Privacy protection system for counting user property information

Through the combination of SM2 algorithm and the random forest model, efficient and secure total user deposit statistics in the double-blind scenario between financial institutions and banks are achieved, and the inefficiency and privacy leakage problems in non-balanced scenarios in the existing technology are solved, and are applicable to cross-institutional property statistics and financial supervision.

CN120372681APending Publication Date: 2025-07-25XIAN UNIV OF POSTS & TELECOMM
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510437377.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-09
Publication Date
2025-07-25

AI Technical Summary

Technical Problem

The existing privacy collection interception technology is inefficient in non-balanced scenarios, has limited application scenarios, and cannot achieve double-blind privacy protection, and there is a problem of correlation leakage of ciphertext aggregation statistics.

Method used

Homomorphic encryption technology based on SM2 algorithm is adopted, combined with the random forest model and joint public key, and the identity information and amount are encrypted layer by layer, the total user deposit statistics in the double-blind scenarios between financial institutions and banks are realized.

Benefits of technology

While protecting user privacy, it improves computing efficiency and security, and is suitable for cross-institutional property statistics and financial supervision scenarios, realizing privacy collection requests in non-balanced scenarios, ensuring that the data privacy of all parties is not leaked.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120372681A_ABST
    Figure CN120372681A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of privacy set intersection, and discloses a privacy protection system for counting user property information, which comprises an identity information encryption module, a bank internal query module and a joint encryption transmission module, and the identity information encryption module performs homomorphic encryption on identity information by a financial institution based on an SM2 algorithm, and the financial institution encrypts an identity card number of a user to be inquired and sends a result to a bank. According to the method, a privacy set intersection protocol is designed by utilizing a national cryptographic SM2 algorithm, and the calculation efficiency is ensured while the privacy of set elements is protected. A financial institution queries the total amount of deposit of a certain person in each bank, each bank is unknown to the information of a queried user, and a query institution is unknown to the deposit information of the queried user in each bank, so that the practical application of the privacy set intersection system in an unbalanced scene is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of private set intersection, and specifically relates to a privacy protection system for statistically analyzing user property information. Background Art

[0002] With the development of the digital society, the privacy and security issues of data have become increasingly important. Many organizations and individuals hope to share some data for cooperation and analysis, but do not want to disclose other sensitive information of their own. Therefore, in recent years, the domestic and international demand for data security and privacy protection has been increasing, and the research on secure multi-party computing technology has gradually deepened. With the continuous optimization and breakthrough of homomorphic encryption technology and its gradual commercialization, PSI, as an important application of secure multi-party computing, has been widely used in privacy computing and has important theoretical and practical significance.

[0003] The national cryptographic algorithms are a series of cryptographic algorithm standards promulgated by the Cryptography Administration of China. They are widely used because they have the characteristics of high efficiency while ensuring security. Private set intersection is a common privacy protection technology that can compare the common elements in two data sets without exposing individual information.

[0004] The existing inventions on using national cryptography to implement private set intersection technology mainly generate public keys, private keys, and random numbers for two sets of national cryptographic algorithms; encrypt the local data of both parties using national cryptographic algorithms to generate the first and second ciphertexts; then extract different parts of the ciphertext and achieve ciphertext interaction through secondary encryption and combination operations; finally, calculate the intersection result based on the ciphertext segments, which is applicable to the private set intersection of two parties.

[0005] Existing drawbacks:

[0006] 1. Inefficient optimization for unbalanced scenarios: For the "unbalanced data volume" in actual financial scenarios, such as when the querying party only queries a single user and the data party stores a large amount of user data, its design may be less efficient when the data volume difference is large.

[0007] 2. Limited application scenarios: It is not clearly applicable to the multi-party cross-institutional property statistics scenario.

[0008] 3. Failure to solve the problem of associated leakage in ciphertext aggregation statistics: When statistically analyzing the total deposits of users in multiple institutions, the existing technology design exposes the associated details of users' cross-institutional data.

[0009] 4. Lack of support for "double-blind" privacy constraints: That is, it cannot achieve that the querying party does not know the data distribution and the data party does not know the identity of the queried user.

[0010] Therefore, it needs to be optimized and improved. Summary of the Invention

[0011] The purpose of the present invention is to provide a privacy protection system for statistically analyzing user property information to solve the problems raised in the above background art.

