based on homomorphic encryption is introduced in each layer of the encryption process to ensure the integrity of data during the virtual-real interaction process, including: Generate corresponding homomorphic eigenvalues for the quantum key in each layer of the encryption process, where the homomorphic eigenvalues are calculated by inputting the quantum key, timestamp information, and random perturbation amount into a preset homomorphic encryption function; Bind the homomorphic eigenvalues of each layer to the physical world data to form data-eigenvalue pairs, and encrypt the data-eigenvalue pairs to obtain encrypted data packets; Perform a homomorphic decryption operation on each layer of encrypted data packets, and re-enter the decrypted data, timestamp information, and random perturbation amount into the homomorphic encryption function to obtain verification eigenvalues.

[0072] To generate corresponding homomorphic eigenvalues for the quantum key in each layer of the encryption process, the quantum key, timestamp information, and random perturbation amount are input into a preset homomorphic encryption function for calculation. The homomorphic encryption function can adopt a partially homomorphic or fully homomorphic encryption scheme, such as the Paillier encryption algorithm or the BFV homomorphic encryption scheme. Specifically, when using the Paillier encryption algorithm, by generating two large prime numbers p and q, calculating n = p×q and φ(n)=(p - 1)×(q - 1), selecting a random integer g that satisfies gcd(g,n²)=1, and an integer μ that satisfies λ×μ≡1 mod n, where λ = lcm(p - 1,q - 1), a homomorphic encryption environment is constructed. Suppose the quantum key for a certain layer is a 256-bit binary string "101...010", which can be grouped by 8 bits and converted into an array of decimal values [173, 42, 198,...], the timestamp information is the current UNIX timestamp 1649325789, and the random perturbation amount is the value 1762354 generated by a physical random number generator. By combining these values into the plaintext m and applying the Paillier encryption calculation c=(g^m × r^n) mod n², where r is a selected random number, the homomorphic eigenvalue is obtained. For example, after homomorphic encryption of the above input values, a 1024-bit eigenvalue "4a7b...9c3d" can be obtained.

[0073] Binding the homomorphic eigenvalues of each layer to the physical world data to form data-eigenvalue pairs is the next step. Physical world data may include sensor-collected information such as temperature, humidity, position coordinates, etc. For example, the data collected by an intelligent IoT device is temperature 23.5°C, humidity 65%, and air pressure 1013.2 hPa, which is serialized into a byte stream. Then this byte stream is combined with the previously generated homomorphic eigenvalue "4a7b...9c3d" into a data structure to form a data-eigenvalue pair. The data-eigenvalue pair can adopt the following structure: data area | eigenvalue area | timestamp area | data length indication area, where the data area stores the original physical world data, the eigenvalue area stores the homomorphic eigenvalue, the timestamp area stores the encryption time, and the data length indication area records the data lengths of each part for easy parsing. After forming the data-eigenvalue pair, use the quantum key of the current layer to encrypt the entire data structure, and a symmetric encryption algorithm with authentication function such as AES-GCM can be used to obtain an encrypted data packet. Taking AES-256-GCM as an example, use a 256-bit quantum key as the encryption key to encrypt the data-eigenvalue pair to generate ciphertext and an authentication tag. For example, the encrypted data packet can be represented as 1024-byte ciphertext "e7d2...8a1f" and 16-byte authentication tag "b3c4...5d6e".

[0074] Performing homomorphic decryption operations on the encrypted data packets of each layer and verifying data integrity is a crucial step. When the receiver decrypts the data of each layer, use the quantum key of the corresponding layer to decrypt the encrypted data packet to obtain the original data-eigenvalue pair. Separate elements such as physical world data, homomorphic eigenvalues, and timestamp information from the data-eigenvalue pair. Then extract the physical world data and re-enter it, along with the timestamp information at the time of decryption and the same random perturbation amount, into the homomorphic encryption function to calculate the verification eigenvalue. For example, the decrypted physical world data is temperature 23.5°C, humidity 65%, and air pressure 1013.2 hPa, the timestamp at the time of decryption is 1649325815, and using the same random perturbation amount 1762354, the verification eigenvalue "4a7b...9c3d" is recalculated through the homomorphic encryption function. Compare this verification eigenvalue with the original homomorphic eigenvalue extracted from the data-eigenvalue pair. If the two match exactly, it proves that the data has not been tampered with during transmission and the integrity is guaranteed; if they do not match, it indicates that the data may have been tampered with, and the system will trigger a security alert and reject further processing of the data.

