A distributed simulation method and system supporting privacy protection and threshold authorization

Abstract

This invention discloses a distributed simulation method and system supporting privacy protection and threshold authorization. The method includes: generating a threshold encryption key pair, containing a public key and a corresponding private key share; securely distributing the private key share to authorized decryption nodes; each participating node loading its own private simulation input data; each participating node collaboratively executing a secure multi-party computation protocol, maintaining the privacy of each party's input data during the computation process; encrypting the simulation result using the threshold encryption public key to generate encrypted result ciphertext, storing the ciphertext result in a ciphertext result storage unit; and when a user initiates a decryption request, a threshold decryption service module sends a collaboration notification to the authorized decryption party, and reconstructs the plaintext simulation result based on the collaboration of the authorized decryption party, sending it to the user. This invention integrates privacy computation with result access control and achieves distributed key management through a threshold mechanism, effectively reducing the risk of single-point concentration of decryption keys.

Description

Technical Field

[0001] This invention relates to digital simulation technology, and in particular to a distributed simulation method and system that supports privacy protection and threshold authorization. Background Technology

[0002] With the widespread application of digital simulation technology in highly sensitive fields such as emergency command, smart cities, and financial risk control, the demand for multi-organizational collaborative distributed simulations is growing. In such scenarios, each participant typically needs to provide private data involving their core business or confidential information as simulation input, such as COVID-19 infection data or business performance indicators. Simultaneously, the simulation output is often highly sensitive, accessible only to specific authorized personnel under compliance conditions.

[0003] To protect the privacy of input data from participating parties, secure multi-party computation (MPC) technology has been introduced into simulation systems. This technology allows multiple parties to collaboratively compute a pre-defined function without revealing their individual original data, thus achieving "data usable but not visible." Currently, various MPC protocols and frameworks support privacy-preserving computation of complex logic, meeting the computational needs of most simulation models.

[0004] On the other hand, to control access permissions to simulation results, threshold mechanisms in cryptography are used for key management. Threshold cryptography divides a key into multiple shares and distributes them to different authorized entities, stipulating that only when a preset number of share holders (i.e., the threshold value) cooperate can critical operations such as decryption or signing be completed. This approach avoids the security risks associated with a single entity centrally storing the key, improving the system's resilience and the flexibility of access control.

[0005] However, in practical applications, current distributed simulation systems typically use MPC for privacy-preserving computation, while delegating result access control to the application layer or a centralized permission system, resulting in a lack of deep integration between the two. Although some systems attempt to encrypt simulation results, the decryption key is still managed by a single node, failing to achieve truly decentralized authorized access to the results. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a distributed simulation method and system that supports privacy protection and threshold authorization. It can integrate privacy computing with result access control and realize distributed key management through a threshold mechanism, effectively reducing the risk of single-point concentration of decryption keys.

[0007] The objective of this invention is achieved through the following technical solution: a distributed simulation method supporting privacy protection and threshold authorization, comprising the following steps:

[0008] Step S1. Generate a threshold encryption key pair, which includes a public key and a corresponding private key share, and securely distribute the private key share to authorized decryption nodes;

[0009] Step S2. Each participating node loads its own private simulation input data, and the input data does not leave the local node;

[0010] Step S3. Each participating node collaboratively executes a secure multi-party computation protocol to calculate the results of a preset simulation function, while keeping the input data of each party private during the computation process;

[0011] Step S4. Encrypt the simulation results using the threshold encryption public key to generate encrypted result ciphertext, and store the result ciphertext in the ciphertext result storage unit;

[0012] Step S5. The user initiates a decryption request. The threshold decryption service module sends a collaboration notification to the authorized decryption party and, based on the collaboration of the authorized decryption party, reconstructs the plaintext simulation result and sends it to the user.

[0013] A distributed simulation system supporting privacy protection and threshold authorization includes a threshold key management module, an MPC simulation computing engine, an encrypted output module, a ciphertext result storage unit, a threshold decryption service module, and multiple participating node nodes. , , …, P n and multiple authorized decryption nodes , , …, D n ;

[0014] The participating node is responsible for providing private simulation input and calling the MPC simulation calculation engine to perform simulation calculations.

[0015] The authorized decryption node holds a share of the private key for subsequent joint decryption;

[0016] The threshold key management module generates threshold encryption key pairs during the system initialization phase and securely distributes the private key share to authorized decryption parties.

[0017] The MPC simulation computing engine performs simulation function calculations under privacy protection.

[0018] The encrypted output module is used to encrypt the simulation function calculation results using a threshold public key;

[0019] The encrypted result storage unit is used to store the encrypted result;

[0020] The threshold decryption service module is used to coordinate authorized nodes to complete joint decryption.