[0012] To achieve the above object, the present invention provides the following technical solution: A privacy protection system for statistically analyzing user property information, the privacy protection system includes an identity information encryption module, an internal bank query module, and a joint encryption transmission module;

[0013] The identity information encryption module uses the financial institution to perform homomorphic encryption of identity information based on the SM2 algorithm. The financial institution encrypts the ID number of the user to be queried and sends the result to the bank;

[0014] The internal bank query module uses the bank to perform an equivalence test using a random forest model, compares the received user information, records the weight of the matching information, and filters out the deposit amount that matches the user;

[0015] The joint encryption transmission module encrypts the deposit amount by each bank using the joint public key, then sums them up by the financial institution, and then decrypts them step by step by each bank. Finally, the total amount is decrypted by the financial institution.

[0016] Preferably, the method steps for performing homomorphic encryption of identity information based on the SM2 algorithm include:

[0017] Step 1, system establishment: Let A and B be the financial institution and the bank respectively. The financial institution and the bank pre-set the same public parameter p as a large prime number in advance, E is an elliptic curve defined over the finite field F p On, G is the q-order base point of the elliptic curve, and the bit length mlen of the plaintext space M is set to 32 bits; If [k]G represents the k-fold point of the elliptic curve base point G, and the cryptographic hash function H: {0,1} → Z q. x||y represents the concatenation of the byte string or bit string x and y; Represents the exclusive OR operation between bit strings;

[0018] Step 2, key generation: User B randomly selects a private key Discloses the public key pk = [sk]G;

[0019] Step 3, encryption: Suppose the message that user A needs to send to user B is m, and user A needs to perform the following calculations:

[0020] Calculate the elliptic curve point S = [h]pk, where h = 1. If S is an infinite point, report an error and exit;

[0021] Select a random number Calculate c1 = [k]G = (x1, y1), [k]pk = (x2, y2);

[0022] Calculate c2 = [m]G + [k]pk;

[0023] c3 = H(x2 || m || y2);

[0024] Output the ciphertext c = (c1, c2, c3);

[0025] The financial institution encrypts the ID number of the person to be queried in five parts through this algorithm and sends the encryption result to the bank;

[0026] Step Four, Decryption: User B receives the ciphertext c = (c1, c2, c3) sent by User A and performs arithmetic decryption.

[0027] Preferably, the decryption method in the said Step Four includes:

[0028] Step One, Verify whether c1 satisfies the elliptic curve equation. If it does not satisfy, report an error and exit;

[0029] Step Two, Calculate the elliptic curve point S = [h]c1, where h = 1. If S is the infinite point, report an error and exit;

[0030] Step Three, Calculate the elliptic curve point [sk]c1 = (x2, y2), [m]G = c2 - [sk]c1;

[0031] Step Four, Decrypt to obtain m[G], and thus use this value for data comparison.

[0032] Preferably, the identity information encryption module includes information that uniquely identifies the user's identity with the ID number, shortening the time used in the subsequent equivalent test process;

[0033] Specifically, divide the 18 - bit identity information. The division criteria are: province, city - district, birth year, birth month - day, and other remaining information; improve the operation efficiency of each bank in matching the identity information of the user to be queried through a hierarchical method; the financial institution encrypts the user's ID number using the homomorphic encryption scheme of the SM2 standard and sends the encrypted result to the bank for comparison to improve the matching efficiency and protect the user's privacy. The bank stratifies the personal information of the deposit users in the same way for subsequent query work.

[0034] Preferably, the random forest model is set based on the ID number, where each layer of nodes respectively corresponds to the hierarchical identity information. The bank compares the internal customer information with the received encrypted identity information, records the weight of the matching information, and encrypts the weight using the public key of the financial institution. By traversing the random forest, select the amount corresponding to the route with the smallest weight as the query result to ensure the confidentiality of the customer information of the bank.

[0035] Preferably, the equivalence test includes encrypting the node information using the same encryption method, taking a random number r, encrypting r, and performing calculations based on the ID card information provided by the financial institution, the information represented by each layer of nodes in the random forest, and the random number. The encryption algorithms involved are all homomorphic encryption schemes conforming to the national cryptographic standard SM2;

[0036] When E(x i -a i ) r takes 0, it means that the equivalence test result is a node with a weight E(w) of 0, and the result (∑w i = 0, E(m)) is sent to the financial institution;

[0037] where x i corresponds to the ID card information provided by the financial institution, and a i corresponds to the corresponding information represented by each layer in the random forest.

[0038] Preferably, the joint encryption transmission module includes the private keys provided by the financial institution and each bank respectively. Based on the elliptic curve algorithm, the independent public keys of the institution and the bank are obtained from the private keys, and finally the joint public key of the financial institution and the bank is obtained;

[0039] This joint public key is used by each bank to encrypt the deposit amount; the SM2 encryption algorithm is used to encrypt the compared amount, where the encryption public key is provided by the financial institution as k1 and by each bank as k2......k n+1 , and by modifying the k value in the SM2 algorithm, the encryption of the joint public key is achieved.