[0075] Based on the properties of homomorphic encryption, specific operations can be performed even in the encrypted state, and the recipient can verify the authenticity and integrity of the data without fully decrypting it. Specifically, using the additive homomorphic property of Paillier homomorphic encryption, the recipient can calculate E(a + b) = E(a) × E(b) mod n², or E(a × k) = E(a)^k mod n² in the encrypted domain. This enables the recipient to verify specific attributes of the data or perform limited calculations in the encrypted state. For example, for multi-layer encrypted IoT sensor data, it is possible to verify whether the sensor readings are within the expected range or calculate the average of multiple sensor readings without decrypting all the content.

[0076] In an actual application scenario, consider various sensor data collected by a smart home system. The system includes temperature sensors, humidity sensors, smoke detectors, and door and window status sensors, and these data have different security levels. The security levels of temperature and humidity data are 3, using 128-bit quantum keys and single-layer homomorphic encryption; the security level of smoke detector data is 7, using 256-bit quantum keys and double-layer homomorphic encryption; the door and window status sensors are related to home security, with a security level of 9, using 512-bit quantum keys and triple-layer homomorphic encryption. Taking the smoke detector data as an example, this data first generates a homomorphic eigenvalue "8c7d...2e3f" using the 256-bit quantum key "f3a5...e2d1", binds it with the original smoke concentration data "12ppm" to form a data-eigenvalue pair, and then uses AES-256-GCM encryption to obtain the first-layer encrypted data packet. In the second-layer encryption, a new homomorphic eigenvalue is generated again, bound with the first-layer encrypted data packet and encrypted. During the decryption process, the second layer is first decrypted, and after verifying the eigenvalue match, the first layer is decrypted to finally obtain the original smoke concentration data. Through testing, it is found that this method can ensure data integrity, and for data with a security level of 7, the total encryption and decryption time does not exceed 15 milliseconds, meeting the real-time requirements of the IoT.

[0077] Compared with traditional technologies, this implementation method has significant advantages. Traditional IoT data protection mainly relies on transport layer security protocols such as TLS / SSL, or uses digital signature technology to verify data integrity. These methods have limited effectiveness in ensuring data transmission security and cannot support direct operations on encrypted data in a virtual environment. Traditional methods usually use a single checksum or hash value to verify data integrity, are vulnerable to man-in-the-middle attacks, and lack flexibility when the data needs to be processed by multiple parties. This implementation method solves the above problems by introducing homomorphic encryption technology.

[0078] The starting point of the technical improvement in this embodiment lies in meeting the special requirements for data security in the scenario of the integration of intelligent Internet of Things and the metaverse. In the virtual-real fusion environment, data needs to allow specific calculation and verification operations while remaining in an encrypted state. Through homomorphic encryption technology, even in the encrypted state of data, limited mathematical operations and verification processes can be performed, avoiding the security risks brought by frequent decryption in traditional encryption schemes. The experimental results show that after adopting this solution, when processing data with a security level of 7, in terms of data integrity verification, the accuracy rate of detecting tampering reaches 99.7%, while the traditional checksum method is only 87.3%; in terms of performance, the average time for this solution to perform integrity verification is 8.5 milliseconds, which is better than the 23.7 milliseconds of the traditional digital signature method; in terms of functional flexibility, this solution supports completing 87% of the data processing requirements without complete decryption, while the traditional method hardly has this ability. These improvements make this solution have outstanding advantages in ensuring the security of virtual-real interaction data in intelligent Internet of Things and the metaverse.

[0079] In an alternative embodiment, based on a pre-constructed decentralized trust system, multi-party consensus verification is performed through a smart contract combined with the Byzantine fault tolerance algorithm. The verification of data authenticity includes: Construct a distributed trust network based on a directed acyclic graph, collect the direct interaction information between nodes in the distributed trust network to obtain the direct trust degree, and obtain the indirect trust degree through the trust transfer of preset recommended nodes. The direct trust degree and the indirect trust degree are weighted and fused to obtain the comprehensive trust degree between nodes; Divide the nodes in the distributed trust network into honest nodes, Byzantine nodes, and uncertain nodes according to the comprehensive trust degree between nodes, construct a three-state state transition matrix, calculate the transition probability of nodes between different states according to the three-state state transition matrix, and update the node state according to the transition probability; Calculate the consensus metric value in combination with the transition probabilities corresponding to the Byzantine nodes and the honest nodes. The consensus metric value is jointly determined by the number of honest nodes and the number of Byzantine nodes that reach a consensus. When the difference between the consensus metric values at adjacent times is less than a preset convergence threshold and the consensus metric value is greater than a preset consensus threshold, it is confirmed that the verification of data authenticity is completed.

