Methods for Data Security and Privacy Protection in Jewelry Transactions Based on 5G Convergence Applications

Abstract

This invention discloses a method for protecting the security and privacy of jewelry transaction data based on 5G converged applications. The method includes: establishing a distributed database; generating a first public key and a first private key for both parties using a pre-established first asymmetric encryption algorithm; calculating a random multivariate function on the first public and private keys, and then de-identifying and storing the random multivariate function in the distributed database using an encryption algorithm; obtaining a random seed for the random multivariate function using a key derivation function of a second hash algorithm, which serves as the second private key for both parties; during the encryption phase, the transaction data of both parties is encrypted using an encryption algorithm, and the encrypted data is de-identified and stored; during the decryption phase, both parties obtain the plaintext data using the second private key. This invention, based on existing encryption algorithms, introduces a neural network algorithm to generate the buyer's key, while simultaneously de-identifying the encrypted data, making jewelry transaction data more secure.

Description

Technical Field

[0001] This invention relates to the field of cryptography, and more specifically, to a method for protecting the security and privacy of jewelry transaction data based on 5G converged applications. Background Technology

[0002] With the rapid development of 5G technology, the jewelry trading industry has ushered in new opportunities and challenges. 5G networks support higher data transmission rates and lower latency, leading to the transmission of large amounts of data and increasing the risk of data breaches. Furthermore, the interconnectedness of everything in the 5G era, with numerous terminal devices connecting to the network and more diverse data sources, further complicates privacy protection.

[0003] Data security and privacy protection are paramount in jewelry transactions. These transactions involve a wealth of sensitive information, such as customer personal details, transaction records, and jewelry appraisal certificates. The leakage of this information can harm not only customers' interests but also the reputation and business of jewelry companies. Therefore, the jewelry trading industry needs to implement stricter data security and privacy protection measures to ensure the security and privacy of data during transmission, storage, and use.

[0004] When encrypting data in jewelry transactions, the security encryption for the buyer and the service provider requires different processing. For example, in jewelry transactions, the buyer usually conducts the transaction offline, which makes it very easy for transaction data to be leaked. Therefore, when encrypting data, the existing technology does not adopt differentiated encryption processing methods, which can easily lead to data leakage.

[0005] Most existing jewelry transaction data uses asymmetric encryption algorithms. Encrypting jewelry transaction data with asymmetric encryption algorithms is relatively secure. However, with the continuous advancement of science and technology, especially the emergence of quantum technology and side-channel attacks, asymmetric encryption algorithms face great threats, and their security is constantly decreasing. Summary of the Invention

[0006] The purpose of this invention is to provide a method for protecting the security and privacy of jewelry transaction data based on 5G converged applications, in order to solve the above-mentioned problems existing in the prior art.

[0007] The application is as follows:

[0008] A method for protecting the security and privacy of jewelry transaction data based on 5G converged applications includes the following steps:

[0009] S1. Establish a distributed database, which is used to store important encrypted data using an anonymized and encrypted method;

[0010] S2. Generate a key pair for the buyer and a key pair for the service provider using a pre-established first asymmetric encryption neural network model. The key pair includes a public key and a private key. The public key is regarded as the first transaction public key of the buyer or the service provider, and the private key is regarded as the first transaction private key of the buyer or the service provider.

[0011] S3. Calculate a random multivariate function on the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function includes the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function is then de-identified and saved to the distributed database using the AES encryption algorithm.

[0012] S4. Obtain the random seed of the random multivariate function through the key derivation function KDF of the second hash algorithm, and use the obtained random seed as the second transaction private key of the buyer or service provider.

[0013] S5. During the encryption period, the buyer queries and obtains the service provider's first transaction public key. Using the obtained service provider's first transaction public key, the buyer encrypts the plaintext data to be encrypted using the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm. The service provider queries and obtains the buyer's first transaction public key. Using the obtained buyer's first transaction public key, the service provider encrypts the plaintext data to be encrypted using the first asymmetric encryption algorithm to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm.

[0014] S6. During the decryption period, the buyer obtains plaintext data by entering the buyer's second transaction private key, and the service provider obtains plaintext data by entering the service provider's second transaction private key.

