Jewelry transaction data security and privacy protection method based on 5G fusion application

Abstract

The invention discloses a jewelry transaction data security and privacy protection method based on a 5G fusion application. The method comprises the following steps: establishing a distributed database; generating a first transaction public key and a first transaction private key of two transaction parties through a pre-established first asymmetric encryption algorithm; carrying out random multivariate function calculation on the first transaction public key and the first transaction private key of the two transaction parties, and desensitizing the random multivariate function through an encryption algorithm and storing the random multivariate function in a distributed database; solving a random seed of the random multivariate function as a second transaction private key of the two transaction parties through a key derivation function of a second Hash algorithm; in the encryption period, the transaction data of the two transaction parties are encrypted through an encryption algorithm, and meanwhile, the encrypted data are desensitized and stored; in the decryption period, the two transaction parties obtain plaintext data through the second transaction private key; on the basis of an existing encryption algorithm, the neural network algorithm is introduced to generate the purchaser key, and meanwhile, desensitization processing is performed on the ciphertext data, so that jewelry transaction data is safer.

Description

Technical Field

[0001] The present invention relates to the field of cryptography technology, and in particular to a method for protecting jewelry transaction data security and privacy based on 5G converged applications. Background Art

[0002] The rapid development of 5G technology presents new opportunities and challenges for the jewelry trading industry. 5G networks support higher data rates and lower latency, which increases the risk of data leakage by transmitting large amounts of data across the network. Furthermore, in the 5G era, with the interconnection of everything and the vast number of devices connected to the network, data sources are becoming more diverse, further increasing the difficulty of protecting privacy.

[0003] Data security and privacy protection are particularly important in the jewelry trade. Jewelry transactions involve a vast amount of sensitive information, such as personal customer information, transaction records, and jewelry appraisal certificates. Leakage of this information can not only harm the interests of customers but also impact the reputation and business of jewelry companies. Therefore, the jewelry trade 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 of the buyer and the service provider requires different processing. For example, in jewelry transactions, the buyer mostly conducts transactions 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 easily leads to data leakage.

[0005] Most existing jewelry transaction data uses asymmetric encryption algorithms. Encrypting jewelry transaction data using asymmetric encryption algorithms is relatively safe, but 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 the security of asymmetric encryption algorithms is constantly decreasing. Summary of the Invention

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

[0007] The specific application is as follows:

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

[0009] S1. Establishing a distributed database, wherein the distributed database is used to store important encrypted data in a desensitizing and encryption manner;

[0010] S2. Generate a key pair for the buyer using a pre-established first asymmetric encryption neural network model and a key pair for the service provider using a first asymmetric encryption algorithm. The key pair includes a public key and a private key. The public key is considered the first transaction public key of the buyer or the service provider, and the private key is considered 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 purchaser and the service party, where the random multivariate function includes the first transaction public key and the first transaction private key of the purchaser and the service party, and desensitize the random multivariate function using the AES encryption algorithm and save it in the distributed database;

[0012] S4. Obtaining a random seed of the random multivariate function using a key derivation function KDF of a second hash algorithm, and using the obtained random seed as a second transaction private key of the purchaser or service provider;

[0013] S5. During the encryption period, the purchaser queries and obtains the first transaction public key of the service provider, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the service provider through the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database; the service provider queries and obtains the first transaction public key of the purchaser, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the purchaser through the first asymmetric encryption algorithm process to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database;

[0014] S6. During the decryption period, the purchaser obtains the plaintext data by inputting the purchaser's second transaction private key, and the service provider obtains the plaintext data by inputting 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 by the first asymmetric encryption neural network model to generate a key pair for the purchaser:

[0018] Input the jewelry information purchased by the buyer through the input layer;

[0019] Extracting features of the jewelry information purchased by the buyer through the first convolutional layer to obtain a first key feature;

[0020] Performing feature extraction on the first key feature through the second convolutional layer to obtain a second key feature;

[0021] Performing a nonlinear transformation on the second key feature through the activation layer, and performing feature extraction on the nonlinear transformation result through the third convolution layer to obtain a third key feature;