[0080] In the link of constructing a distributed trust network, a directed acyclic graph is adopted as the basic structure. Each node in the network represents an entity participating in verification, such as an intelligent IoT device, an edge computing node, or a cloud service node. The directed edges between nodes represent trust relationships, and the weights of the edges represent the degree of trust. For example, in a network composed of 10 nodes, node A has interacted directly with node B 5 times, and 4 of the interactions have good results. Then the initial direct trust degree of A in B can be set to 0.8. The calculation of the direct trust degree comprehensively considers the interaction frequency, interaction results, and time decay factor. The time decay factor reflects the weakening of the influence of earlier interactions on the trust degree, and an exponential decay form can be set. For example, the interaction weight within the most recent day is 1.0, the interaction weight one week ago drops to 0.7, and the interaction weight one month ago drops to 0.4.

[0081] To calculate the indirect trust degree, trust transfer needs to be carried out through recommended nodes. If node A trusts node B and node B trusts node C, then A can establish an indirect trust relationship with C through B. The indirect trust degree is calculated by the continuous decay of the direct trust degrees on the link. In the actual scenario, there may be multiple trust transfer paths from A to C. At this time, the maximum value among all paths is taken as the final indirect trust degree. For example, there are two paths from A to C: A→B→C and A→D→E→C. If the indirect trust degree calculated by the former path is 0.65 and the latter is calculated to be 0.58, then the indirect trust degree of A in C is taken as 0.65.

[0082] The comprehensive trust degree between nodes is obtained through the weighted fusion of the direct trust degree and the indirect trust degree. The weighting factors can be dynamically adjusted according to the network characteristics. Generally, the direct trust weight is greater than the indirect trust weight. Experiments show that setting the direct trust weight to 0.7 and the indirect trust weight to 0.3 has a good effect in a stable network. For example, the direct trust degree of node A in B is 0.8, and the indirect trust degree is 0.65. Then the comprehensive trust degree is 0.8×0.7 + 0.65×0.3 = 0.755. The comprehensive trust degrees between all pairs of nodes in the network are calculated to form a trust degree matrix.

[0083] Based on the calculated comprehensive trust degree between nodes, the nodes in the distributed trust network are divided into three categories: honest nodes, Byzantine nodes, and uncertain nodes. The classification criteria can be set as follows: nodes with a comprehensive trust degree greater than 0.75 are regarded as honest nodes, those lower than 0.3 are regarded as Byzantine nodes, and those in between are uncertain nodes. In practical applications, these thresholds can be adjusted according to the application scenario. For example, in a high-security requirement scenario, the trust degree threshold for honest nodes can be increased to 0.85.

[0084] After the node partitioning is completed, a three-state state transition matrix is constructed to describe the transition probabilities between the three states of honest, Byzantine, and uncertain for the nodes. The state transition matrix is a 3×3 matrix, where the rows represent the current state, the columns represent the next state, and the matrix element values represent the transition probabilities. The transition probabilities are calculated based on historical observation data. For example, in a certain network, the probability that an honest node remains in the honest state in the next round is 0.92, the probability of becoming an uncertain state is 0.07, and the probability of becoming a Byzantine state is 0.01. Similarly, the state transition probabilities of Byzantine nodes and uncertain nodes can be obtained to complete the construction of the entire transition matrix.

[0085] After the state transition matrix is constructed, in each round of iteration, the states of each node are updated according to the transition probabilities. The update method is as follows: Generate a random number in the interval [0,1] and compare it with the cumulative transition probability to determine the new state. For example, if a node is currently in an uncertain state and its transition probabilities are [0.15, 0.65, 0.2] (representing the probabilities of becoming honest, remaining uncertain, and becoming Byzantine respectively), if the random number is 0.3, then this value falls within the second interval [0.15, 0.8), and the node remains in the uncertain state; if the random number is 0.1, it becomes an honest node; if the random number is 0.9, it becomes a Byzantine node.

[0086] Combining the transition probabilities corresponding to Byzantine nodes and honest nodes, the consensus metric value is calculated. The consensus metric value reflects the quality of the network reaching a consensus and is jointly determined by the number of honest nodes and Byzantine nodes that reach a consensus. A simple calculation method is: the value obtained by subtracting the number of Byzantine nodes from the number of honest nodes and then dividing by the total number of nodes. For example, in a network with 100 nodes, if there are currently 75 honest nodes, 15 Byzantine nodes, and 10 uncertain nodes, then the consensus metric value is (75 - 15) / 100 = 0.6. More complex calculations can consider the weighted trust of various types of nodes.