[0015] Furthermore, the first asymmetric encryption neural network model in step S2 includes:

[0016] Input layer, first convolutional layer, second convolutional layer, activation layer, third convolutional layer, pooling layer, and output layer;

[0017] The following steps are performed using the first asymmetric encryption neural network model to generate the buyer's key pair:

[0018] The buyer inputs the jewelry information they purchased through the input layer.

[0019] The first key feature is obtained by extracting features from the jewelry information purchased by the buyer through the first convolutional layer.

[0020] The second key features are obtained by extracting features from the first key features through the second convolutional layer;

[0021] The second key feature is nonlinearly transformed by the activation layer, and the third key feature is obtained by feature extraction from the nonlinear transformation result by the third convolutional layer.

[0022] The third key feature is pooled through the pooling layer, and the pooling result is output through the output layer. The output of the output layer is used as input to the second asymmetric encryption algorithm to generate the first transaction public key and the first transaction private key.

[0023] Furthermore, feature extraction is performed on the jewelry information purchased by the buyer. The jewelry information includes digital images and product information, wherein the product information includes jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a six-sided image of the jewelry.

[0024] A first jewelry image is generated based on the jewelry information, wherein the first jewelry image is used to encrypt and store the product information in the form of an image, and is stored in the distributed database;

[0025] A jewelry product image is generated based on the first jewelry image and the digital image, wherein the digital image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, and image features of the people involved in the transaction;

[0026] The jewelry product image is input into a pre-trained Xception model to extract key features, resulting in a key feature vector of jewelry information; wherein, the Xception model is a variant of the Inception series, which introduces depthwise separable convolution.

[0027] Further, the calculation of the first transaction public key and the first transaction private key of the buyer and the service provider in step S3 includes: the multivariate function is a random multivariate function generated by the system: R = f(x, y), and the first transaction public key and the first transaction private key of the buyer and the service provider are substituted into the multivariate function for calculation; wherein, in the multivariate function R = f(x, y), x is regarded as the first transaction public key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, y is regarded as the first transaction private key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, and R is the result of the multivariate function.

[0028] Further, step S4 includes: performing a key derivation function KDF calculation on the multivariate function R = f(x, y) using the key derivation function KDF of the second hash algorithm, converting the multivariate function into a random seed, and using the obtained random seed as the second transaction private key of the buyer or service provider; the key derivation function KDF calculation on the multivariate function R = f(x, y) refers to performing the key derivation function KDF calculation on the entire function expression of the multivariate function, rather than only performing the key derivation function KDF calculation on the result R of the multivariate function.

[0029] Furthermore, step S4 also includes: calculating a multivariate function R = f(x, y) through the key derivation function KDF, storing the result in a distributed database, and simultaneously generating a second transaction private key key field. The second transaction private key key field is desensitized and protected for privacy using the AES encryption algorithm, and the key field represents the address of each column of the data table in the database.

[0030] Further, step S5 includes: the buyer refers to a jewelry purchaser, and the service provider refers to a jewelry purchase service provider; during the encryption period, the buyer obtains the service provider's first transaction public key through query, and uses the obtained service provider's first transaction public key to encrypt the plaintext data to be sent through the first asymmetric encryption neural network model encryption process, and the encrypted ciphertext data generates a ciphertext key field, and saves the generated ciphertext key field to the distributed database, and uses the AES encryption algorithm to de-identify and protect the privacy of the ciphertext key field.

[0031] Further, step S6 includes: during the decryption period, the service provider inputs the second transaction private key, and the system performs verification and decryption; the system performs verification and decryption by first locking the second transaction private key key field of the buyer, and verifying the multivariate function R = f(x, y) of the second transaction private key key field, verifying whether the random seed transformed by the multivariate function R = f(x, y) is the same as the buyer's second transaction private key, and if they are the same, it proves that the second transaction private key input by the buyer is correct.

[0032] Furthermore, step S6 also includes: verifying the correctness of the service provider's second transaction private key through the multivariate function R = f(x, y) stored in the second transaction private key key field, where x is the first transaction public key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, and y is the first transaction private key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm; since the service provider's first transaction public key x has been made public, the service provider's first transaction private key y is obtained through the multivariate function R = f(x, y).