[0021] The beneficial effects of this invention are: (1) Deeply coupling the privacy computing capability of MPC with the access control capability of threshold cryptography to form an end-to-end security process; (2) Through role separation design, the responsibilities of data providers (participants) and result authorizers (authorized decryption parties) are decoupled to avoid conflicts of interest; (3) An asynchronous threshold decryption mechanism is introduced to adapt to the weak network conditions commonly found in distributed simulations; (4) A lightweight threshold homomorphic encryption scheme is selected to control computation and communication overhead while ensuring functionality. Attached Figure Description

[0022] Figure 1 This is a flowchart of the method of the present invention;

[0023] Figure 2 This is a schematic diagram of the system principle of the present invention. Detailed Implementation

[0024] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0025] like Figure 1 As shown, a distributed simulation method supporting privacy protection and threshold authorization includes the following steps:

[0026] Step S1. Generate a threshold encryption key pair, which includes a public key and a corresponding private key share, and securely distribute the private key share to authorized decryption nodes;

[0027] During system initialization, the threshold key management module generates a public-key encryption scheme that supports threshold decryption (preferably the threshold Paillier encryption scheme, as it supports additive homomorphism and is suitable for most simulation functions such as summation, averaging, and linear combinations). This scheme includes a public encryption key pk and a private key consisting of m shares. ,satisfy The threshold property states that at least t shares are required to reconstruct the private key and complete decryption; if fewer than t shares are required, no information about the private key can be obtained. Private key shares are distributed to m authorized decryption nodes through a secure channel, while the public key pk is made public to all participants.

[0028] Step S2. Each participating node loads its own private simulation input data, and the input data does not leave the local node;

[0029] Each participating node Its private simulation input data Loaded locally, without transmitting in plaintext to any other node.

[0030] Step S3. Each participating node collaboratively executes a secure multi-party computation protocol to calculate the results of a preset simulation function, while keeping the input data of each party private during the computation process;

[0031] All participants jointly invoke the MPC simulation engine to execute preset simulation functions. The engine is based on secure multi-party computation protocols (such as SPDZ or ABY-like protocols), ensuring that no participant can deduce the input of other parties throughout the computation process. It can only obtain the encrypted form of the final function output y or the intermediate ciphertext state.

[0032] Step S4. Encrypt the simulation results using the threshold encryption public key to generate encrypted result ciphertext, and store the result ciphertext in the ciphertext result storage unit;

[0033] After the MPC calculation is completed, the designated output proxy node (which can be any participant or an independent coordinating node) calls the encryption output module to encrypt the plaintext result y using the aforementioned public key pk, generating ciphertext. The ciphertext 'c' is written to the ciphertext result storage unit, which can be deployed locally, on an edge node, or in the cloud for subsequent authorized access.

[0034] Step S5. The user initiates a decryption request. The threshold decryption service module sends a collaboration notification to the authorized decryption party and, based on the collaboration of the authorized decryption party, reconstructs the plaintext simulation result and sends it to the user.

[0035] When plaintext simulation results are needed, a user with the required business permissions initiates a decryption request. The threshold decryption service module receives this request and broadcasts a decryption collaboration notification to authorized decryption party nodes. Upon receiving the notification, use the private key share held by them. A partial decryption operation is performed on the ciphertext c to generate a partial decryption share, which is then sent to the threshold decryption service module over the network. This invention specifically supports an asynchronous submission mechanism: the authorizing party does not need to be online simultaneously, and the system can cache received shares until the cumulative number reaches the threshold value t. This mechanism significantly improves the system's availability in weak network environments such as high latency and intermittent connections.

[0036] Once at least t valid decryption shares are collected, the threshold decryption service module executes a threshold decryption protocol (such as a reconstruction algorithm based on Shamir secret sharing) to combine these shares and recover the plaintext result y, which is then returned to the requester. Throughout the process, the plaintext y only appears in the final output stage after decryption and is only visible to entities that have completed the authorized collaboration.

[0037] like Figure 2 As shown, a distributed simulation system supporting privacy protection and threshold authorization includes a threshold key management module, an MPC simulation computing engine, an encryption output module, a ciphertext result storage unit, a threshold decryption service module, and multiple participating node nodes. , , …, P n and multiple authorized decryption nodes , , …, D n ;

[0038] The participating node is responsible for providing private simulation input and calling the MPC simulation calculation engine to perform simulation calculations.

[0039] The authorized decryption node holds a share of the private key for subsequent joint decryption;

[0040] The threshold key management module generates threshold encryption key pairs during the system initialization phase and securely distributes the private key share to authorized decryption parties.

[0041] The MPC simulation computing engine performs simulation function calculations under privacy protection.

[0042] The encrypted output module is used to encrypt the simulation function calculation results using a threshold public key;

[0043] The encrypted result storage unit is used to store the encrypted result;

[0044] The threshold decryption service module is used to coordinate authorized nodes to complete joint decryption.

