A pairless identity-traceable ring signature cryptography method and system for VANET
By using elliptic curve Diffie-Hellman key negotiation and partial private key mechanism, combined with AOS-type ring signature, the problem of high computation and communication overhead in VANET is solved, realizing low-latency, high-efficiency anonymous communication and identity traceability, which is suitable for secure communication in intelligent connected vehicles.
Patent Information
- Application Number
- CN202611094331.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-22
- Publication Date
- 2026-11-10
AI Technical Summary
Existing ring signature schemes have high computational and communication overhead in VANET, making them difficult to adapt to the requirements of high-frequency message transmission, dynamic ring formation, and low-latency processing. At the same time, key management and identity tracking mechanisms are difficult to achieve anonymous communication of legitimate vehicles and traceability of malicious vehicles while avoiding key escrow.
It employs elliptic curve Diffie-Hellman key negotiation, authentication encryption, partial private key mechanism, and AOS-type ring signature, combined with an unpaired identity-traceable ring signature method. By using elliptic curve scalar multiplication and point-addition hashing operations, it reduces computational complexity and eliminates key escrow risks through a partial private key generation mechanism, thereby achieving anonymous communication and identity traceability.
It reduces onboard computing overhead, eliminates key escrow security risks, achieves conditional privacy protection and traceability, and enhances security and reliability, making it suitable for secure communication in intelligent connected vehicles.
Smart Images

Figure CN123074722A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to, but is not limited to, the fields of vehicle network communication security, cryptographic authentication and privacy protection technology, and particularly relates to a pairwise identity-traceable ring signature method and system for VANET. Background Technology
[0002] With the development of intelligent connected vehicles and intelligent transportation systems, vehicular ad hoc networks have become an important communication method for information exchange between vehicles and between vehicles and roadside infrastructure. Vehicles periodically transmit location, speed, braking status, road events, and collaborative perception information through their onboard units, providing data support for collision warnings, traffic management, and autonomous driving decisions. Because these messages are characterized by high-frequency transmission, high timeliness, and dynamic network topology changes, their security and processing efficiency are directly related to road traffic safety.
[0003] VANET primarily uses open wireless channels for communication, making messages vulnerable to attacks such as eavesdropping, tampering, forgery, replay attacks, and identity impersonation during transmission. Furthermore, if a vehicle consistently uses a fixed identity to send messages, external attackers could potentially obtain the vehicle's driving trajectory and behavioral characteristics by correlating continuous communication records. Therefore, VANET communication must simultaneously guarantee message confidentiality, integrity, source authenticity, and vehicle identity privacy.
[0004] Ring signature cryptography combines encryption, authentication, and anonymous signing into a single cryptographic process, enabling the receiver to verify that a message was generated by a legitimate vehicle within the ring, while preventing direct identification of the actual sender. Some existing solutions employ identity-based cryptography to reduce the overhead of traditional certificate management, but these typically rely on bilinear pairing to bind identities, keys, and ring members. Bilinear pairing is computationally complex and can easily increase the computational burden and communication latency of the onboard unit in scenarios with dense vehicle traffic, frequent message transmission, or a large number of ring members.
[0005] Furthermore, existing identity-based schemes typically rely on a key generation center to generate the vehicle's complete private key, which poses a risk of key escrow. If the key generation center is attacked or its permissions are abused, attackers could impersonate legitimate vehicles to generate valid signatures. Meanwhile, normal vehicle communication needs to remain anonymous, but after a vehicle publishes false traffic information or commits malicious acts, traffic management agencies need to trace the true sender. Existing schemes struggle to simultaneously ensure identity privacy, autonomous key control, and accountability.
[0006] Based on the above analysis, existing technologies urgently need to address two major issues: First, the computational and communication overhead of existing ring signature schemes is high, making it difficult to meet the actual needs of high-frequency message transmission, dynamic ring formation, and low-latency processing in VANETs; second, existing key management and identity tracking mechanisms cannot achieve anonymous communication for legitimate vehicles and traceability of malicious vehicle identities while avoiding key escrow, thus limiting their practical application in secure communication in vehicle networks. Summary of the Invention
[0007] To address the problems of high computational overhead in bilinear pairing, complete private key possession by the key generation center, difficulty in balancing anonymity and traceability, and insufficient real-time performance in dynamic vehicular networks (VANETs) in existing technologies, this invention provides a pairless, identity-traceable ring signature method for VANETs. This method combines elliptic curve Diffie-Hellman key negotiation, authentication encryption, partial private key mechanisms, AOS-type ring signatures, and dedicated tracking tags into a unified signature structure. This allows vehicles to complete secure authentication communication without exposing their true sender identity, and enables the traffic management center to perform identity tracing based on the master key only in the event of a malicious incident.
[0008] This invention is implemented as follows: a pairless identity-traceable ring signature method for VANET, comprising:
[0009] Step 1, System initialization;
[0010] The key generation center selects an elliptic curve cyclic group and a generator to determine the master key and the system public key, wherein the system public key is the scalar product of the master key and the generator; and configures the identity hash function, the key derivation function, the ring challenge hash function, and the authentication encryption algorithm.
[0011] Step 2, vehicle key generation;
[0012] The key generation center calculates a hash value using the identity hash function based on the vehicle's identity identifier, generates a partial private key, and sends it to the vehicle. The vehicle autonomously selects a secret value and forms a complete private key based on the secret value and the partial private key. The vehicle generates an encryption public key and a ring verification public key. The encryption public key is the scalar product of the secret value and the generator. The ring verification public key is obtained by adding the group element obtained by multiplying the hash value of the identity identifier and the generator by a scalar to the encryption public key.
[0013] Step 3, generate signature;
[0014] The sending vehicle determines the ring member set, selects a random number, and calculates a temporary public key, which is the scalar product of the random number and the generator. The sending vehicle performs a dot product of the random number and the receiver's encryption public key to obtain a shared secret. The sending vehicle performs a dot product of its own secret value and the temporary public key to obtain a tracking anchor, and performs a dot product of its partial private key and the tracking anchor to obtain a tracking tag. The symmetric key derived from the shared secret is used to authenticate and encrypt the message to obtain ciphertext. The sending vehicle uses its ring private key and the ring verification public keys of each member in the ring member set to generate a ring signature for the message, which includes the temporary public key, the tracking anchor, the tracking tag, and the ciphertext, and outputs the ring signature cryptography. The sending vehicle's ring private key is determined based on the sum of the hash value of the vehicle's identity identifier and its secret value.