[0033] Furthermore, step S6 also includes: determining the ciphertext key field generated after being encrypted by the first transaction public key of the service provider using the obtained first transaction private key y of the service provider, decrypting the ciphertext data through the decryption process of the first asymmetric encryption algorithm, finally obtaining plaintext data, and outputting the obtained plaintext data to the service provider; the plaintext data obtained by the buyer is the same as the aforementioned process.

[0034] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0035] This invention provides a method for establishing a distributed database. A key pair for the buyer is generated using a pre-established first asymmetric encryption neural network model, and a key pair for the service provider is generated using a first asymmetric encryption algorithm. Each key pair includes a public key and a private key. The public key is considered the first transaction public key of the buyer or service provider, and the private key is considered the first transaction private key of the buyer or service provider. A random multivariate function is calculated on the first transaction public key and first transaction private key of the buyer and service provider. The random multivariate function includes the first transaction public key and first transaction private key of both the buyer and service provider. The random multivariate function is then de-identified and stored in the distributed database using the AES encryption algorithm. A random seed for the random multivariate function is obtained using the key derivation function (KDF) of a second hash algorithm, and the obtained random seed is used as the second transaction private key of the buyer or service provider. During the encryption period, the buyer queries and obtains the first transaction private key of the service provider. A transaction public key is used to encrypt plaintext data to be encrypted using a first asymmetric encryption neural network model, obtaining ciphertext data. The ciphertext data is then de-identified and stored in the distributed database using the AES encryption algorithm. The service provider queries and obtains the buyer's first transaction public key, and uses this key to encrypt the plaintext data to be encrypted using the first asymmetric encryption algorithm, obtaining ciphertext data. The ciphertext data is then de-identified and stored in the distributed database using the AES encryption algorithm. During decryption, the buyer obtains the plaintext data by inputting their second transaction private key, and the service provider obtains the plaintext data by inputting their second transaction private key. This invention, based on existing encryption algorithms, introduces a neural network algorithm to generate the buyer's key and simultaneously de-identifies the ciphertext data, making jewelry transaction data more secure. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the method for protecting jewelry transaction data security and privacy based on 5G converged applications provided in this embodiment of the invention. Detailed Implementation

[0037] The present invention will now be described in detail with reference to the accompanying drawings.

[0038] Example 1

[0039] This invention provides a method for protecting the security and privacy of jewelry transaction data based on 5G converged applications, such as... Figure 1 This includes the following steps:

[0040] A method for protecting the security and privacy of jewelry transaction data based on 5G converged applications, characterized by the following steps:

[0041] S1. Establish a distributed database, which is used to store important encrypted data using an anonymized and encrypted method;

[0042] It should be noted that the distributed database includes all jewelry transaction data collected before the system runs. The data in the distributed database will be continuously updated over time, but the original data content needs to be redundantly backed up to ensure that the original data can be queried at any time or to ensure that the original data is not lost. In this embodiment, the distributed database is created using Redis, a high-performance key-value store database that supports various data structures such as strings, hashes, and lists.

[0043] S2. Generate a key pair for the buyer and a key pair for the service provider using a pre-established first asymmetric encryption neural network model. The key pair includes a public key and a private key. The public key is regarded as the first transaction public key of the buyer or the service provider, and the private key is regarded as the first transaction private key of the buyer or the service provider.

[0044] Specifically, the first asymmetric encryption algorithm is any first asymmetric encryption algorithm that meets the requirements of the problem. Meeting the requirements means that the first asymmetric encryption algorithm forms a key pair, which includes a public key and a private key. Usually, the public key is used for encryption and the private key is used for decryption; or the private key is used for signing and the public key is used for verifying the signature. In this invention, the public key generated by both parties to the transaction is used as the first transaction public key, and the private key generated by both parties to the transaction is used as the first transaction private key. The first transaction public key is publicly available to the entire network.

[0045] In this embodiment, for ease of understanding, the asymmetric RSA encryption algorithm and the asymmetric ECC encryption algorithm (elliptic curve algorithm) are selected first to specifically illustrate the embodiments of the present invention.