[0045] In summary, this invention effectively achieves result access control: through a threshold decryption mechanism, simulation results can only be obtained when ≥t authorized parties collaborate, preventing unauthorized access. For example, under a 3 / 5 threshold, any two authorized parties cannot decrypt, improving security by 100% (compared to single-key schemes). It eliminates single-point-of-failure risks: distributed storage of private keys eliminates single-point-of-failure; even if one authorized party's device is compromised, the system remains secure (as long as the number of compromised devices < t). It improves availability in weak networks: the asynchronous decryption design allows the system to complete decryption even when node offline rates reach 40% (experiments based on NS-3 simulations, t=3, m=5), increasing the protocol success rate from 62% in traditional synchronous schemes to over 95%. It achieves end-to-end privacy protection: input is protected by MPC, and output is controlled by thresholds, satisfying the dual requirements of "data minimization" and "access control." It also keeps computational overhead manageable: threshold Paillier encryption only increases MPC online computation time by about 15% (tested on Intel i7, 16GB RAM), which is within an acceptable range.

[0046] The foregoing description illustrates and describes a preferred embodiment of the present invention. However, as previously stated, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the inventive concept described herein through the foregoing teachings or techniques or knowledge in related fields. Any modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A distributed simulation method supporting privacy protection and threshold authorization, characterized in that: Includes the following steps: Step S1. Generate a threshold encryption key pair, which includes a public key and a corresponding private key share, and securely distribute the private key share to authorized decryption nodes; Step S2. Each participating node loads its own private simulation input data, and the input data does not leave the local node; Step S2 includes: Each participating node Its private simulation input data Loaded locally, without transmitting in plaintext to any other node, where n represents the number of participating nodes; Step S3. Each participating node collaboratively executes a secure multi-party computation protocol to calculate the results of a preset simulation function, while keeping the input data of each party private during the computation process; All participating nodes jointly invoke the MPC simulation engine to execute preset simulation functions. The MPC simulation computing engine is implemented based on a secure multi-party computation protocol; Step S4. Encrypt the simulation results using the threshold encryption public key to generate encrypted result ciphertext, and store the result ciphertext in the ciphertext result storage unit; Step S5. The user initiates a decryption request. The threshold decryption service module sends a collaboration notification to the authorized decryption party and, based on the collaboration of the authorized decryption party, reconstructs the plaintext simulation result and sends it to the user. Step S5 includes: When plaintext simulation results are needed, a user with the necessary business permissions initiates a decryption request. The threshold decryption service module receives decryption requests and broadcasts decryption collaboration notifications to authorized decryption nodes. Each authorized decryption node Upon receiving the notification, use the private key share held by them. Perform partial decryption on the ciphertext c, generate a partial decryption value, and send this portion to the threshold decryption service module over the network; The threshold decryption service module caches the received shares until the cumulative number reaches the threshold value t. Once at least t valid partial decryption shares are collected, the threshold decryption service module executes the threshold decryption protocol, combines the received shares, recovers the plaintext result y, and returns it to the requester.

2. The distributed simulation method supporting privacy protection and threshold authorization according to claim 1, characterized in that: Step S1 includes: S101. During the system initialization phase, the threshold key management module generates a public-key encryption scheme that supports threshold decryption. This scheme includes a public key pk and a private key consisting of m shares. ,satisfy The threshold property states that any number of shares not less than t can reconstruct the private key and complete decryption; if fewer than t are used, no information about the private key can be obtained. S102. Distribute the private key share to m authorized decryption nodes through a secure channel, while the public key pk is made public to all participants.

3. The distributed simulation method supporting privacy protection and threshold authorization according to claim 1, characterized in that: Step S4 includes: The encrypted output module is invoked through the specified output proxy node, and the plaintext result y is encrypted using the public key pk to generate ciphertext. ; The specified output proxy node can be any participating node or a preset independent coordination node; Write the ciphertext c into the ciphertext result storage unit.

4. A distributed simulation system supporting privacy protection and threshold authorization, employing the method described in any one of claims 1 to 3, characterized in that: It includes a threshold key management module, an MPC simulation calculation engine, an encryption output module, a ciphertext result storage unit, a threshold decryption service module, and multiple participating node nodes. , , …, P n and multiple authorized decryption nodes , , …, D n ; The participating node is responsible for providing private simulation input and calling the MPC simulation calculation engine to perform simulation calculations. The authorized decryption node holds a share of the private key for subsequent joint decryption; The threshold key management module generates threshold encryption key pairs during the system initialization phase and securely distributes the private key share to authorized decryption parties. The MPC simulation computing engine performs simulation function calculations under privacy protection. The encrypted output module is used to encrypt the simulation function calculation results using a threshold public key; The encrypted result storage unit is used to store the encrypted result; The threshold decryption service module is used to coordinate authorized nodes to complete joint decryption.

Images