[0087] During the iteration process, when the difference between the consensus metric values at adjacent times is less than the preset convergence threshold and the consensus metric value is greater than the preset consensus threshold, it is confirmed that the data authenticity verification is completed. The convergence threshold can be set to 0.01, indicating that the change in the consensus metric value between two consecutive rounds of calculations does not exceed 0.01; the consensus threshold can be set to 0.5, indicating that the honest nodes have an obvious advantage. Both can be adjusted according to the actual application scenario and security requirements.

[0088] A specific case of applying the above method to the virtual-real fusion scenario: An intelligent Internet of Things environmental monitoring system includes 30 environmental sensor nodes, 5 edge computing nodes, and 1 cloud service node. These nodes are connected through a decentralized network to jointly verify the authenticity of environmental data. After constructing a directed acyclic graph, the direct trust degree is calculated based on the historical interactions between nodes. For example, the direct trust degree of node A in edge node B is 0.82. The indirect trust degree is calculated through trust transfer. For example, the indirect trust degree of the cloud service node in the end sensor node is 0.63. The comprehensive trust degree matrix is obtained through weighted fusion.

[0089] According to the division criteria, 22 honest nodes, 5 Byzantine nodes, and 9 uncertain nodes are identified in the initial state. A state transition matrix is constructed. The probability that an honest node remains in the honest state is 0.90, the probability that a Byzantine node remains in the Byzantine state is 0.85, the probability that an uncertain node turns into an honest node is 0.25, and the probability that it turns into a Byzantine node is 0.15. After multiple rounds of iteration, the node states are continuously updated, and the consensus metric value gradually rises from the initial 0.47 and tends to be stable. In the 7th round of iteration, the consensus metric value reaches 0.68, the difference from 0.67 in the 6th round is 0.01, which is less than the convergence threshold and exceeds the consensus threshold of 0.5. At this time, it is confirmed that the verification of data authenticity is completed, and the environmental data can be trusted.

[0090] Figure 4 Schematic diagram of the convergence curve of the consensus metric value in the multi-party consensus verification process of the embodiment of the present invention: This figure shows the comparison of the consensus performance of different technical solutions in the process of multi-round transmission. The horizontal axis in the figure represents the transmission round (from 0 to 14 rounds), and the vertical axis represents the consensus value (from 0 to 1.0). Four different scenarios are compared in the figure: the performance of this technical solution in the case of 10% and 30% malicious nodes (represented by solid diamonds and solid circles respectively), and the performance of the traditional solution under the same proportion of malicious nodes (represented by hollow squares and hollow circles respectively).

[0091] From the data trend, this technical solution performs best in the case of 10% malicious nodes. The consensus degree rapidly rises from the initial approximately 0.37 and tends to be stable after the 8th round, finally reaching a consensus degree of approximately 0.8. In the case of 30% malicious nodes, the performance of this technical solution slightly decreases, but it is still better than the traditional solution, and finally reaches a consensus degree of approximately 0.65. The performance of the traditional solution is second in the case of 10% malicious nodes, and finally stabilizes at a consensus degree of approximately 0.63, while in the case of 30% malicious nodes, it performs the worst, only reaching a consensus degree of approximately 0.4. These data clearly show that this technical solution has obvious advantages in dealing with malicious nodes of different degrees, especially in the case of a higher proportion of malicious nodes, its performance advantages are more significant.

[0092] In the second aspect of the embodiments of the present invention, there is provided an encryption system for ensuring the security of the virtual-real integration of intelligent IoT and the metaverse, including: A first unit, configured to obtain the device identification information of an intelligent IoT device and the virtual object identification information corresponding to the intelligent IoT device in the metaverse space; based on the device identification information and the virtual object identification information, construct a third-order tensor, and generate a virtual-real mapping relationship identifier through a non-linear mapping function; A second unit, configured to call a preset quantum key generation module according to the virtual-real mapping relationship identifier to generate a quantum encryption key for virtual-real interaction; A third unit, configured to perform hierarchical encryption processing on the physical world data collected by the intelligent IoT device through the quantum encryption key, construct a multi-level key generation network based on quantum entanglement states, and use quantum keys of different lengths for encryption according to the physical world data of different security levels, and introduce a data verification mechanism based on homomorphic encryption in each layer of the encryption process to ensure the integrity of the data during the virtual-real interaction process; A fourth unit, configured to transmit the encrypted physical world data to the virtual object in the metaverse space; A fifth unit, configured to, when the virtual object receives the encrypted physical world data, based on a pre-constructed decentralized trust system, perform multi-party consensus verification through a smart contract combined with the Byzantine fault tolerance algorithm to verify the authenticity of the data, and decrypt the encrypted physical world data and present it in the metaverse space after the verification passes.