[0022] The third key feature is pooled through the pooling layer, and the pooling operation result is output through the output layer, wherein the output of the output layer is used to input 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, wherein the jewelry information includes a digitized image and product information, wherein the product information includes the jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a jewelry hexahedron image;

[0024] generating a first jewelry image 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] Generating a jewelry product image based on the first jewelry image and the digitized image, wherein the digitized image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, and image features of both parties to the transaction;

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

[0027] Furthermore, in step S3, the calculation of a random multivariate function on the first transaction public key and the first transaction private key of the purchaser and the service party 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 purchaser and the service party are substituted into the multivariate function for calculation; wherein, x in the multivariate function R = f(x, y) 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] Furthermore, step S4 includes: performing a key derivation function KDF calculation on the multivariate function R=f(x, y) through the key derivation function KDF of the second hash algorithm, converting the multivariate function into a random seed, and treating the obtained random seed as the second transaction private key of the purchaser or service provider; the performing a key derivation function KDF calculation on the multivariate function R=f(x, y) refers to performing a key derivation function KDF calculation on the entire functional formula of the multivariate function, rather than only performing a key derivation function KDF calculation on the multivariate function result R.

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

[0030] Furthermore, step S5 includes: the purchaser refers to the jewelry purchaser, and the service provider refers to the jewelry purchasing service provider; during the encryption period, the purchaser obtains the first transaction public key of the service provider through query, and encrypts the plaintext data to be sent through the first asymmetric encryption neural network model encryption process using the obtained first transaction public key of the service provider, and the encrypted ciphertext data regenerates the ciphertext key field, and the generated ciphertext key field is saved in the distributed database, and the ciphertext key field is desensitized and protected for privacy through the AES encryption algorithm.

[0031] Furthermore, step S6 includes: during the decryption period, the service party inputs the second transaction private key, and the system verifies and decrypts; the system verifies and decrypts by first locking the second transaction private key key field of the purchaser, and verifies the multivariate function R=f(x, y) of the second transaction private key key field, and verifies whether the random seed converted by the multivariate function R=f(x, y) is the same as the second transaction private key of the purchaser. If they are the same, it proves that the second transaction private key entered by the purchaser is correct.

[0032] Furthermore, step S6 also includes: verifying that the second transaction private key of the service party is correct through the multivariate function R=f(x, y) saved in the second transaction private key key field, wherein 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; if the first transaction public key x of the service party has been made public, then the first transaction private key of the service party is obtained as y through the multivariate function R=f(x, y).

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

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

[0035] An embodiment of the present invention provides a method for establishing a distributed database, generating a key pair of a buyer through a pre-established first asymmetric encryption neural network model and a key pair of a service provider through a first asymmetric encryption algorithm, wherein 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; a random multivariate function is calculated 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, and the random multivariate function is desensitized and saved in the distributed database through the AES encryption algorithm; a random seed of the random multivariate function is obtained through a 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 the service provider; during the encryption period, the buyer queries and obtains the first transaction public key and the first transaction private key of the service provider A transaction public key is used to encrypt the plaintext data to be encrypted by using the first asymmetric encryption neural network model with the obtained first transaction public key of the service party to obtain ciphertext data, and the ciphertext data is desensitized and saved in the distributed database through the AES encryption algorithm; the service party queries and obtains the first transaction public key of the buyer, and encrypts the plaintext data to be encrypted by using the obtained first transaction public key of the buyer through the first asymmetric encryption algorithm process to obtain ciphertext data, and the ciphertext data is desensitized and saved in the distributed database through the AES encryption algorithm; during the decryption period, the buyer obtains the plaintext data by inputting the buyer's second transaction private key, and the service party obtains the plaintext data by inputting the service party's second transaction private key; on the basis of the existing encryption algorithm, the present invention introduces a neural network algorithm to generate the buyer's key, and at the same time desensitizes the ciphertext data, so as to make the jewelry transaction data more secure. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a jewelry transaction data security and privacy protection method based on 5G converged applications provided by an embodiment of the present invention. DETAILED DESCRIPTION

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

[0038] Example 1

[0039] The embodiment of the present invention provides a jewelry transaction data security and privacy protection method based on 5G converged applications, such as Figure 1 , including the following steps:

[0040] The jewelry transaction data security and privacy protection method based on 5G converged application is characterized by comprising the following steps:

[0041] S1. Establishing a distributed database, wherein the distributed database is used to store important encrypted data in a desensitizing and encryption manner;

[0042] It should be noted that the distributed database includes: all jewelry transaction data collected before the system is run; the data in the distributed database will be continuously iterated and 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. The distributed database of this embodiment uses the Redis database creation method to create a distributed database. Redis is a high-performance key-value storage database that supports multiple data structures, such as strings, hashes, lists, etc.