[0046] The process of generating public and private keys using the asymmetric RSA encryption algorithm:

[0047] 1. Randomly generate two large prime numbers p and q using RSAV version 2.2;

[0048] 2. Calculate n = p * q, Euler's totient function DD(n) = (p-1) × (q-1), and eliminate p and q;

[0049] 3. Randomly select an integer e that is coprime to DD(n), i.e., (e,DD(n))=1;

[0050] 4. Find a number d such that e and d are combined, satisfying ed;

[0051] 5. Generate a public key (n, e) and a private key (n, d).

[0052] The process of generating public and private keys using the ECC encryption algorithm:

[0053] a) Randomly select an elliptic curve Fp(a, b) and take a point on the elliptic curve as the base point H;

[0054] b) Randomly select a large number m as the private key and generate a public key G = mH.

[0055] The public and private keys of the service provider are generated by using the first asymmetric encryption algorithms RSA and ECC, which in turn generate the first transaction public key and first transaction private key of the service provider in this invention.

[0056] In the above embodiments, specifically, the process for generating the first transaction public key and the first transaction private key of the purchaser in this invention is as follows:

[0057] The first asymmetric encryption neural network model includes:

[0058] Input layer, first convolutional layer, second convolutional layer, activation layer, third convolutional layer, pooling layer, and output layer;

[0059] The following steps are performed using the first asymmetric encryption neural network model to generate the buyer's key pair:

[0060] The buyer inputs the jewelry information they purchased through the input layer.

[0061] The first feature is obtained by extracting features from the jewelry information purchased by the buyer through the first convolutional layer;

[0062] The second feature is obtained by extracting features from the first key features through the second convolutional layer;

[0063] The second feature is nonlinearly transformed by the activation layer, and the third feature is obtained by feature extraction from the nonlinear transformation result by the third convolutional layer.

[0064] The third feature is pooled through the pooling layer, and the pooling result is output through the output layer. The output of the output layer is used as input to the second asymmetric encryption algorithm to generate the first transaction public key and the first transaction private key.

[0065] It should be noted that the first convolutional layer has a kernel size of 2*1, a stride of 2, and an output vector size of 12*1; the second convolutional layer has a kernel size of 3*1, a stride of 2, and an output vector size of 15*1; the third convolutional layer has a kernel size of 2*1 and a stride of 1; and the pooling layer has a kernel size of 3*1 and a stride of 1. Therefore, this embodiment generates the first transaction public key and the first transaction private key of the buyer using the first asymmetric encryption neural network model for encryption, thereby obtaining an encryption key for jewelry transactions. This effectively improves the security of the encryption key, increases the difficulty of cracking it, and enhances the reliability of jewelry transactions. This is one of the technical highlights of this invention.

[0066] It should be noted that the process of using the output layer of the first asymmetric encryption neural network model to input the second asymmetric encryption algorithm, thereby generating the first transaction public key and the first transaction private key, is similar to the process of the service provider selecting the asymmetric RSA encryption algorithm and the ECC encryption algorithm (elliptic curve algorithm) for encryption. This embodiment will not be described in detail.

[0067] In the above embodiments, specifically, feature extraction is performed on the jewelry information purchased by the buyer. The jewelry information includes digital images and product information, wherein the product information includes jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a six-sided image of the jewelry.

[0068] A first jewelry image is generated based on the jewelry information, wherein the first jewelry image is used to encrypt and store the product information in the form of an image, and is stored in the distributed database;

[0069] A jewelry product image is generated based on the first jewelry image and the digital image, wherein the digital image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, image features of the transacting parties, and image features of the jewelry certification certificate;

[0070] The jewelry product image is input into a pre-trained Xception model to extract key features, resulting in a key feature vector of jewelry information; wherein, the Xception model is a variant of the Inception series, which introduces depthwise separable convolution.

[0071] It should be noted that in this embodiment, the first transaction public key and the first transaction private key of the buyer of the present invention are generated by the first asymmetric encryption neural network model and encrypted. By introducing the digital image of jewelry information and the feature extraction corresponding to the product information, the security of the encryption key can be effectively improved, the difficulty of cracking the encryption key is increased, and the reliability of jewelry transactions is improved.