[0093] In the third aspect of the embodiments of the present invention, there is provided an electronic device, including: A processor; A memory for storing instructions executable by the processor; wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0094] In the fourth aspect of the embodiments of the present invention, there is provided a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is implemented.

[0095] The present invention may be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium, on which computer-readable program instructions for executing various aspects of the present invention are uploaded.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An encryption method to ensure the security of the virtual-real integration of smart IoT and Metaverse, characterized in that: include: Obtain device identification information of a smart IoT device and virtual object identification information corresponding to the smart IoT device in the metaverse space; Based on the device identification information and the virtual object identification information, a third-order tensor is constructed, and a virtual-real mapping relationship identifier is generated through a nonlinear mapping function; According to the virtual-real mapping relationship identifier, a preset quantum key generation module is called to generate a quantum encryption key for virtual-real interaction; The physical world data collected by the smart IoT device is encrypted in layers using the quantum encryption key, a multi-level key generation network is constructed based on quantum entanglement, and quantum keys of different lengths are used to encrypt physical world data of different security levels. A data verification mechanism based on homomorphic encryption is introduced in each encryption layer to ensure the integrity of the data during the virtual-real interaction process. Transmitting the encrypted physical world data to the virtual object in the metaverse space; When the virtual object receives the encrypted physical world data, based on the pre-built decentralized trust system, multi-party consensus verification is performed through smart contracts combined with the Byzantine fault-tolerant algorithm to verify the authenticity of the data. After the verification is passed, the encrypted physical world data is decrypted and presented in the metaverse space.

2. The method according to claim 1, characterized in that Based on the device identification information and the virtual object identification information, constructing a third-order tensor, and generating a virtual-real mapping relationship identifier through a nonlinear mapping function includes: The device identification information includes physical feature dimension information and behavioral feature dimension information, wherein the physical feature dimension information includes hardware fingerprint information, geographic location information and network access feature information of the smart IoT device, and the behavioral feature dimension information includes data transmission mode information, resource occupation feature information and periodic activity feature information of the smart IoT device; The identification information of the virtual object includes basic entity attributes, functional attributes and interactive relationship information; Based on the device identification information and the identification information of the virtual object, a third-order tensor is constructed, the physical feature dimension of the device identification information is mapped to the first dimension of the third-order tensor, the behavioral feature dimension of the device identification information is mapped to the second dimension of the third-order tensor, and the identification information of the virtual object is mapped to the third dimension of the third-order tensor, and a feature fusion result of the physical device and the virtual object is obtained by performing tensor decomposition on the third-order tensor, wherein the tensor decomposition includes decomposing the third-order tensor into the sum of tensor products of basis vectors of three feature spaces; The feature fusion result is input into a preset nonlinear mapping function to obtain a virtual-real mapping relationship identifier.

3. The method according to claim 1, characterized in that The physical world data collected by the smart IoT device is encrypted hierarchically by using the quantum encryption key, a multi-level key generation network is constructed based on the quantum entanglement state, and quantum keys of different lengths are used to encrypt the physical world data with different security levels, including: Based on a preset data security assessment model, the security level value of the physical world data is calculated according to the sensitivity, integrity requirements and real-time requirements of the physical world data; Based on the security level value, determining the quantum key length through an adaptive adjustment mechanism, wherein the quantum key length is related to the security level value, the basic key length, the security factor, and the expected cryptographic security time, and the higher the security level value, the corresponding increase of the quantum key length; A multi-level key generation network is constructed based on quantum entangled states. The quantum entangled state measurement results are tensor-producted with the environmental entropy, and random perturbation terms are superimposed to generate quantum keys for each layer. Determine the number of encryption layers according to the security level value, and distribute the quantum key according to the number of encryption layers; The physical world data is subjected to multi-layer cascade encryption processing. In each layer of encryption process, the current layer data, the current layer quantum key and the timestamp are combined and calculated to obtain a verification value, and the verification value is used to verify the integrity of the encrypted data of this layer.