[0043] S2. Generate a key pair for the buyer using a pre-established first asymmetric encryption neural network model and a key pair for the service provider using a first asymmetric encryption algorithm. The key pair includes a public key and a private key. The public key is considered the first transaction public key of the buyer or the service provider, and the private key is considered 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 question. The requirements of the question refer to that the first asymmetric encryption algorithm forms a key pair, and the key 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 to verify the signature; the present invention uses the generated public keys of the two transaction parties as the first transaction public key, and the generated private keys of the two transaction parties as the first transaction private key, and the first transaction public key is disclosed 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 first selected 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. Generate two large prime numbers p and q randomly using RSAV 2.2.

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

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

[0050] 4. Find a number d such that the combination of e and d satisfies ed;

[0051] 5. Generate a public key of (n, e) and a private key of (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 key and private key of the service provider are generated by the above-mentioned first asymmetric encryption algorithm RSA and ECC, that is, the first transaction public key and the first transaction private key of the service provider of the present invention are generated.

[0056] In the above embodiment, specifically, the process of generating the first transaction public key and the first transaction private key of the buyer of the present 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 by the first asymmetric encryption neural network model to generate a key pair for the purchaser:

[0060] Input the jewelry information purchased by the buyer through the input layer;

[0061] Extracting features of the jewelry information purchased by the buyer through the first convolutional layer to obtain a first feature;

[0062] Performing feature extraction on the first key feature through the second convolutional layer to obtain a second feature;

[0063] Performing a nonlinear transformation on the second feature through the activation layer, and performing feature extraction on the nonlinear transformation result through the third convolution layer to obtain a third feature;

[0064] performing a pooling operation on the third feature through the pooling layer, and outputting the pooling operation result through the output layer, wherein the output of the output layer is used to input a second asymmetric encryption algorithm to generate a first transaction public key and a first transaction private key;

[0065] It should be noted that the convolution kernel size of the first convolution layer is 2*1, the convolution step size is 2, and the output vector size is 12*1; the convolution kernel size of the second convolution layer is 3*1, the convolution step size is 2, and the output vector size is 15*1; the convolution kernel size of the third convolution layer is 2*1, the convolution step size is 1; and the convolution kernel size of the pooling layer is 3*1, and the convolution step size is 1. Therefore, this embodiment generates the first transaction public key and the first transaction private key of the buyer of the present invention through the first asymmetric encryption neural network model for encryption to obtain the encryption key for jewelry transactions. This can effectively improve the security of the encryption key, increase the difficulty of cracking the encryption key, and improve the reliability of jewelry transactions. This is also one of the technical highlights of the present invention.

[0066] It should be noted that the output layer output of the first asymmetric encryption neural network model is used to input the second asymmetric encryption algorithm, and the implementation process of generating the first transaction public key and the first transaction private key is similar to the service party selecting the asymmetric RSA encryption algorithm and the ECC encryption algorithm (elliptic curve algorithm) for encryption, which will not be repeated in this embodiment.

[0067] In the above embodiment, specifically, feature extraction is performed on the jewelry information purchased by the buyer, wherein the jewelry information includes a digitized image and product information, wherein the product information includes the jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a jewelry hexahedron image;

[0068] generating a first jewelry image 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] Generating a jewelry product image based on the first jewelry image and the digitized image, wherein the digitized image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, image features of both parties to the transaction, and image features of the jewelry authentication certificate;

[0070] The jewelry product image is input into a pre-trained Xception model to extract key features and obtain a key feature vector of the jewelry information; wherein the Xception model is a variant of the Inception series and 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 for encryption. By introducing the digitized image of the 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 purchaser and the service party, where the random multivariate function includes the first transaction public key and the first transaction private key of the purchaser and the service party, and desensitize the random multivariate function using the AES encryption algorithm and save it in the distributed database;