[0072] S3. Calculate a random multivariate function on the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function includes the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function is then de-identified and saved to the distributed database using the AES encryption algorithm.

[0073] Preferably, taking the asymmetric RSA encryption algorithm as an example, the randomly generated multivariate function is as follows: x is regarded as the first transaction public key obtained by the asymmetric RSA encryption algorithm, i.e., x = (n, e), and y is regarded as the first transaction private key of the asymmetric RSA encryption algorithm, i.e., y = (n, d); the random multivariate function R = f(x, y) is a function including variables x = (n, e) and y = (n, d).

[0074] Preferably, the functional relationship satisfying the above can be: F = h(x)f(y), where x = (n, e), y = (n, d), and h(x) and f(y) are multivariate functions.

[0075] Preferably, taking the asymmetric ECC encryption algorithm as an example, the randomly generated multivariate function is as follows: x is regarded as the first transaction public key obtained by the asymmetric ECC encryption algorithm, i.e., x = G; y is regarded as the first transaction private key obtained by the asymmetric ECC encryption algorithm, i.e., y = k; the random multivariate function R = f(x, y) is a function including variables x and y.

[0076] Preferably, assuming the coordinates of point G are (a, b), the functional relationship that satisfies the above can be: DD = i(x)j(y), where x = G, y = k, i(x) is a quadratic function, and j(y) is a linear function, that is, the functional relationship can be expressed as: DD = i(G)j(k).

[0077] It should be noted that any function or component of a multivariate function R = f(x, y), F = h(x)f(y), and DD = i(x)j(y) is not required to be a function of a specified degree. It is only necessary that the random function contains the first transaction public key x and the first transaction private key y. For example, the random function can be expressed as: R = f(x, y, d, ...). It can be seen that the random multivariate function R = f(x, y, d, ...) includes variables other than the first transaction public key x and the first transaction private key y.

[0078] In the above embodiments, specifically, the calculation of the first transaction public key and the first transaction private key of the buyer and the service provider using a random multivariate function includes: the multivariate function is a random multivariate function generated by the system: R = f(x, y), and the first transaction public key and the first transaction private key of the buyer and the service provider are substituted into the multivariate function for calculation; wherein, in the multivariate function R = f(x, y), x is regarded as the first transaction public key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, y is regarded as the first transaction private key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, and R is the result of the multivariate function.

[0079] It should be noted that the principle behind using different encryption methods for the buyer and the service provider is as follows: In jewelry transactions, the buyer uses the first asymmetric encryption neural network model to encrypt transaction data. By capturing the buyer's jewelry transaction characteristics and introducing digital images of jewelry information and feature extraction corresponding to product information, the security of the encryption key can be effectively improved, increasing the difficulty of cracking the encryption key and enhancing the reliability of the jewelry transaction. Furthermore, the service provider does not need to capture transaction features as input to the asymmetric encryption algorithm, nor does it need to expend neural network resources for complex encryption calculations. By employing differentiated encryption methods for the buyer and the service provider, the problem of easy leakage of buyer data is solved, which is one of the technical highlights of this invention.

[0080] S4. Obtain the random seed of the random multivariate function through the key derivation function KDF of the second hash algorithm, and use the obtained random seed as the second transaction private key of the buyer or service provider.

[0081] Specifically, the method for obtaining the random seed of the random multivariate function and using the obtained random seed as the second transaction private key for the buyer or service provider is as follows: The multivariate function R = f(x, y) is calculated using the key derivation function (KDF) of the second hash algorithm, transforming the multivariate function into a random seed. The obtained random seed is then used as the second transaction private key for the buyer or service provider. This second transaction private key is sent to the buyer or service provider, and it is required that only the buyer or service provider knows it. The key derivation function (KDF) calculation for the multivariate function R = f(x, y) refers to performing the key derivation function (KDF) calculation on the entire multivariate function expression, rather than only on the result R of the multivariate function.

[0082] Preferably, a key derivation function (KDF) is used to calculate the multivariate function R = f(x, y), which is then converted into a 64-bit random seed. This random seed is used as the second transaction private key for the buyer or service provider.