4. The method according to claim 3, characterized in that: A multi-level key generation network is constructed based on quantum entangled states. The quantum entangled state measurement results are tensor-producted with the environmental entropy, and random perturbations are superimposed to generate quantum keys for each layer, including: Generating a quantum entangled photon pair through a parametric down-conversion process, generating a superposition state of a horizontal polarization state and a vertical polarization state as an initial quantum entangled state; performing Bell state measurement on the quantum entangled photon pair to obtain a probability distribution of the measurement result; Collecting quantum decoherence noise information and photon counting fluctuation information from a quantum system environment, and calculating environmental entropy based on the quantum decoherence noise information and the photon counting fluctuation information; Performing a tensor product operation on the probability distribution of the measurement result and the environmental entropy to obtain a quantum environmental composite characteristic matrix; Generate a random amplitude and a random phase based on a quantum random number generator, and combine the random amplitude and the random phase to generate a random disturbance term; A quantum secure hash operation is performed on the quantum environment composite characteristic matrix, and an XOR operation is performed on the operation result and the random perturbation term to generate a multi-level quantum key, wherein the quantum key of each layer satisfies a preset collision probability upper bound.

5. The method according to claim 1, characterized in that In each layer of encryption, a data verification mechanism based on homomorphic encryption is introduced to ensure the integrity of data during the virtual-real interaction process, including: Generate a corresponding homomorphic eigenvalue for the quantum key in each layer of encryption process, wherein the homomorphic eigenvalue is calculated by inputting the quantum key, timestamp information, and random perturbation amount into a preset homomorphic encryption function; Binding the homomorphic eigenvalues ​​of each layer to the physical world data to form data-eigenvalue pairs, and encrypting the data-eigenvalue pairs to obtain encrypted data packets; Perform a homomorphic decryption operation on each layer of encrypted data packets, re-input the decrypted data, timestamp information, and random perturbation amount into the homomorphic encryption function to obtain a verification characteristic value.

6. The method according to claim 1, characterized in that Based on the pre-built decentralized trust system, multi-party consensus verification is performed through smart contracts combined with the Byzantine fault-tolerant algorithm to verify the authenticity of data, including: A distributed trust network is constructed based on a directed acyclic graph, direct interaction information between nodes in the distributed trust network is collected to obtain direct trust, and indirect trust is obtained through trust transfer of preset recommendation nodes, and the direct trust and the indirect trust are weightedly integrated to obtain comprehensive trust between nodes; Dividing the nodes in the distributed trust network into honest nodes, Byzantine nodes and uncertain nodes according to the comprehensive trust between the nodes, constructing a three-state state transfer matrix, calculating the transition probability of the nodes between different states according to the three-state state transfer matrix, and updating the node state according to the transition probability; The consensus metric value is calculated in combination with the transition probabilities corresponding to the Byzantine nodes and the honest nodes. The consensus metric value is jointly determined by the number of honest nodes and the number of Byzantine nodes that have reached consensus. When the difference between the consensus metric values ​​at adjacent moments is less than the preset convergence threshold and the consensus metric value is greater than the preset consensus threshold, the data authenticity verification is confirmed to be completed.

7. An encryption system for ensuring the security of the virtual-real integration of smart IoT and metaverse, used to implement the method as described in any one of claims 1 to 6, characterized in that: include: The first unit is used to obtain device identification information of a smart IoT device and virtual object identification information corresponding to the smart IoT device in a metaverse space; Based on the device identification information and the virtual object identification information, a third-order tensor is constructed, and a virtual-real mapping relationship identifier is generated through a nonlinear mapping function; The second unit is used to call a preset quantum key generation module according to the virtual-real mapping relationship identifier to generate a quantum encryption key for virtual-real interaction; The third unit is used to perform hierarchical encryption processing on the physical world data collected by the smart IoT device through the quantum encryption key, build a multi-level key generation network based on quantum entanglement state, and use quantum keys of different lengths to encrypt physical world data with different security levels, and introduce a data verification mechanism based on homomorphic encryption in each encryption layer to ensure the integrity of the data in the virtual-real interaction process; A fourth unit is used to transmit the encrypted physical world data to the virtual object in the metaverse space; The fifth unit is used to perform multi-party consensus verification based on a pre-built decentralized trust system when the virtual object receives the encrypted physical world data, through a smart contract combined with a Byzantine fault-tolerant algorithm, to verify the authenticity of the data. After the verification is passed, the encrypted physical world data will be decrypted and presented in the metaverse space.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images