[0073] Preferably, taking the asymmetric RSA encryption algorithm as an example, the randomly generated multivariate function is: x is regarded as the first transaction public key obtained by the asymmetric RSA encryption algorithm, that is, x = (n, e), and y is regarded as the first transaction private key of the asymmetric RSA encryption algorithm, that is, 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 may be: F=h(x)f(y), wherein 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: x is regarded as the first transaction public key obtained by the asymmetric ECC encryption algorithm, that is, x=G; y is regarded as the first transaction private key obtained by the asymmetric ECC encryption algorithm, that is, y=k; the random multivariate function R=f(x, y) is a function including variables x and y.

[0076] Preferably, assuming that the coordinates of point G are (a, b), the functional relationship satisfying 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 of the multivariate functions R = f(x, y), F = h(x)f(y) and DD = i(x)j(y) or its components are not required to be functions of a specified number of times. It is only required 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 other variables in addition to the first transaction public key x and the first transaction private key y.

[0078] In the above embodiment, specifically, the calculation of a random multivariate function on the first transaction public key and the first transaction private key of the purchaser and the service party 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 purchaser and the service party are substituted into the multivariate function for calculation; wherein, x in the multivariate function R = f(x, y) 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, by using the first asymmetric encryption neural network model to encrypt transaction data for the buyer, and by using the first asymmetric encryption algorithm to encrypt transaction data for the service provider, the principle of the two parties using different forms of encryption is that in jewelry transactions, the buyer encrypts transaction data by adopting the first asymmetric encryption neural network model, by capturing the buyer's jewelry transaction characteristics, and by introducing the digital image of the jewelry information and the feature extraction corresponding to the product information, it can effectively improve the security of the encryption key, increase the difficulty of cracking the encryption key, and improve the reliability of jewelry transactions. In addition, the service provider does not need to capture the features in the transaction as input to the asymmetric encryption algorithm, and the service provider does not need to consume neural network resources for complex encryption calculations. By using differentiated encryption processing methods by the buyer and the service provider, the problem of easy leakage of the buyer's data is solved. This is also one of the technical highlights of the present invention.

[0080] S4. Obtaining a random seed of the random multivariate function using a key derivation function KDF of a second hash algorithm, and using the obtained random seed as a second transaction private key of the purchaser or service provider;

[0081] Specifically, a method for obtaining a random seed of the random multivariate function and using the obtained random seed as the second transaction private key of the purchaser or the service provider is as follows: performing KDF calculation on the multivariate function R=f(x, y) through the key derivation function KDF of the second hash algorithm, converting the multivariate function into a random seed, and treating the obtained random seed as the second transaction private key of the purchaser or the service provider, sending the second transaction private key to the purchaser or the service provider, and requiring only the purchaser or the service provider to know it; performing 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 functional formula of the multivariate function, rather than only performing the key derivation function KDF calculation on the multivariate function result R.

[0082] Preferably, a key derivation function KDF is used to perform a key derivation function KDF calculation on the multivariate function R=f(x, y), and the multivariate function is converted into a 64-bit random seed, which is regarded as the second transaction private key of the purchaser 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 saved in a distributed database to generate a second transaction private key field, and the second transaction private key field is desensitized and privacy protected by the AES encryption algorithm.

[0084] It should be noted that the desensitization privacy protection in this embodiment cannot use hash desensitization. Hash desensitization is irreversible and the original data cannot be restored. The desensitization method in this embodiment can use encryption desensitization: reversible, a key is required to restore the original data; replacement desensitization: reversible, the replacement rules and original data need to be recorded; mask desensitization: reversible, the mask rules and original data need to be recorded; obfuscation desensitization: reversible, the obfuscation rules and original data need to be recorded.

[0085] It should be noted that the specific process of using the key derivation function KDF to generate a random seed is as follows:

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

[0087] 1. Choose a hash function

[0088] Select an appropriate 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.