[0040] Preferably, the specific steps for the financial institution to obtain the total deposit amount are as follows: each bank encrypts the amount results obtained from its own queries using the joint public key with additive homomorphic property, and then sends the results to the financial institution respectively. After the financial institution sums up the ciphertexts of each amount, it sends it to one of the banks for partial decryption, and then gradually decrypted by other banks, and the last step is handed over to the financial institution for decryption to obtain the final total amount M.

[0041] Preferably, this privacy protection system also includes an efficiency optimization method for unbalanced data volume, specifically by using a five-layer hierarchical method for the node information stored inside the bank, combining with the random forest model for equivalence testing, reducing unnecessary computational amounts, improving the query efficiency in the scenario where the financial institution queries a single user while the bank has a large amount of user data, and ensuring the response speed and performance of the system under high load.

[0042] Preferably, the privacy protection system further includes an enhanced security feature implementation method, specifically using an improved SM2 algorithm with homomorphic properties during the identity information encryption stage, combining the digest generated by the cryptographic hash function for ciphertext verification to enhance the security and integrity of the ciphertext; during the joint public key encryption process, generating a multi-party joint public key by modifying the parameters in the SM2 algorithm to ensure the security of the encryption process and prevent information leakage.

[0043] The beneficial effects of the present invention are as follows:

[0044] 1. The present invention designs a private set intersection protocol using the national cryptography SM2 algorithm, which ensures computational efficiency while protecting the privacy of set elements. By reconstructing the underlying architecture of the PSI protocol with the national cryptography SM2 algorithm, the above technical bottlenecks are overcome, and a systematic balance is achieved among algorithm compliance, efficiency in non-equilibrium scenarios, double-blind privacy isolation, and ciphertext aggregation security. Specifically, the present invention uses this protocol to implement a privacy protection system for statistical user property information, which is used by financial institutions to query the total deposit amount of a person in various banks, while each bank is unaware of the information of the queried user, and the querying institution is unaware of the deposit information of the queried user in each bank, thus realizing the practical application of the private set intersection system in non-equilibrium scenarios;

[0045] 2. The present invention constructs a decision tree by hierarchically encrypting identity information, combines joint public key collaborative decryption and homomorphic encryption technology to realize the statistical calculation of the total user deposit amount by financial institutions and banks in a "double-blind" scenario. Financial institutions only submit encrypted user identities, banks cannot know the queried object, and financial institutions also cannot know the specific deposit distribution. By using hierarchical matching to optimize the query efficiency, the amount is encrypted with additive homomorphic encryption and decrypted step by step through a multi-party joint public key, and finally the ciphertext data is safely aggregated to obtain the total amount. This solution takes into account national cryptography compliance and privacy protection intensity, and is applicable to scenarios such as cross-institutional property statistics and financial supervision. While protecting user privacy and bank data sovereignty, in the identity information encryption stage of our solution, only an improved SM2 algorithm with homomorphic properties is used, which solves the problem of secure data collaboration under non-equilibrium permissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the encryption algorithm of the present invention;

[0047] Figure 2 It is a flowchart of the decryption algorithm of the present invention;

[0048] Figure 3 It is a system flowchart of the present invention;

[0049] Figure 4 It is a flowchart of the implementation of the joint public key of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] As Figures 1 to 4 shown, the embodiment of the present invention provides a privacy protection system for statistically analyzing user property information, and the privacy protection system is composed of an identity information encryption module, an internal bank query module, and a joint encryption transmission module;

[0052] The identity information encryption module uses the financial institution to perform homomorphic encryption on the identity information based on the SM2 algorithm. The financial institution encrypts the ID number of the user to be queried and sends the result to the bank;

[0053] The internal bank query module uses the bank to perform an equivalent test using a random forest model, compares the received user information, records the weight of the matching information, and filters out the deposit amount that matches the user;

[0054] The joint encryption transmission module encrypts the deposit amount by each bank using the joint public key respectively, then sums them up by the financial institution, and then hands them over to each bank for step-by-step decryption. Finally, the total amount is decrypted by the financial institution.