[0015] Step 4, decrypt the token;
[0016] The receiver uses its own secret value to perform a dot product with the temporary public key to recover the shared secret, uses the recovered shared secret to decrypt and authenticate the ciphertext, and verifies the ring signature;
[0017] Step 5, Identity Tracing;
[0018] When tracing, the key generation center calculates a partial private key for each member in the ring member set using the master key and the member's identity identifier. The calculated partial private key is then multiplied by the tracking anchor. If a multiplication result matches the tracking tag, the member's identity is determined to be the real sender.
[0019] Furthermore, in the vehicle key generation step, after the vehicle receives the partial private key, it verifies whether the dot product of the partial private key and the generator is equal to the dot product of the hash value of the identity identifier and the system public key. Only after the verification is successful is the partial private key saved and a secret value selected.
[0020] Furthermore, before using the vehicle public key, it is verified whether the ring verification public key is equal to the hash value of the identity identifier multiplied by the scalar of the generator plus the encryption public key, and the ring verification public key is accepted only if the equation is true.
[0021] Furthermore, the step of generating the signature to determine the ring member set includes: removing vehicles that have been revoked, expired, or failed the ring verification public key consistency verification from the nearby active vehicle directory published by the roadside unit; selecting several vehicles according to the communication range, road direction, message type, and preset ring size, and forming the ring member set with themselves; and arranging the ring members in order.
[0022] Furthermore, the generation of the ring signature includes: using the ring private key of the sending vehicle as the signing private key, for each other member in the ring member set other than the sending vehicle, selecting a random number and calculating an intermediate commitment value using the ring verification public key of that member, and calculating the challenge value of the next member based on the ring challenge hash function and the intermediate commitment value, thus forming a challenge chain; the sending vehicle calculates a response value based on its own corresponding challenge value and ring private key, so that the challenge chain is closed at both ends; the signature message is composed of the ring member set, temporary public key, tracking anchor, tracking tag, ciphertext, recipient identity, timestamp, and protocol context through unambiguous encoding.
[0023] Furthermore, the decryption step also includes: checking whether the timestamp in the received ring signature is within a preset time window, checking whether the random number in the ring signature already exists in the replay buffer, verifying the validity period of the ring member public key and the consistency of the ring verification public key; and accepting the message only when the checks, decryption authentication and ring signature verification are all passed.
[0024] Furthermore, the identity tracing step also includes: verifying the validity of the signature of the ring to be traced before tracing, and combining the identity of the real sender, the hash of the signature of the ring to be traced, the tracing request number, the approval information and the timestamp to generate an auditable tracing record after successful tracing.
[0025] Furthermore, in the ring signature generation step, the sending vehicle mixes an unrelated extension term based on the session random factor into the ring private key each time it signs a signature, so that the ring signatures generated by the same sending vehicle in different signature sessions satisfy computational unlinkability, thereby preventing external attackers from performing behavioral profiling or trajectory tracking on the sending vehicle by associating the ring signature features in different signatures.
[0026] Furthermore, it also includes a forward-secure key update step: a portion of the vehicle's private key and autonomous secret value are unidirectionally evolved and updated according to a preset time period or event triggering conditions. The updated key and the historical key satisfy the unidirectional irreversible property. Even if the key material of the current period is leaked, attackers cannot use the current key to decrypt or forge the signature of the historical period, thereby ensuring the forward security of historical communication content.
[0027] Furthermore, in the identity tracing step, the system master key The t-of-n threshold partitioning method is used to distribute the data across multiple independent regulatory nodes. When identity tracing is required, at least t regulatory nodes collaborate to participate in the calculation. Each node generates a partial tracing result based on its own threshold share, and the results are combined to restore the complete tracing capability. After tracing is completed, a verifiable tracing credential containing the tracing request number, approval information, and timestamp is generated to prevent a single management node from abusing its tracing authority or from making internal erroneous tracing decisions.
[0028] Furthermore, in the designing and verification step, when the receiver needs to verify multiple ring signatures in batches, a signature aggregation verification algorithm is adopted to merge the scalar multiplications in the ring verification of multiple ring signatures in batches. Fixed window sliding scalar multiplication optimization is adopted to significantly reduce the average computational cost of a single batch signature verification, which is suitable for the processing needs of receiving a large number of BSM security messages in a short time under high-density traffic flow scenarios.
[0029] Furthermore, the system initialization step is also configured with post-quantum secure ring signature primitives or post-quantum secure key encapsulation mechanisms as alternative cryptographic components; when the system detects an increase in the level of quantum computing threat, it can smoothly migrate to a lattice-based ring signature scheme or an encoding-based key encapsulation scheme to replace or extend the elliptic curve component, thereby enabling the system to have long-term security against quantum attacks.
[0030] Another objective of this invention is to provide a pairless identity-traceable ring signature system for VANET, comprising a key generation center, a roadside unit, and a vehicle-mounted unit;
[0031] The key generation center is used to select elliptic curve cyclic groups and generators, determine the master key and system public key, configure identity hash function, key derivation function, ring challenge hash function and authentication encryption algorithm; issue partial private keys based on identity identifiers to vehicle units; and, during tracing, use the master key and the tracking anchor, tracking tag and ring member identity identifier in the ring signature to determine the real sender.
[0032] The roadside unit is used to broadcast system parameters, maintain and publish the public key directory of active vehicles, and report abnormal signatures.
[0033] The vehicle-mounted unit includes:
[0034] The key generation module is used to select a secret value and generate an encryption public key and a ring verification public key. The ring verification public key is obtained by multiplying the hash value of the vehicle identification identifier by the generator to obtain a group element, and then adding it to the encryption public key.