[0089] 2. Prepare input data

[0090] Prepare one or more input data. These data 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] Use the selected key derivation function KDF to hash the input data and generate a hash value of fixed length;

[0093] 4. Extract random seeds

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

[0095] S5. During the encryption period, the purchaser queries and obtains the first transaction public key of the service provider, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the service provider through the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database; the service provider queries and obtains the first transaction public key of the purchaser, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the purchaser through the first asymmetric encryption algorithm process to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database;

[0096] Preferably, this embodiment is specifically described through the encryption process of the asymmetric RSA encryption algorithm and the asymmetric ECC encryption algorithm.

[0097] First, we need to explain in detail the encryption process of the asymmetric RSA encryption algorithm and the asymmetric ECC encryption algorithm:

[0098] The encryption process of the asymmetric RSA encryption algorithm is as follows: the sender encrypts the plaintext data to be sent using the first transaction public key (n, e) of the receiver; 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: the method flow of generating the first transaction public key G by the asymmetric ECC encryption algorithm, the purchaser or service provider encodes the plaintext data to be transmitted to a point Q on Fp(i, j), and generates a random integer d, and generates ciphertext data 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 embodiment, the sending end encrypts the plaintext data to be sent through 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 through the ciphertext data, and desensitizes the ciphertext key field through the AES encryption algorithm and saves it in a distributed database.

[0101] S6, during the decryption phase, the buyer obtains the plaintext data by inputting the buyer's second transaction private key, and the service provider obtains the plaintext data by inputting the service provider's second transaction private key;

[0102] Preferably, for ease of understanding, this embodiment specifically describes the decryption process of the asymmetric RSA encryption algorithm in the present invention.

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

[0104] Through the decryption 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 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 field is b=k e 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 embodiment, preferably, the process of decrypting the second transaction private key using the asymmetric RSA encryption algorithm is as follows:

[0107] The purchaser or service provider inputs the second transaction private key. First, the purchaser or service provider's second transaction private key determines the second transaction private key key field and extracts the multivariate function F=h(x)f(y) of 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, and the random seed converted by the multivariate function is compared to see whether it is the same as the second transaction private key input by the purchaser or service provider. If they are the same, it proves that the second transaction private key input by the purchaser or service provider is correct. If they are not the same, it proves that the second transaction private key input by the purchaser or service provider is incorrect. If the second transaction private key input by the purchaser or service provider is correct, then the purchaser or service provider's first transaction private key y(n, d) is obtained based on the multivariate function form and the known first transaction public key x(n, e).

[0108] In the above technical solution, by obtaining the first transaction private key x(n, d) of the receiving end, the private key field generated by encrypting the first transaction public key x(n, e) of the receiving end is determined in the distributed database, the ciphertext data b in the private key field is extracted, and the ciphertext data b is decrypted through the decryption process of the asymmetric RSA encryption algorithm. The decryption process function of the asymmetric RSA encryption algorithm is: k = b d modn, the obtained k is the plaintext data, and the system transmits the plaintext data k to the receiving end, which is represented by the purchaser 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 thereof based on the present invention; the variations and modifications made by ordinary technicians in this industry through the present invention without making groundbreaking innovations all fall within the scope of protection of the present invention.

Claims

1. A jewelry transaction data security and privacy protection method based on 5G converged applications is characterized by: The following steps are involved: S1. Establishing a distributed database, wherein the distributed database is used to store important encrypted data in a desensitizing and encryption manner; S2. Generate a key pair for the buyer using a pre-established first asymmetric encryption neural network model and a key pair for the service provider using a first asymmetric encryption algorithm. The key pair includes a public key and a private key. The public key is considered the first transaction public key of the buyer or the service provider, and the private key is considered 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 purchaser and the service party, where the random multivariate function includes the first transaction public key and the first transaction private key of the purchaser and the service party, and desensitize the random multivariate function using the AES encryption algorithm and save it in the distributed database; S4. Obtaining a random seed of the random multivariate function using a key derivation function KDF of a second hash algorithm, and using the obtained random seed as a second transaction private key of the purchaser or service provider; S5. During the encryption period, the purchaser queries and obtains the first transaction public key of the service provider, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the service provider through the encryption process of the first asymmetric encryption neural network model to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database; the service provider queries and obtains the first transaction public key of the purchaser, encrypts the plaintext data to be encrypted using the obtained first transaction public key of the purchaser through the first asymmetric encryption algorithm process to obtain ciphertext data, desensitizes the ciphertext data using the AES encryption algorithm and saves it in the distributed database; S6. During the decryption period, the purchaser obtains the plaintext data by inputting the purchaser's second transaction private key, and the service provider obtains the plaintext data by inputting the service provider's second transaction private key.