[0083] After obtaining the random seed of the multivariate function, the multivariate function R = f(x, y) of the key derivation function KDF is stored in the distributed database, and a second transaction private key key field is generated. The second transaction private key key field is de-identified and protected for privacy using the AES encryption algorithm.

[0084] It should be noted that the privacy protection of this embodiment cannot use hash desensitization, as hash desensitization is irreversible and the original data cannot be recovered. The desensitization methods in this embodiment can be encryption desensitization: reversible, but requires a key to recover the original data; replacement desensitization: reversible, but requires recording the replacement rules and the original data; mask desensitization: reversible, but requires recording the masking rules and the original data; and obfuscation desensitization: reversible, but requires recording the obfuscation rules and the original data.

[0085] It should be noted that the specific process of generating a random seed using the Key Derivation Function (KDF) is as follows:

[0086] Key derivation functions (KDFs) can be used to generate random seeds because they can transform input data of arbitrary length into a fixed-length output value, which can then be used as the seed for a random number generator. The following is the specific process of generating a random seed using a KDF:

[0087] 1. Choose a hash function

[0088] Choose a suitable key derivation function (KDF), such as PBKDF2 (Password-Based Key Derivation Function 2) or HKDF (HMAC-based Key Derivation Function): an HMAC-based key derivation function, etc.

[0089] 2. Prepare input data

[0090] Prepare one or more input data, which can be any type of information, such as timestamps, user input, random strings, etc. The diversity and unpredictability of the input data will affect the quality of the generated random seed.

[0091] 3. Calculate the hash value

[0092] The input data is hashed using the selected key derivation function (KDF) to generate a fixed-length hash value.

[0093] 4. Extract random seeds

[0094] One or more random seeds are extracted from the hash value. Typically, the hash value is a sequence of bytes that can be converted into an integer or other format suitable for a random number generator. This embodiment uses the format converted to an integer.

[0095] S5. During the encryption period, the buyer queries and obtains the service provider's first transaction public key. Using the obtained service provider's first transaction public key, the buyer encrypts the plaintext data to be encrypted using the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm. The service provider queries and obtains the buyer's first transaction public key. Using the obtained buyer's first transaction public key, the service provider encrypts the plaintext data to be encrypted using the first asymmetric encryption algorithm to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm.

[0096] Preferably, this embodiment will be specifically described using the encryption process of asymmetric RSA encryption algorithm and asymmetric ECC encryption algorithm.

[0097] First, the encryption processes of asymmetric RSA and asymmetric ECC encryption algorithms need to be explained in detail:

[0098] The encryption process of the asymmetric RSA encryption algorithm is as follows: the sender encrypts the plaintext data to be sent using the receiver's first transaction public key (n, e); specifically, the encryption process is performed by the encryption algorithm function: b = k e modn encrypts plaintext data, where k is the plaintext data and b is the encrypted ciphertext data.

[0099] The encryption process of the asymmetric ECC encryption algorithm is as follows: following the method flow of generating the first transaction public key G using the asymmetric ECC encryption algorithm, the buyer or service provider encodes the plaintext data to be transmitted onto a point Q on Fp(i,j) and generates a random integer d. The ciphertext data is generated through the encryption function: B = {dP, Q + dQ}, where P is the base point and B is the generated ciphertext data.

[0100] In the above embodiments, the sending end encrypts the plaintext data to be sent using the first transaction public key of the asymmetric RSA encryption algorithm or the asymmetric ECC encryption algorithm to generate ciphertext data, generates a ciphertext key field from the ciphertext data, and de-identifies and privately saves the ciphertext key field to the distributed database using the AES encryption algorithm.

[0101] S6. During the decryption period, the buyer obtains plaintext data by entering the buyer's second transaction private key, and the service provider obtains plaintext data by entering the service provider's second transaction private key.

[0102] Preferably, for ease of understanding, this embodiment provides a detailed description of the decryption process of the asymmetric RSA encryption algorithm in this invention.

[0103] The decryption process of the asymmetric RSA encryption algorithm is as follows:

[0104] By decrypting the algorithm function: k = b d modn decrypts the ciphertext data to obtain the plaintext data, where k is the plaintext data, b is the ciphertext data, and (n, d) is the first transaction private key of the buyer or service provider.