[0035] The ring signature generation module is used to determine the ring member set, generate a temporary public key, calculate a tracking anchor and a tracking tag, calculate a shared secret based on the receiver's encryption public key and encrypt the message, and generate a ring signature using the sending vehicle's ring private key and the ring verification public keys of each member in the ring member set, and output the ring signature. The tracking anchor is the dot product of the sending vehicle's secret value and the temporary public key; the tracking tag is the dot product of the sending vehicle's partial private key and the tracking anchor; and the ring private key is determined based on the sum of the hash value of the sending vehicle's identity identifier and its secret value.
[0036] The deseal module is used to recover the shared secret by multiplying its own secret value with the temporary public key, decrypting and authenticating the message, and verifying the ring signature.
[0037] Furthermore, the master key of the key generation center is stored in a hardware security module, or it is divided into multiple management nodes using a threshold method, and identity tracing is performed collaboratively after a preset approval threshold is reached.
[0038] Furthermore, the vehicle unit also includes a secure storage module, wherein the secret value is stored in a trusted execution environment, a TPM, or an automotive-grade security chip.
[0039] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:
[0040] This invention reduces onboard computing overhead: the solution does not perform bilinear pairing, but only uses elliptic curve scalar multiplication, point addition, hashing and symmetric authentication encryption, making it suitable for deployment in onboard units with limited computing power, power consumption and communication resources.
[0041] This invention eliminates key escrow security risks: This invention uses a partial private key generation mechanism to split the user's private key into a partial private key generated by KGC. With user-selected secret value It consists of two parts. The KGC only issues a portion of the user's private key, while the user independently chooses the secret value. This ensures that even if the KGC is completely compromised, attackers cannot impersonate honest users and forge valid signatures. Signature generation requires the use of a portion of the private key. and secret value To generate ring signature response value and calculate tracking anchor points An attacker who only possesses a portion of the private key and does not know the secret value cannot generate a legitimate signature. This design achieves effective separation of key escrow and signing authority, eliminating the risk of a single point of failure in the KGC (Key Control Center) in traditional identity-based cryptography.
[0042] This invention achieves conditional privacy protection and traceability. This scheme uses the AOS ring signature mechanism to achieve unconditional anonymity for the sender; no external adversary or even the receiver can determine the true sender's identity from a legally generated signature. Simultaneously, a portion of the sender's private key is embedded in the signature. With secret value Traceable labels This tag can only be used by the holder of the master key. KGC uses the Trace algorithm for unique verification and traceability. In the event of malicious behavior, KGC can uniquely identify the sender using its master key and thereby execute certificate revocation or legal action, achieving an effective balance between vehicle privacy protection and traffic management accountability.
[0043] This invention enhances security and reliability. Under the random oracle model, this scheme can be proven to satisfy multiple security properties: confidentiality is reduced to the elliptic curve computational Diffie-Hellman problem (ECCDHP), achieving the IND-CCA2 security level; unforgeability and traceability are reduced to the elliptic curve discrete logarithm problem (ECDLP), achieving the EUF-CMA security level; anonymity is directly guaranteed by the unconditional anonymity of AOS ring signatures; and non-framing ensures that even a malicious KGC cannot incorrectly attribute the signature to an honest user. Using AES-256-GCM certified encryption instead of XOR operations makes the security reduction more direct and transparent. The scheme simultaneously satisfies a complete set of security properties at the 128-bit security level, including confidentiality, unforgeability, anonymity, traceability, non-framing, and key escrow resistance.
[0044] (1) The expected benefits and commercial value of the technical solution of this invention after transformation are as follows:
[0045] This solution can be integrated into in-vehicle communication security chips or OBU firmware, providing secure communication capabilities for both the pre-installed and aftermarket intelligent connected vehicles. With the advancement of mandatory V2X communication regulations globally, the V2X secure communication market is projected to exceed tens of billions of US dollars by 2030. The solution's characteristics of requiring no bilinear pairing and high computational efficiency allow for direct deployment on existing in-vehicle hardware platforms, eliminating the need for additional security coprocessors and resulting in extremely low incremental licensing costs per OBU. The conditional privacy protection capabilities provided by the solution meet the core requirements of GDPR, China's Personal Information Protection Law, and other privacy regulations regarding the anonymization of connected vehicle data, providing compliant technical assurance for connected vehicle operators and OEMs.
[0046] (2) The technical solution of the present invention solves a technical problem that people have long wanted to solve but have never been able to solve successfully:
[0047] This invention addresses the coexisting problems of excessively high bilinear pairing computation overhead and key escrow security vulnerabilities in existing ring signaturecryption schemes. It proposes a solution combining elliptic curve-based no-pairing construction with a partial private key generation mechanism. Traditional schemes, in order to achieve identity-based anonymous authentication, generally rely on bilinear pairing operations, becoming a performance bottleneck for onboard unit (OBU) deployments. Simultaneously, the key custodian (KGC) holds the user's complete private key, introducing key escrow security risks. This solution completely eliminates bilinear pairing, downgrading all cryptographic operations to efficient elliptic curve scalar multiplication, achieving an efficient authentication and encryption mechanism in ring signaturecryption that requires no pairing operations. Through the partial private key generation mechanism, the KGC's role is downgraded from a full key custodian to a partial key service provider, fundamentally eliminating key escrow security risks.
[0048] (3) The technical solution of the present invention overcomes technical bias:
[0049] This invention splits the user's private key into two parts: a partial private key issued by the KGC and a user-selected secret value. This ensures that even if the KGC is completely compromised, it cannot forge any legitimate signature from an honest user. Simultaneously, the scheme retains the KGC's exclusive ability to trace the source of malicious messages. The KGC can uniquely identify the sender of the signature using its master key, and no other entity, including ring members themselves, can perform tracing or forge traceability tags to frame others. This design achieves a delicate balance between eliminating key escrow risks and maintaining traceability, providing a reliable security foundation for real-world deployment environments in the Internet of Vehicles (IoV) characterized by multiple domains, multiple parties, and a lack of a unified trust anchor.