2. The jewelry transaction data security and privacy protection method based on 5G converged application according to claim 1 is characterized in that: 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 by the first asymmetric encryption neural network model to generate a key pair for the purchaser: Input the jewelry information purchased by the buyer through the input layer; Extracting features of the jewelry information purchased by the buyer through the first convolutional layer to obtain a first key feature; Performing feature extraction on the first key feature through the second convolutional layer to obtain a second key feature; Performing a nonlinear transformation on the second key feature through the activation layer, and performing feature extraction on the nonlinear transformation result through the third convolution layer to obtain a third key feature; The third key feature is pooled through the pooling layer, and the pooling operation result is output through the output layer, wherein the output of the output layer is used to input the second asymmetric encryption algorithm to generate the first transaction public key and the first transaction private key.

3. The jewelry transaction data security and privacy protection method based on 5G converged applications according to claim 2 is characterized in that: Extracting features from jewelry information purchased by the buyer, wherein the jewelry information includes a digitized image and product information, wherein the product information includes the jewelry name, jewelry material, jewelry model, jewelry weight, jewelry production time, and a jewelry hexahedron image; generating a first jewelry image 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; Generating a jewelry product image based on the first jewelry image and the digitized image, wherein the digitized image includes spatial features, color features, texture features, brightness and contrast features, resolution features, sales location features, and image features of both parties to the transaction; The jewelry product image is input into a pre-trained Xception model to extract key features and obtain a key feature vector of the jewelry information; wherein the Xception model is a variant of the Inception series and introduces depthwise separable convolution.

4. The jewelry transaction data security and privacy protection method based on 5G converged applications according to claim 1 is characterized in that: In step S3, the calculation of a random multivariate function on the first transaction public key and the first transaction private key of the purchaser and the service party 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 purchaser and the service party are substituted into the multivariate function for calculation; wherein, x in the multivariate function R = f(x, y) 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.

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

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

7. The jewelry transaction data security and privacy protection method based on 5G converged application according to claim 1 is characterized in that: Step S5 includes: the purchaser refers to the jewelry purchaser, and the service provider refers to the jewelry purchasing service provider; during the encryption period, the purchaser obtains the first transaction public key of the service provider through query, and encrypts the plaintext data to be sent using the first asymmetric encryption neural network model encryption process using the obtained first transaction public key of the service provider. The encrypted ciphertext data then generates a ciphertext key field, and the generated ciphertext key field is saved in a distributed database, and the ciphertext key field is desensitized and protected for privacy using the AES encryption algorithm.

8. The jewelry transaction data security and privacy protection method based on 5G converged application according to claim 1 is characterized in that: Step S6 includes: during the decryption period, the service party inputs the second transaction private key, and the system verifies and decrypts; the system verifies and decrypts by first locking the second transaction private key key field of the purchaser, and verifies the multivariate function R=f(x, y) of the second transaction private key key field, and verifies whether the random seed converted by the multivariate function R=f(x, y) is the same as the second transaction private key of the purchaser. If they are the same, it proves that the second transaction private key entered by the purchaser is correct.

9. The jewelry transaction data security and privacy protection method based on 5G converged applications according to claim 8 is characterized in that: Step S6 also includes: verifying that the second transaction private key of the service party is correct through the multivariate function R=f(x, y) saved 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; if the first transaction public key x of the service party has been made public, then the first transaction private key of the service party is obtained as y through the multivariate function R=f(x, y).

10. The jewelry transaction data security and privacy protection method based on 5G converged application according to claim 9 is characterized in that: Step S6 also includes: determining the ciphertext key field generated after encryption by the service party's first transaction public key through the obtained service party's first transaction private key y, decrypting the ciphertext data through the decryption process of the first asymmetric encryption algorithm, and finally obtaining the plaintext data, and outputting the obtained plaintext data to the service party; the plaintext data obtained by the purchaser is the same as the aforementioned process.

Images