[0055] Among them, the method steps for performing homomorphic encryption on the identity information based on the SM2 algorithm include:

[0056] Step 1, system establishment: Let A and B be the financial institution and the bank respectively. The financial institution and the bank pre-set the same public parameter p as a large prime number in advance. E is an elliptic curve defined over the finite field F p p. Let G be the q-order base point of the elliptic curve, and the bit length mlen of the plaintext space M is set to 32 bits; If [k]G represents the k-fold point of the elliptic curve base point G, and the cryptographic hash function H: {0, 1} → Z q .x||y represents the concatenation of the byte string or bit string x and y; represents the exclusive OR operation between bit strings;

[0057] Step 2, key generation: User B randomly selects a private key and publishes the public key pk = [sk]G;

[0058] Step 3, encryption: Suppose the message m that user A needs to send to user B, user A needs to perform the following calculations:

[0059] Calculate the elliptic curve point S = [h]pk, where h = 1. If S is the point at infinity, report an error and exit.

[0060] Select a random number Calculate c1 = [k]G = (x1, y1), [k]pk = (x2, y2);

[0061] Calculate c2 = [m]G + [k]pk;

[0062] c3 = H(x2||m||y2);

[0063] Output the ciphertext c = (c1, c2, c3).

[0064] The c2 part of the standard SM2 scheme is t = KDF(x2||y2, mlen), KDF: {0, 1} * ×mlen → {0, 1} mlen , but since the value of t is not used in the subsequent calculation process of this scheme, there is no need to calculate this value in this scheme. The financial institution encrypts the ID number of the person being queried in five parts through this algorithm and sends the encryption result to the bank;

[0065] Step 4, Decryption: User B receives the ciphertext c = (c1, c2, c3) sent by User A and performs operations for decryption.

[0066] Among them, the decryption method in Step 4 includes:

[0067] Step 1, Verify whether c1 satisfies the elliptic curve equation. If it does not satisfy, report an error and exit;

[0068] Step 2, Calculate the elliptic curve point S = [h]c1, where h = 1. If S is the point at infinity, report an error and exit;

[0069] Step 3, Calculate the elliptic curve point [sk]c1 = (x2, y2), [m]G = c2 - [sk]c1;

[0070] Step 4, Decrypt to obtain m[G], and thus use this value for data comparison.

[0071] Among them, the identity information encryption module includes information that uniquely identifies the user's identity with the ID number, shortening the time used in the subsequent equal value test process;

[0072] Specifically, the 18-digit identity information is divided according to the following criteria: province, city and district, year of birth, month and day of birth, and other remaining information; the operation efficiency of each bank in matching the identity information of the user to be queried is improved through a hierarchical method; financial institutions use the homomorphic encryption scheme of the SM2 standard to encrypt the user's ID number and send the encrypted result to the bank for comparison, so as to improve the matching efficiency and protect the user's privacy. The bank stratifies the personal information of the depositor in the same way for subsequent query work.

[0073] Among them, the random forest model is set based on the ID number. Each layer of nodes corresponds to the stratified identity information respectively. The bank compares the internal customer information with the received encrypted identity information, records the weight of the matching information, and encrypts the weight using the public key of the financial institution. By traversing the random forest, the amount corresponding to the route with the smallest weight is selected as the query result to ensure the confidentiality of the bank's customer information.

[0074] Among them, the equivalent test includes encrypting the node information in the same encryption method, taking a random number r and encrypting r, and calculating based on the ID information provided by the financial institution, the information represented by each layer of nodes in the random forest, and the random number. The encryption algorithms involved are all the homomorphic encryption scheme of the national cryptography SM2 standard;

[0075] When E(x i -a i ) r takes 0, it means that the result of the equivalent test is the node with the weight E(w) of 0, and the result (∑w i =0, E(m)) is sent to the financial institution;

[0076] Among them, x i corresponds to the ID information provided by the financial institution, and a i corresponds to the corresponding information represented by each layer in the random forest.

[0077] Among them, the joint encryption transmission module includes the private keys provided by the financial institution and each bank respectively. Based on the elliptic curve algorithm, the independent public keys of the institution and the bank are obtained according to the private keys, and finally the joint public key of the financial institution and the bank is obtained;

[0078] This joint public key is used by each bank to encrypt the deposit amount; the SM2 encryption algorithm is used to encrypt the compared amount, and the encryption public key is provided by the financial institution k1 and each bank provides k2......k n+1 , and the joint public key encryption is realized by modifying the k value in the SM2 algorithm.

[0079] Among them, the specific steps for the financial law institution to obtain the total deposit amount are as follows: After each bank encrypts the amount results obtained from their respective queries using the joint public key with additive homomorphic properties, the results are sent to the financial institution respectively. After the financial institution sums up the ciphertexts of each amount, it sends it to one of the banks for partial decryption, and then gradually decrypted by other banks. The last step is handed over to the financial institution for decryption to obtain the final total amount M.