[0050] The pairwise identity-traceable ring signature method for VANET proposed in this invention achieves a good balance between security, efficiency, and privacy protection, and is suitable for vehicular communication scenarios that require sender anonymity and traceability of malicious behavior. It can be applied to the construction of prototype systems for specific applications such as V2V secure communication, V2I message authentication, traffic accident liability tracing, and malicious vehicle revocation. Attached Figure Description
[0051] Figure 1 A flowchart of a pairless identity-traceable ring signature method for VANET provided in an embodiment of the present invention;
[0052] Figure 2 A detailed flowchart of the unpaired identity-traceable ring signature method for VANET provided in this embodiment of the invention;
[0053] Figure 3 A flowchart of the dynamic ring-building steps provided in this embodiment of the invention;
[0054] Figure 4 This is a structural diagram of a pairless identity-traceable ring signature system for VANET provided in an embodiment of the present invention;
[0055] Figure 5 This is a schematic diagram illustrating an application scenario of the VANET-based unpaired identity traceable ring signature system provided in an embodiment of the present invention.
[0056] Figure 6 This is a schematic diagram of a pairwise identity-traceable ring signature cryptosystem model for VANET provided in an embodiment of the present invention;
[0057] Figure 7 A dynamic ring size distribution diagram for five types of VANET scenarios;
[0058] Figure 8This is a comparison curve of the signature generation time of the present invention and four comparative schemes under different numbers of ring members;
[0059] Figure 9 This is a comparison curve of the decryption time of the present invention and four comparative schemes under different numbers of ring members;
[0060] Figure 10 This is a comparison curve of the tracking operation time of the present invention and four comparative schemes under different numbers of ring members;
[0061] Figure 11 This is a comparison chart of the total processing time of the proposed solution and four comparative solutions in five real-world VANET scenarios. Detailed Implementation
[0062] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0063] like Figure 1 As shown, this embodiment of the invention provides a pairless identity-traceable ring signature cryptography method for VANET, including:
[0064] S101, System initialization;
[0065] The key generation center selects an elliptic curve cyclic group and a generator to determine the master key and the system public key, wherein the system public key is the scalar product of the master key and the generator; and configures the identity hash function, the key derivation function, the ring challenge hash function, and the authentication encryption algorithm.
[0066] S102, Vehicle key generation;
[0067] The key generation center calculates a hash value using the identity hash function based on the vehicle's identity identifier, generates a partial private key, and sends it to the vehicle. The vehicle autonomously selects a secret value and forms a complete private key based on the secret value and the partial private key. The vehicle generates an encryption public key and a ring verification public key. The encryption public key is the scalar product of the secret value and the generator. The ring verification public key is obtained by adding the group element obtained by multiplying the hash value of the identity identifier and the generator by a scalar to the encryption public key.
[0068] S103, signature generation;
[0069] The sending vehicle determines the ring member set, selects a random number, and calculates a temporary public key, which is the scalar product of the random number and the generator. The sending vehicle performs a dot product of the random number and the receiver's encryption public key to obtain a shared secret. The sending vehicle performs a dot product of its own secret value and the temporary public key to obtain a tracking anchor, and performs a dot product of its partial private key and the tracking anchor to obtain a tracking tag. The symmetric key derived from the shared secret is used to authenticate and encrypt the message to obtain ciphertext. The sending vehicle uses its ring private key and the ring verification public keys of each member in the ring member set to generate a ring signature for the message, which includes the temporary public key, the tracking anchor, the tracking tag, and the ciphertext, and outputs the ring signature cryptography. The sending vehicle's ring private key is determined based on the sum of the hash value of the vehicle's identity identifier and its secret value.
[0070] S104, Deciphering the secret;
[0071] The receiver uses its own secret value to perform a dot product with the temporary public key to recover the shared secret, uses the recovered shared secret to decrypt and authenticate the ciphertext, and verifies the ring signature;
[0072] S105, Identity Tracing;
[0073] When tracing, the key generation center calculates a partial private key for each member in the ring member set using the master key and the member's identity identifier. The calculated partial private key is then multiplied by the tracking anchor. If a multiplication result matches the tracking tag, the member's identity is determined to be the real sender.
[0074] In the vehicle key generation step, after the vehicle receives the partial private key, it verifies whether the dot product of the partial private key and the generator is equal to the dot product of the hash value of the identity identifier and the system public key. Only after the verification is successful is the partial private key saved and a secret value selected.
[0075] This invention provides a method to verify, before using the vehicle public key, whether the ring verification public key is equal to the hash value of the identity identifier multiplied by the scalar of the generator plus the encryption public key, and to accept the ring verification public key only when the equation is true.
[0076] The present invention provides a method for determining the ring member set in the signature generation step, which includes: the sending vehicle removing vehicles that have been revoked, expired, or failed the ring verification public key consistency verification from the nearby active vehicle directory published by the roadside unit; selecting several vehicles according to the communication range, road direction, message type, and preset ring size, and forming the ring member set with itself; and arranging the ring members in order.
[0077] This invention provides a method for generating ring signatures, which includes: using the ring private key of the sending vehicle as the signing private key; for each member in the ring member set other than the sending vehicle, selecting a random number and calculating an intermediate commitment value using the member's ring verification public key; calculating the challenge value of the next member based on the ring challenge hash function and the intermediate commitment value, thus forming a challenge chain; the sending vehicle calculating a response value based on its corresponding challenge value and ring private key, thereby closing the challenge chain; the signing message is composed of the ring member set, temporary public key, tracking anchor, tracking tag, ciphertext, recipient identity, timestamp, and protocol context, all encoded unambiguously.
[0078] The decryption steps provided in this embodiment of the invention further include: checking whether the timestamp in the received ring signature is within a preset time window, checking whether the random number in the ring signature already exists in the replay buffer, verifying the validity period of the ring member public key and the consistency of the ring verification public key; and accepting the message only when the checks, decryption authentication and ring signature verification are all passed.
[0079] The identity tracing steps provided in this embodiment of the invention further include: verifying the validity of the signature of the ring to be traced before tracing, and after successful tracing, combining the identity of the real sender, the hash of the signature of the ring to be traced, the tracing request number, the approval information and the timestamp to generate an auditable tracing record.