[0105] Preferably, the decryption process of the asymmetric RSA encryption algorithm in this embodiment is as follows: taking the second transaction private key key field as F = h(x)f(y) as an example, where x = (n, e), y = (n, d), h(x) and f(y) are multivariate functions, and the private key key field is b = k e If modn, then (n, e) in the private key field is the first transaction public key of the buyer or service provider, k is the plaintext data, and b is the ciphertext data.

[0106] In the above embodiments, preferably, the process of decrypting the second transaction private key using the asymmetric RSA encryption algorithm is as follows:

[0107] The buyer or service provider inputs a second transaction private key. First, by determining the second transaction private key key field, a multivariate function F = h(x)f(y) is extracted from the second transaction private key key field, where x is the first transaction public key and y is the first transaction private key. Then, the multivariate function F = h(x)f(y) is converted into a random seed. The random seed converted by the multivariate function is compared with the second transaction private key input by the buyer or service provider. If they are the same, it proves that the second transaction private key input by the buyer or service provider is correct; if they are different, it proves that the second transaction private key input by the buyer or service provider is incorrect. If the second transaction private key input by the buyer or service provider is correct, the first transaction private key y(n, d) of the buyer or service provider is obtained based on the form of the multivariate function and the known first transaction public key x(n, e).

[0108] In the above technical solution, using the obtained first transaction private key x(n, d) from the receiving end, the private key key field generated after encrypting the first transaction public key x(n, e) from the receiving end is determined in the distributed database. The ciphertext data b is extracted from the private key key field and decrypted using the asymmetric RSA encryption algorithm. The decryption function of the asymmetric RSA encryption algorithm is: k = b d The k obtained by modulo n is the plaintext data. The system transmits the plaintext data k to the receiving end, which is represented by the buyer or the service provider.

[0109] It should be understood that the above embodiments are one or more embodiments of the present invention, and there are many other embodiments and variations based on the present invention; any variations and modifications made by those skilled in the art through the present invention without making pioneering innovations are all within the protection scope of the present invention.

Claims

1. A method for data security and privacy protection in jewelry transactions based on 5G converged applications, characterized in that, Includes the following steps: S1. Establish a distributed database, which is used to store important encrypted data using an anonymized and encrypted method; S2. Generate a key pair for the buyer and a key pair for the service provider using a pre-established first asymmetric encryption neural network model. The key pair includes a public key and a private key. The public key is regarded as the first transaction public key of the buyer or the service provider, and the private key is regarded as the first transaction private key of the buyer or the service provider. S3. Calculate a random multivariate function on the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function includes the first transaction public key and the first transaction private key of the buyer and the service provider. The random multivariate function is then de-identified and saved to the distributed database using the AES encryption algorithm. S4. Obtain the random seed of the random multivariate function through the key derivation function KDF of the second hash algorithm, and use the obtained random seed as the second transaction private key of the buyer or service provider. S5. During the encryption period, the buyer queries and obtains the service provider's first transaction public key, and uses the obtained service provider's first transaction public key to encrypt the plaintext data to be encrypted through the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm. The service provider queries and obtains the buyer's first transaction public key, and uses the obtained buyer's first transaction public key to encrypt the plaintext data to be encrypted through the first asymmetric encryption algorithm process to obtain ciphertext data. The ciphertext data is then de-identified and saved to the distributed database using the AES encryption algorithm. S6. During the decryption period, the buyer obtains plaintext data by entering the buyer's second transaction private key, and the service provider obtains plaintext data by entering the service provider's second transaction private key. The first asymmetric encryption neural network model in step S2 includes: Input layer, first convolutional layer, second convolutional layer, activation layer, third convolutional layer, pooling layer, output layer, and second asymmetric encryption algorithm; The following steps are performed using the first asymmetric encryption neural network model to generate the buyer's key pair: The buyer inputs the jewelry information they purchased through the input layer. The first key feature is obtained by extracting features from the jewelry information purchased by the buyer through the first convolutional layer. The second key features are obtained by extracting features from the first key features through the second convolutional layer; The second key feature is nonlinearly transformed by the activation layer, and the third key feature is obtained by feature extraction from the nonlinear transformation result by the third convolutional layer. The third key feature is pooled through the pooling layer, and the pooling result is output through the output layer. The output of the output layer is used as input to the second asymmetric encryption algorithm to generate the first transaction public key and the first transaction private key.

2. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 1, characterized in that, Feature extraction is performed on the jewelry information purchased by the buyer. The jewelry information includes digital images and product information, wherein the product information includes jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a six-sided image of the jewelry. A first jewelry image is generated based on the jewelry information, wherein the first jewelry image is used to encrypt and store the product information in the form of an image, and is stored in the distributed database; A jewelry product image is generated based on the first jewelry image and the digital image, wherein the digital image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, and image features of the people involved in the transaction; The jewelry product image is input into a pre-trained Xception model to extract key features, resulting in a key feature vector of jewelry information; wherein, the Xception model is a variant of the Inception series, which introduces depthwise separable convolution.

3. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 1, characterized in that, Step S3, which involves calculating the first transaction public key and the first transaction private key of the buyer and the service provider using a random multivariate function, includes: the multivariate function being a system-generated random multivariate function: R = f(x, y), and substituting the first transaction public key and the first transaction private key of the buyer and the service provider into the multivariate function for calculation; wherein, in the multivariate function R = f(x, y), x is regarded as the first transaction public key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, y is regarded as the first transaction private key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, and R is the result of the multivariate function.

4. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 1, characterized in that, Step S4 includes: performing a key derivation function KDF calculation on the multivariate function R = f(x, y) using the key derivation function KDF of the second hash algorithm, converting the multivariate function into a random seed, and using the obtained random seed as the second transaction private key of the buyer or service provider; the key derivation function KDF calculation on the multivariate function R = f(x, y) refers to performing the key derivation function KDF calculation on the entire function expression of the multivariate function, rather than only performing the key derivation function KDF calculation on the result R of the multivariate function.

5. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 4, characterized in that, Step S4 also includes: calculating a multivariate function R = f(x, y) through the key derivation function KDF, storing the result in a distributed database, and generating a second transaction private key key field. The second transaction private key key field is de-identified and protected for privacy using the AES encryption algorithm. The key field represents the address of each column of the data table in the database.

6. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 1, characterized in that, Step S5 includes: the buyer refers to a jewelry purchaser, and the service provider refers to a jewelry purchase service provider; during the encryption period, the buyer obtains the service provider's first transaction public key through query, and uses the obtained service provider's first transaction public key to encrypt the plaintext data to be sent through the first asymmetric encryption neural network model encryption process. The encrypted ciphertext data then generates a ciphertext key field, and the generated ciphertext key field is saved in a distributed database. The ciphertext key field is de-identified and protected for privacy through the AES encryption algorithm.

7. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 1, characterized in that, Step S6 includes: During the decryption period, the service provider inputs the second transaction private key, and the system verifies and decrypts it; the system first locks the second transaction private key key field of the buyer, and verifies the multivariate function R = f(x, y) of the second transaction private key key field to verify whether the random seed transformed by the multivariate function R = f(x, y) is the same as the buyer's second transaction private key. If they are the same, it proves that the second transaction private key input by the buyer is correct.

8. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 7, characterized in that, Step S6 further includes: verifying the correctness of the service provider's second transaction private key through the multivariate function R = f(x, y) stored in the second transaction private key key field, where x is the first transaction public key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm, and y is the first transaction private key obtained by the first asymmetric encryption neural network model or the first asymmetric encryption algorithm; since the service provider's first transaction public key x has been made public, the service provider's first transaction private key y is obtained through the multivariate function R = f(x, y).

9. The method for protecting jewelry transaction data security and privacy based on 5G converged applications according to claim 8, characterized in that, Step S6 further includes: determining the ciphertext key field generated after being encrypted by the first transaction public key of the service provider using the obtained first transaction private key y of the service provider, decrypting the ciphertext data through the decryption process of the first asymmetric encryption algorithm, finally obtaining the plaintext data, and outputting the obtained plaintext data to the service provider; the plaintext data obtained by the buyer is the same as the aforementioned process.

Images