[0080] Among them, the privacy protection system also includes an efficiency optimization method for unbalanced data volume. Specifically, by using the node information stored inside the bank in a five-layer hierarchical manner and combining with the random forest model for equivalence testing, unnecessary computational volume is reduced, and the query efficiency in the scenario where the financial institution queries a single user while the bank has a large amount of user data is improved, ensuring the response speed and performance of the system under high load.

[0081] Among them, the privacy protection system also includes an enhanced security feature implementation method. Specifically, in the identity information encryption stage, a modified SM2 algorithm with homomorphic properties is used, combined with the digest generated by the cryptographic hash function for ciphertext verification to enhance the security and integrity of the ciphertext; in the process of joint public key encryption, the generation of the multi-party joint public key is achieved by modifying the parameters in the SM2 algorithm to ensure the security of the encryption process and prevent information leakage.

[0082] Identity Information Encryption

[0083] Since the ID number can uniquely identify the user's identity information, to shorten the time used in the subsequent equivalence testing process, the 18-digit identity information is divided. The division criteria are: province, city, birth year, birth month and day, and other remaining information. The operating efficiency of each bank in matching the identity information of the user to be queried is improved through a hierarchical method. The financial institution encrypts the user's ID number using the homomorphic encryption scheme of the SM2 standard and sends it to the bank. The bank stratifies the personal information of the deposit users in the same way for subsequent query work. Set the model according to the ID number, where each layer of nodes corresponds to the stratified identity information respectively to generate a decision tree, and the node information is encrypted using the same encryption principle. The specific execution process:

[0084] 1) The client C inputs the ID information W of the person to be queried.

[0085] 2) The client C groups the ID information W by province, city, birth year, birth month and day, and general information.

[0086] 3) The client C selects appropriate public and private keys and sends this public key to the bank side.

[0087] 4) The client C encrypts the grouped information respectively through the national cryptographic SM2 algorithm.

[0088] Process:

[0089] Suppose the message that the client C needs to send to the bank B is m, and the client C needs to perform the following calculations:

[0090] a) Calculate the elliptic curve point S = [h]pk, where h = 1. If S is the infinite point, report an error and exit.

[0091] b) Select a random number Calculate

[0092] c1 = [k]G = (x1, y1), [k]pk = (x2, y2);

[0093] c) Calculate c2 = [m]G + [k]pk;

[0094] d) c3 = H(x2 || m || y2);

[0095] e) Output the ciphertext c = (c1, c2, c3)

[0096] 5) Send the output ciphertext to the bank.

[0097] Internal query in the bank

[0098] The financial institutions and each bank respectively provide the private keys. Based on the elliptic curve algorithm, the independent public keys of the institutions and banks are obtained according to the private keys, and finally the joint public key of the financial institutions and banks is obtained. This joint public key is used for the encryption of the deposit amount by each bank. We use the SM2 encryption algorithm to encrypt the compared amount, where the encryption public key is provided by the financial institutions and each bank respectively with keys. By modifying the k value in the SM2 algorithm, the encryption of the joint public key is realized. The specific execution process:

[0099] 1) Each bank randomly selects its own private key and receives the public key sent by the client.

[0100] 2) Each bank groups all the client ID card information in its own bank by province, city, district, birth year, birth date, and general information.

[0101] 3) Encrypt the identity information using the same public key as the client through the national cipher SM2 algorithm.

[0102] 4) Combine the encrypted internal client information of the bank to form a random forest.

[0103] 5) Combine the encrypted identity information with the deposit amount

[0104] 6) Protocol for comparing the encrypted information sent by the client with the random forest generated by the bank side. 7) Take a random number r and encrypt r. When E(x i -a i ) r takes 0, it means that the equivalent test result is a node with a weight E(w) of 0.

[0105] Process:

[0106] a) Given h = g s .

[0107] b) After encryption, the ciphertext

[0108] c) Node information

[0109] d) It can be obtained that c1:c1' = g r-r′ ,

[0110] e) Therefore, it can be known that if x = y, then: x i -y i = 0

[0111] 8) Use the SM2 homomorphic encryption algorithm to encrypt the amount obtained through the random forest comparison.

[0112] 9) The client and each bank jointly generate a joint public key.