[0080] like Figure 2 As shown, this embodiment of the invention provides a pairless identity-traceable ring signature method for VANET, which includes the following steps:
[0081] The system initialization process involves the key generation center selecting an elliptic curve cyclic group. Generator and master key Public key of computing system Configure the identity hash function. Key derivation function Ring challenge hash function and authentication encryption algorithms;
[0082] The vehicle key generation process involves a key generation center targeting the vehicle's identity. calculate and part of the private key Vehicles autonomously select secret values And form a complete private key Calculate the public key for encryption Ring verification public key ;
[0083] The dynamic ring formation step involves sending a vehicle to select multiple vehicles from the active vehicle directory and adding them to form a ring member set. The signature generation process involves sending a random number to the vehicle. ,calculate Sharing secrets Tracking Anchors and tracking tags ,by The obtained symmetric key is used to authenticate and encrypt the message, and a ring signature is generated based on the ring verification public key set and the ring private key of the sending vehicle. The output includes = , , , Ring signature cryptography, including ciphertext and ring signature;
[0084] The decryption process involves the recipient using their own secret value and... The shared secret is restored, authentication and decryption are completed, and ring signature closure is verified; along with the identity tracing step, the key generation center calculates the ring member's identity. and will satisfy The vehicle's identity was confirmed as that of the true sender.
[0085] The vehicle receives a portion of the private key. Afterwards, verification The private key is saved only when verification is successful, and a self-generated secret value is generated. .
[0086] The ring verification public key satisfies Before using the vehicle public key, the roadside unit or vehicle verifies the identity identifier, the consistency between the encrypted public key and the ring verification public key according to this equation.
[0087] like Figure 3 As shown, the dynamic ring formation step includes: receiving a directory of nearby active vehicles published by the roadside unit; removing vehicles that have been revoked, expired, or whose public key consistency verification has failed; selecting ring members according to the communication range, road direction, message type, and preset ring size; and arranging the ring members in order based on session randomness.
[0088] The tracking anchor point and tracking tags Each satisfies and ,in To send the vehicle's autonomous secret value, This is a partial private key issued by the key generation center for the identity of the sending vehicle. This is the temporary public key used for this signature.
[0089] The ring signature is an AOS ring signature, sent by the vehicle with... For the ring private key, compute for non-signing members and And calculate based on the actual sending vehicles This is to ensure that the ring challenge chain is closed at both ends.
[0090] Signature message Set of members of the ring Temporary public key Tracking Anchors Tracking tags Authentication encrypted text The receiver's identity, random number, timestamp, and protocol context are obtained through unambiguous encoding.
[0091] The decryption step further includes: checking whether the timestamp is within a preset time window, checking whether the random number already exists in the replay cache, and verifying the validity and consistency of the public keys of each ring member; and accepting the message only when the above checks, authentication decryption, and ring signature verification all pass.
[0092] The identity tracing step is executed after verifying the validity of the signature of the ring to be traced, and generates an auditable tracing record by combining the real sender's identity, the signature hash of the ring to be traced, the tracing request number, the approval information, and the timestamp.
[0093] like Figure 4 As shown, this embodiment of the invention provides a pairless identity traceable ring signature system for VANET, including: a key generation center 1, a roadside unit 2, and a vehicle-mounted unit 3;
[0094] The key generation center 1 is used to select the elliptic curve cyclic group and generator, determine the master key and system public key, configure the identity hash function, key derivation function, ring challenge hash function and authentication encryption algorithm; issue a partial private key based on the identity identifier to the vehicle unit; and, in tracing, use the master key and the tracking anchor, tracking tag and ring member identity identifier in the ring signature to determine the real sender.
[0095] The roadside unit 2 is used to broadcast system parameters, maintain and publish the public key directory of active vehicles, and report abnormal signatures.
[0096] The vehicle-mounted unit 3 includes:
[0097] The key generation module is used to select a secret value and generate an encryption public key and a ring verification public key. The ring verification public key is obtained by multiplying the hash value of the vehicle identification identifier by the generator to obtain a group element, and then adding it to the encryption public key.
[0098] The ring signature generation module is used to determine the ring member set, generate a temporary public key, calculate a tracking anchor and a tracking tag, calculate a shared secret based on the receiver's encryption public key and encrypt the message, and generate a ring signature using the sending vehicle's ring private key and the ring verification public keys of each member in the ring member set, and output the ring signature. The tracking anchor is the dot product of the sending vehicle's secret value and the temporary public key; the tracking tag is the dot product of the sending vehicle's partial private key and the tracking anchor; and the ring private key is determined based on the sum of the hash value of the sending vehicle's identity identifier and its secret value.
[0099] The deseal module is used to recover the shared secret by multiplying its own secret value with the temporary public key, decrypting and authenticating the message, and verifying the ring signature.
[0100] The embodiments of the present invention provide that the master key of the key generation center is stored in the hardware security module, or is divided into multiple management nodes using a threshold method, and identity tracing is performed collaboratively after a preset approval threshold is reached.
[0101] The vehicle unit provided in this embodiment of the invention also includes a secure storage module, wherein the secret value is stored in a trusted execution environment, a TPM, or an automotive-grade security chip.
[0102] In the ring signature generation step, the sending vehicle mixes an unrelated extension term based on a session random factor into the ring private key each time it performs a signature, so that the ring signatures generated by the same sending vehicle in different signature sessions satisfy computational unlinkability, thereby preventing external attackers from profiling or tracking the sending vehicle's behavior by associating the ring signature features in different signatures.
[0103] The embodiments of the present invention also provide a forward-secure key update step: a portion of the vehicle's private key and autonomous secret value are unidirectionally evolved and updated according to a preset time period or event triggering conditions, and the updated key and the historical key satisfy the unidirectional irreversible property; even if the key material of the current period is leaked, attackers cannot use the current key to decrypt or forge the signature of the historical period, thereby ensuring the forward security of historical communication content.
[0104] In the identity tracing step provided by this embodiment of the invention, the system master key The t-of-n threshold partitioning method is used to distribute the data across multiple independent regulatory nodes. When identity tracing is required, at least t regulatory nodes collaborate to participate in the calculation. Each node generates a partial tracing result based on its own threshold share, and the results are combined to restore the complete tracing capability. After tracing is completed, a verifiable tracing credential containing the tracing request number, approval information, and timestamp is generated to prevent a single management node from abusing its tracing authority or from making internal erroneous tracing decisions.
[0105] In the above embodiments, when the receiver needs to verify multiple ring signatures in batches during the signature decryption verification step, a signature aggregation verification algorithm is used to batch merge the scalar multiplications in the ring verification of multiple ring signatures. Fixed window sliding scalar multiplication optimization is adopted to significantly reduce the average computational cost of a single batch signature verification, which is suitable for the processing needs of receiving a large number of BSM security messages in a short time under high-density traffic flow scenarios.
[0106] In the system initialization step provided by this embodiment of the invention, a post-quantum-secure ring signature primitive or a post-quantum-secure key encapsulation mechanism is also configured as an alternative cryptographic component. When the system detects an increase in the level of quantum computing threat, it can smoothly migrate to a lattice-based ring signature scheme or an encoding-based key encapsulation scheme, replacing or extending the elliptic curve component, thereby enabling the system to have long-term security against quantum attacks. Table 1: Explanation of Main Symbols
[0107]
[0108] The VANET-oriented unpaired identity-traceable ring signature method provided in this invention includes six steps: system initialization, vehicle registration and key generation, dynamic ring construction, ring signature generation, designing and ring verification, and conditional identity tracing.
[0109] S1, System Initialization
[0110] KGC input security parameters Perform the following operations:
[0111] (1) Select a safe elliptic curve defined on a finite field and its order as . cyclic subgroups Select generator This ensures that the elliptic curve discrete logarithm problem and the elliptic curve computational Diffie-Hellman problem are infeasible under a predetermined security strength.
[0112] (2) Random selection Used as the system master key, and used to calculate the system master public key. .
[0113] (3) Configuration , and ,in To verify the length of the encryption key.
[0114] (4) Configure authentication encryption algorithm Public system parameters Secretly preserved .
[0115] To reduce the risk of single point of leakage, The data can be segmented using a t-of-n threshold method and stored in multiple hardware security modules. During tracing, the tracing computation capability is jointly restored by the management nodes that have reached the threshold number, but this does not change subsequent operations. The verification relationship.
[0116] S2, Vehicle Registration and Key Generation
[0117] vehicle Upon initial system access, the vehicle interacts with the KGC via a protected registration channel. The vehicle identification identifier can be formed by encoding the vehicle's legal identity, device serial number, registration domain, and validity period; alternatively, it can be a long-term identifier whose mapping relationship is only known to the traffic management department. The specific steps are as follows:
[0118] (1) KGC verifies vehicle registration documents and calculates... as well as and will Send to the vehicle via a secure channel.
[0119] (2) Vehicle verification If this equation is true, it means that a portion of the private key matches the current identity and the system's master public key; if it is false, the vehicle registration is terminated.
[0120] (3) Vehicles are randomly selected locally. To form a complete private key . It is not sent to KGC and is protected by onboard security hardware.
[0121] (4) Vehicle calculation and .because Any node can be identified by its identity, Check the consistency of the public key with system parameters.
[0122] (5) Vehicles publish to TMC or public key directory The RSU periodically obtains catalog increments and broadcasts them to vehicles within its coverage area.
[0123] S3, Dynamic Ring Construction
[0124] Before each signature, the sending vehicle constructs a ring based on the current communication environment. Preferably, the active vehicle entries broadcast by the RSU include identity identifier, public key digest, validity period, most recent online time, and management domain. The sending vehicle first removes revoked, expired, or public key consistency check failed entries, and then sorts them according to communication radius, road direction, message type, and preset ring size. choose The other members, together with themselves, form a ring. .
[0125] To reduce long-term associations, the order of ring members can be pseudo-randomized based on session randomness and time windows; the same vehicle can select different ring members in different messages. For high-density road scenarios, a maximum ring size can be set. To limit latency; for low-density scenarios, vehicles or trusted RSU public keys that are still valid within the most recent time window can be used as supplementary members. Ring construction does not require the online participation of other members, nor does it require obtaining the private keys of other members.
[0126] S4, Ring Sign-in Generation
[0127] Send vehicle Hold the complete private key Target recipient The public key for encryption is Send vehicle pair message Perform the following operations:
[0128] (1) Session key generation. Randomly selected. Calculate the temporary public key and sharing secrets .Will With communication context enter To obtain the symmetric key The communication context may include the protocol version, transmission time slot, receiver identifier, and algorithm identifier.
[0129] (2) Authentication encryption. Generate a unique random number. and timestamp ,Will As additional authentication data, calculation The authenticated encrypted output includes a ciphertext body and an authentication tag.
[0130] (3) Tracking tag generation. Calculation as well as . The vehicle's autonomous secret value will be bound to a random variable in this session. Further bind a portion of the private key issued by KGC for the identity.
[0131] (4) Construction of the signature message. After obtaining... and Then update the additional authentication data and construct... Encode is an unambiguous encoding function that includes the type, length, and order of each field, avoiding ambiguity caused by concatenation.
[0132] (5) AOS Ring Signature. Let the index of the actual sending vehicle in the ring be s, and its ring private key be... ,and The sending vehicle is randomly selected. And calculate ,make Then, each non-signature member in the ring is randomly selected in circular order. ,calculate and After looping back to the member before the actual sending vehicle, calculate... This restores the actual sending vehicle location. .
[0133] (6) Signature output. Output and encapsulate The vehicle will be dispatched. Send it in the payload of a V2V or V2I message.
[0134] In one implementation, first calculate and This is then added to the authentication encryption data, so that the encrypted authentication tag is also bound to the tracking field; in another implementation, authentication encryption is performed first, and then uniformly bound by the ring signature. , and Both methods should ensure that desigma verification fails if any field is modified.
[0135] S5, Decryption and Ring Verification
[0136] The recipient received Next, check the protocol version, ring size, field length, and timestamp. and random numbers .like Exceeding the allowed time window, or If the file already exists in the replay buffer of the same sending domain, it will be rejected directly. Then, the following steps will be executed:
[0137] (1) Calculation and derived According to the agreed AD call recover If the authentication label verification fails, an invalid identifier will be output.
[0138] (2) According to Retrieve members from the local directory or RSU. Verify member validity and public key consistency, and construct a key identical to the sender's. .
[0139] (3) Analysis ,make .right to Calculate in sequence and and order If ultimately Then the ring signature is closed.