[0113] After each bank encrypts the amount result queried by itself with the joint public key with additive homomorphic property, the results are sent to the financial institution respectively. After the financial institution sums up the ciphertexts of each amount, the result is sent to one of the banks for partial decryption, and then gradually decrypted by other banks. The last step is handed over to the financial institution for decryption to obtain the final total amount M. The specific process is as follows:

[0114] a) The client uses its own private key k1 and performs the operation A1 = [k1]G on it.

[0115] b) Bank 1 uses its own private key k2 and performs the operation A2 = [k2]G on it.

[0116] c) Bank 2 uses its own private key k3 and performs the operation A3 = [k3]G on it.

[0117] d) Bank 3 uses its own private key k4 and performs the operation A4 = [k4]G on it.

[0118] e) Bank n uses its own private key kn+1 and performs the operation A = [kn+1]G on it.

[0119] f) Generate the combined public key A = [k1 + k2 + k3 + kn + 1]G

[0120] 10) Each bank uses the combined public key (each bank and the financial department) to perform a homomorphic encryption operation on the amount M to obtain the encrypted amount E(M).

[0121] 11) After each bank encrypts the amount M using the combined public key with additive homomorphic property, it sends the result to the server.

[0122] 12) After the server sums up the ciphertexts of each amount, it sends it to one of the banks for partial decryption.

[0123] 13) After one bank decrypts, it sends it to another bank.

[0124] 14) After all banks decrypt and send it to the server, the server performs the final decryption to obtain the total user deposit amount.

[0125] 15) Decryption process:

[0126] a) The bank receives the ciphertext c = (c1, c2, c3) sent by the client and performs the following operations for decryption:

[0127] b) Verify whether c1 satisfies the elliptic curve equation. If not, report an error and exit;

[0128] c) Calculate the elliptic curve point S = [h]c1, where h = 1. If S is the infinite point, report an error and exit;

[0129] d) Calculate the elliptic curve points [sk]c1 = (x2, y2), [m]G = c2 - [sk]c1;

[0130] Decrypt to obtain m[G], and thus use this value for data comparison.

[0131] Bank internal query process:

[0132] The specific implementation steps of the bank are as follows:

[0133] 1) The bank uses a five-layer hierarchical method to perform an equivalent test on the internal personnel information and the received ciphertext information;

[0134] 2) Denote the weight of the line where E(x - b) = 0 as 0, and encrypt the weight using the public key of the financial institution to keep the personal information confidential for the bank;

[0135] 3) By traversing the random forest, select the amount corresponding to the route with the smallest weight as the final corresponding amount M;

[0136] 4) The ciphertext equality test scheme usually includes the following six algorithms: Setup, KeyGen, Enc, Dec, Aut, Test, where M and C are the plaintext space and the ciphertext space respectively. Among them:

[0137] Setup(λ) → pp: The initialization algorithm, which takes the security parameter λ as input and outputs the system parameter pp.

[0138] KeyGen(pp) → (pk, sk): The key generation algorithm, which takes the system parameter pp as input and outputs the public key and private key (pk, sk).

[0139] Enc(m, pk) → c: The encryption algorithm, which takes the plaintext m ∈ M to be encrypted and the public key pk as input and outputs the ciphertext c.

[0140] Dec(c, sk) → m: The decryption algorithm, which takes the ciphertext c ∈ C to be decrypted and the private key sk as input and outputs the plaintext m.

[0141] Aut(sk,.) → td: The authorization generation algorithm, which takes the private key sk, etc. as input and outputs the authorization td. Test(c, t, c′, td′) → result: The ciphertext equality test algorithm, which takes two groups of ciphertexts and the corresponding authorizations (c, td) and (c′, td′) as input and outputs the matching result: If the

[0142] corresponding plaintexts of c and c′ are the same, then output 1; otherwise output 0.

[0143] In addition, the algorithm consistency is defined.

[0144] Definition 1 (Decryption correctness): For any m ∈ M,

[0145] Definition 2 (Test consistency): For any m and m′ ∈ M, if m ∈ m′, and td = Aut(sk, ·), td′ = Aut(sk′, ·), then Pr[Test(Enc(m, pk), td, Enc(m′, pk′), td′) = 1] = 1.

[0146] Definition 3 (Test reliability): For any m and m′ ∈ M, if m ∈ m′, and td = Aut(sk, ·), td′ = Aut(sk′, ·), then Pr[Test(Enc(m, pk), td, Enc(m′, pk′), td′) = 1] ≤ negl(λ).

[0147] 5) x i corresponds to the ID card information provided by the financial institution, and a i corresponds to the corresponding information represented by each layer in the random forest. After encrypting the node information using the same encryption method, we perform a comparison protocol, where r takes a random number and r is encrypted.