[0140] (4) Output a message only when authentication and decryption are successful, ring signature is closed, replay protection check passes, and all public keys are valid. Accept; otherwise, delete the temporary data and output an invalid flag.
[0141] S6. Conditional Identity Tracing
[0142] When the receiving vehicle, RSU, or independent anomaly detection module detects false location, falsified braking, malicious congestion information, or cooperative perception deception, it saves the original signature, its reception time, detection evidence, and the signature of the reporting node, and initiates a traceability request to the TMC. The KGC executes the following after completing authorization and approval:
[0143] (1) Verify the ring signature and message structure of the signature to be traced and parse it. , and ,examine It is not an infinite point and belongs to the correct subgroup.
[0144] (2) Calculate for each IDi in u and .
[0145] (3) Inspection If the equation is true, output... And trace the evidence summary; if multiple members simultaneously satisfy the equation, the input is judged to be abnormal and enters manual audit; if no member satisfies the equation, the output cannot be traced.
[0146] (4) The traceability results can be combined with message hash, request number, authorization approval information and timestamp to form an audit record, which can be used for certificate revocation, reputation update, accident investigation or judicial evidence collection.
[0147] Evidence related to the technical effects obtained by the embodiments of the present invention:
[0148] like Figure 5 , Figure 6To verify the technical effectiveness of this invention, a comprehensive performance evaluation was conducted on a hardware platform based on SECP256k1 elliptic curve cryptography and AES-256-GCM authentication encryption, running Windows 11, an AMD Ryzen 7 5800H CPU with Radeon Graphics CPU @ 3.20GHz, and 16GB of memory. The experiment used the libsecp256k1 library to measure the ECC operation time, with each operation repeated 5000 times and the average taken. The pairing operation time was taken from the BLS12-381 benchmark data of the RELIC / mcl library on equivalent x86-64 hardware. A single ECC scalar multiplication took approximately 0.0482ms, SHA-256 took approximately 0.0005ms, AES-256-GCM encryption / decryption of a 300B message took approximately 0.0016 / 0.0017ms respectively, and BLS12-381 pairing operation took approximately 1.50ms. The present invention scheme was compared with four representative schemes: Cai et al. scheme
[20] (IB-RSC), Du et al. scheme
[22] (improved CP-RSC), Wei et al. scheme
[23] (IDTRSC framework) and Zhan et al. scheme
[29] (multi-mode CL-RSC). All schemes were implemented at a unified 128-bit security level and uniformly adopted AES-256-GCM authentication encryption.
[0149] The experiment was conducted using three real-world vehicle trajectory datasets: LuST, Tapascologne, and VeReMi. Five VANET scenarios were constructed, generating a total of 3000 communication events. Ring members were dynamically selected based on neighboring vehicles within a 300m communication radius, with a maximum ring size of n_max=50. The five scenarios covered diverse traffic conditions: LuST Urban (mean ring member value 13.5, maximum 43); LuST Suburban (mean ring member value 12.7, maximum 32); Tapascologne (mean ring member value 18.4, maximum 50); VeReMi Highway (mean ring member value 14.7, maximum 44); and VeReMi Misbehavior (mixed traffic including 5% malicious nodes, mean ring member value 8.9, maximum 29).
[0150] Figure 7This figure shows the dynamic ring size distribution across five VANET scenarios. It illustrates the distribution of naturally formed ring member numbers within a 300m communication radius, based on three real-world vehicle trajectory datasets: LuST, Tapascologne, and VeReMi. The horizontal axis represents the number of ring members (n), and the vertical axis represents the frequency of occurrence. The figure reveals that in the LuST Urban morning rush hour scenario, ring sizes are concentrated in the range of 6–22 members; in the Tapascologne high-density urban scenario, ring sizes can reach a maximum of 50; in the VeReMi Highway scenario, ring sizes range from 3–44; and in the VeReMi Misbehavior scenario, due to the presence of malicious nodes resulting in fewer effective neighbors, ring sizes are concentrated in the range of 3–15. This figure visually reflects the dynamic changes in ring size in real-world traffic environments, validating the adaptive ring formation capability of this proposed solution under different traffic densities.
[0151] Figure 8 The graph shows the comparison of signature generation time between the present invention and four comparative schemes under different ring member numbers. The horizontal axis represents the number of ring members n, and the vertical axis represents the signature generation time. The vertical axis uses a logarithmic scale to simultaneously show schemes of different magnitudes. The graph shows two unpaired ECC curves of the present invention and CL-RSC
[29] , and three paired curves of IBRSC
[20] , IBRSC
[22] , and IDTRSC
[23] based on BLS12-381. It can be seen from the graph that the signature time of the three paired schemes increases superlinearly with n; the signature time of the present invention increases approximately linearly with n, only 1.21ms when n=3 and only 19.48ms when n=50, which is about 10.0 times faster than scheme
[23] . Scheme
[29] has a lower signature time due to the use of unbound tracking tag construction, but the additional signature overhead of the present invention is entirely used to generate bound OR proofs in exchange for key security upgrades such as non-fragmentation and O(1) publicly verifiable tracking.
[0152] Figure 9 The graph shows the decryption time comparison between the present invention's scheme and four comparative schemes under different ring member numbers. The horizontal axis represents the number of ring members n, and the vertical axis represents the decryption time. The decryption time trend is basically symmetrical with the decryption time: the present invention's scheme requires only 19.48ms when n=50, scheme
[23] reaches 211.00ms, and schemes
[20] and
[22] reach 104.43ms and 106.05ms respectively. The present invention's scheme and scheme
[29] are both unpaired constructions, requiring only elliptic curve scalar multiplication and hash functions in decryption, without the need for pairing verification. The low latency of the receiver's decryption is quite important for the real-time processing of a large number of BSM messages in VANET.
[0153] Figure 10The graph shows the comparison of the tracking operation time of the present invention scheme and four comparative schemes under different ring member numbers. The horizontal axis represents the number of ring members n, and the vertical axis represents the tracking operation time. The present invention scheme uses ElGamal public ciphertext points for binding tracking. The tracking operation only requires one scalar multiplication and one dot subtraction, followed by a constant-time table lookup in the public key directory. The total tracking time is always 0.05ms, which is completely independent of the number of ring members n. In contrast, the tracking of scheme
[29] requires O(n) ECC scalar multiplications, while the tracking of the three pairing-based schemes requires O(n) pairing operations. The tracking characteristics of the present invention scheme have a good efficiency advantage in the forensic scenario where thousands of messages need to be traced in batches after an incident.