[0148] 1) When E(x i -a i ) r takes 0, it means that the equivalent test result is the node with weight E(w) being 0

[0149] 2) Send the result (∑w i = 0, E(m)) to the financial institution, where the encryption of the weight is done by the financial department

[0150] 6) Algorithm implementation:

[0151] a. If h = g s

[0152] b. After encryption exists, the ciphertext node information

[0153] c. c1:c1' = g r-r′ ,

[0154] Therefore, it can be known that if x = y, then: x i -y i = 0.

[0155] Joint encryption transmission:

[0156] The specific implementation steps of the jointly encrypted amount are as follows:

[0157] 1) Each bank has its own public key A i , and each bank encrypts using the joint public key A shared by all banks and financial institutions;

[0158] 2) The final ciphertext deposit amount is handed over to the financial institution. With the agreement of all banks, the server adds up the ciphertext amounts and sends them to one of the banks;

[0159] 3) Each bank decrypts a part using its own private key and hands it over to the next bank, and finally to the financial institution for the last step of decryption;

[0160] 4) The financial institution decrypts to obtain the total amount and can obtain the total deposit information of someone;

[0161] 5) For the deposits of each bank, confidentiality is achieved. Each bank performs a homomorphic encryption operation on the amount M using the joint public key (each bank and the financial department) to obtain the encrypted amount E(M); 6) We use the SM2 encryption algorithm to encrypt the compared amount, where the encryption public key is provided by the financial institution k1 and each bank provides k2......k n+1 , and by modifying the k value in the SM2 algorithm, the joint public key encryption is achieved. The process is as follows:

[0162] A) The client uses its own private key k1 and performs the operation A1 = [k1]G on it;

[0163] B) Bank n uses its own private key kn+1 and performs the operation A = [kn+1]G on it;

[0164] C) Generate the joint public key A = [k1 + k2 + k3 + kn+1]G

[0165] D) Each bank uses the joint public key (each bank and the financial department) to perform a homomorphic encryption operation on the amount M to obtain the encrypted amount E(M).

[0166] E) After each bank encrypts the amount M using the joint public key with additive homomorphic property, it sends the result to the server.

[0167] F) The server sums up the ciphertexts of each amount and then sends it to one of the banks for partial decryption.

[0168] G) After one bank decrypts it, it sends it to another bank.

[0169] H) After all banks decrypt it and send it to the server, the server performs the final decryption to obtain the total deposit amount of the user. Thus, under the condition that all banks are authorized, the financial institution can obtain the deposit amount information of the person being queried, and each bank does not know the personal information in actual applications.

[0170] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0171] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A privacy protection system for counting user property information, characterized in that: The privacy protection system consists of an identity information encryption module, an internal bank query module, and a joint encryption transmission module; The identity information encryption module is used by financial institutions to homomorphically encrypt identity information based on the SM2 algorithm. The financial institution encrypts the ID number of the user to be queried and sends the result to the bank; The internal bank query module is used by the bank to perform an equivalent test using a random forest model, compare the received user information, record the weight of the matching information, and filter out the deposit amount that matches the user; The joint encryption transmission module is used by each bank to encrypt the deposit amount separately using a joint public key, then the financial institution sums them up, and then each bank decrypts them step by step. Finally, the financial institution decrypts to obtain the total amount.

2. The privacy protection system for counting user property information according to claim 1, characterized in that: The method steps for homomorphically encrypting identity information based on the SM2 algorithm include: Step 1, System establishment: Let A and B be a financial institution and a bank respectively. The financial institution and the bank pre-set the same public parameter p as a large prime number in advance, E is an elliptic curve defined over the finite field F p p, G is a base point of order q of the elliptic curve, and the bit length mlen of the plaintext space M is set to 32 bits; if [k]G represents the k-fold point of the elliptic curve base point G, the cryptographic hash function H: {0, 1} → Z q. x||y represents the concatenation of the byte string or bit string x and y; represents the exclusive OR operation between bit strings; Step 2, Key Generation: User B randomly selects a private key and publishes the public key pk = [sk]G; Step three, encryption: Suppose the message that user A needs to send to user B is m. User A needs to perform the following calculations: Calculate the elliptic curve point S = [h]pk, where h = 1. If S is an infinite point, report an error and exit; Select a random number Calculate c1 = [k]G = (x1, y1), [k]pk = (x2, y2); Calculate c2 = [m]G + [k]pk; c3 = H(x2||m||y2); Output the ciphertext c = (c1, c2, c3); The financial institution encrypts the ID number of the person to be queried in five parts using this algorithm and sends the encryption result to the bank; Step four, decryption: User B receives the ciphertext c = (c1, c2, c3) sent by user A and performs operation decryption.

3. The privacy protection system for statistically counting user property information according to claim 2, characterized in that: The decryption method in step four includes: Step one, verify whether c1 satisfies the elliptic curve equation. If not, report an error and exit; Step two, calculate the elliptic curve point S = [h]c1, where h = 1. If S is an infinite point, report an error and exit; Step three, calculate the elliptic curve point [sk]c1 = (x2, y2), [m]G = c2 - [sk]c1; Step four, decrypt to obtain m[G], and thus use this value for data comparison.

4. A privacy protection system for statistically analyzing user property information according to claim 1, characterized in that: The identity information encryption module includes information that uniquely identifies the user's identity with the ID number, shortening the time used in the subsequent equivalent test process; Specifically, the 18-bit identity information is divided. The division standard is: province, city, year of birth, month and day of birth, and other remaining information; The operation efficiency of each bank to match the identity information of the user to be queried is improved through a hierarchical method; The financial institution encrypts the user's ID number using the homomorphic encryption scheme of the SM2 standard and sends the encrypted result to the bank for comparison to improve the matching efficiency and protect the user's privacy. The bank stratifies the personal information of the deposit users in the same way for subsequent query work.

5. The privacy protection system for counting user property information according to claim 1, characterized in that: The random forest model is set based on the ID number, where each layer of nodes corresponds to the stratified identity information. The bank compares the internal customer information with the received encrypted identity information, records the weight of the matching information, and encrypts the weight using the public key of the financial institution. The amount corresponding to the route with the smallest weight is selected as the query result by traversing the random forest to ensure the confidentiality of the bank's customer information.

6. The privacy protection system for counting user property information according to claim 1, characterized in that: The equivalent value test includes encrypting node information using the same encryption method, taking a random number \(r\), encrypting \(r\), and performing calculations based on the ID card information provided by the financial institution, the information represented by each layer of nodes in the random forest, and the random number. The encryption algorithms involved are all homomorphic encryption schemes of the national cryptography SM2 standard; When E(x i -a i ) r takes 0, it means that the equivalent test result is a node with a weight E(w) of 0, and the result (∑w i = 0, E(m)) is sent to the financial institution; where x i corresponds to the ID card information provided by the financial institution, a i corresponds to the corresponding information represented by each layer in the random forest.

7. A privacy protection system for statistically analyzing user property information according to claim 1, characterized in that: The joint encryption transmission module includes the financial institution and each bank respectively providing private keys, obtaining the independent public keys of the institution and the bank based on the elliptic curve algorithm according to the private keys, and finally obtaining the joint public key of the financial institution and the bank; The combined public key is used by each bank to encrypt the deposit amount; the SM2 encryption algorithm is used to encrypt the compared amount, where the encryption public key is provided by the financial institution as k1 and by each bank as k2......k n+1 , and the combined public key encryption is achieved by modifying the k value in the SM2 algorithm.

8. A privacy protection system for statistically analyzing user property information according to claim 7, characterized in that: The specific steps for the financial law institution to obtain the total deposit amount are as follows: After each bank encrypts the amount results obtained from their respective inquiries using the joint public key with additive homomorphic property, the results are sent to the financial institution respectively. After the financial institution sums up the ciphertexts of each amount, it sends it to one of the banks for partial decryption, and then the other banks decrypt it step by step. The last step is handed over to the financial institution for decryption to obtain the final total amount M.

9. A privacy protection system for statistically counting user property information according to claim 1, characterized in that: This privacy protection system also includes an efficiency optimization method for unbalanced data volumes. Specifically, by using a five-layer hierarchical method for the node information stored within the bank and combining it with the random forest model for equivalent value testing, unnecessary computational amounts are reduced, and the query efficiency is improved in scenarios where the financial institution queries a single user while the bank has a vast amount of user data, ensuring the response speed and performance of the system under high load conditions.

10. A privacy protection system for counting user property information according to claim 1, characterized in that: This privacy protection system also includes an enhanced security feature implementation method. Specifically, in the identity information encryption stage, a modified SM2 algorithm with homomorphic properties is used, combined with the digest generated by the cryptographic hash function for ciphertext verification, enhancing the security and integrity of the ciphertext; during the joint public key encryption process, the generation of the multi-party joint public key is achieved by modifying the parameters in the SM2 algorithm, ensuring the security of the encryption process and preventing information leakage.