[0154] Figure 11 The figure shows a comparison of the total processing time of the present invention and four comparative schemes in five real VANET scenarios. The horizontal axis represents the five scenarios: LuST Urban, LuST Suburban, TAPAS Cologne, VeReMi Highway, and VeReMi Misbehavior, and the vertical axis represents the total processing time. As can be seen from the figure, the total processing time of the present invention in all five scenarios is much lower than that of the three paired schemes: in the high-density urban scenario of TAPAS Cologne, the present invention takes only 14.42ms, while scheme
[23] takes 153.24ms, a difference of about 10.6 times; in the lightweight VeReMi Misbehavior scenario, the present invention takes only 7.00ms, while scheme
[23] still takes 76.80ms. Scheme
[29] has a simpler construction and a total processing time in the range of 3~5ms, but it pays the price of lacking non-fragmentability. The figure intuitively shows the stable performance of the unpaired construction in different real traffic scenarios.
[0155] Application Area 1: V2V Emergency Braking Warning. After detecting an emergency braking event, the sending vehicle selects vehicles traveling in the same direction from the list of nearby active vehicles to form a loop. The location, speed, braking intensity, and timestamp are encoded into message 'm', and the message is then decrypted using the public key of the target following vehicle. After the following vehicle completes the decryption, it can confirm that the message originated from a nearby legitimate vehicle but cannot identify the specific vehicle. If the warning is proven to be maliciously forged, TMC can trace the true sender.
[0156] Application Area 2: V2I Road Accident Reporting. Accident vehicles send the accident location, sensor summary, and image hash to the RSU. The RSU, as the receiver, performs desigma verification to confirm the message's integrity and that the sender belongs to the current set of legitimate vehicles on the road. For legitimate reports, the RSU does not obtain the true identity; for intentional false alarms, the RSU submits the desigma and evidence, and the TMC performs identity tracing.
[0157] Application Area 3: Dynamic Ring Formation in High-Density Urban Roads. In densely trafficked areas, RSU broadcasts a valid public key digest every time window. Sending vehicles randomly select members based on the communication radius and the maximum ring size nmax, preventing the ring size from growing indefinitely. Scalar multiplication in ring signature verification can be computed in parallel on a member-by-member basis, and RSU caches Wᵢ to reduce directory lookup overhead.
[0158] Application Area 4: Cross-Management Domain V2I Communication. When a vehicle moves from one management domain to another, the target RSU verifies the domain signature and validity period of the vehicle's public key directory, bringing vehicles from different domains that conform to the mutual trust policy into the ring. Multiple KGCs can each maintain a portion of their own domain's private keys and traceability permissions. A domain field is added to the signature and incorporated into the M and AD fields to prevent cross-domain context substitution.
[0159] Application Area 5: KGC Master Key Leakage Scenario. Even if an attacker obtains... It can only calculate the individual vehicles. Unable to obtain vehicle local The attacker cannot compute a legitimate loop response. Required It is also unable to generate a secret value consistent with the target vehicle. Therefore, it is impossible to impersonate a vehicle to generate a signature that simultaneously passes loop verification and traceability consistency checks. The system can be updated... and reissue The master key rotation is completed, and the vehicle can retain or update synchronously. .
[0160] Application Area Six: Multi-party Authorization and Traceability. To prevent a single administrator from abusing traceability permissions, a master key... A threshold-based approach is employed, requiring at least t nodes from the incident investigation, cybersecurity, and privacy compliance nodes to approve traceability requests. All nodes collaborate to complete this process. The inspection ultimately only outputs the identity result linked to the case number, and does not disclose it to ordinary verification nodes. or all candidates .
[0161] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.
[0162] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A lightweight, fault-tolerant private sample alignment method for low-altitude wireless networks, characterized in that, include: The steps for establishing a cryptographic system are as follows: multiple client nodes initialize the system with the central server, generate global public parameters based on a ring fault-tolerant learning encryption scheme accelerated by number theory transformation, and use the Fiat-Shamir verifiable secret sharing mechanism to split the private keys of each client node into multiple key shares and distribute them to other nodes to establish a cryptographic system that supports threshold decryption. Encrypted upload steps: Each client node processes its local private sample set in batches, performs differential noise-based encoding on each batch of samples, and uses the ring fault-tolerant learning encryption scheme to encrypt the encoded data, generating encrypted samples which are then uploaded to the central server. Aggregation and decryption steps: The central server aggregates encrypted samples from each client, solicits key shares of no less than a preset threshold value from the currently online client nodes to reconstruct the aggregated key, performs decryption and noise filtering, and extracts candidate information from the intersection of samples and returns it to the client nodes; The differential noise-based encoding operation is as follows: for each sample identifier in the sample set, if the sample exists, it is encoded as a random number from a first numerical range; if the sample does not exist, it is encoded as a random number from a second numerical range. The interval between the first numerical range and the second numerical range is greater than the noise amplitude of the ring fault-tolerant learning encryption system, so that after the encoded data of all client nodes are aggregated and decrypted, the values corresponding to the intersection samples fall into a distinguishable statistical range.
2. The method according to claim 1, characterized in that, The Fiat-Shamir verifiable secret sharing mechanism specifically includes: each client node distributes its private key to other nodes by randomly splitting it into a number of shares equal to the total number of client nodes using a polynomial, and broadcasts a commitment to the polynomial coefficients; the receiving nodes verify the validity of the received shares non-interactively based on the commitment, thus forming a threshold cryptosystem, wherein the minimum number of shares required for decryption is the threshold value, which is less than the total number of client nodes.
3. The method according to claim 1, characterized in that, Before the cryptographic system establishment step, a threshold judgment step is also performed: the size of the sample intersection is probabilistically estimated based on the Goldwasser-Sipser protocol, and the cryptographic system establishment step, the encryption upload step, and the aggregation decryption step are triggered only when the estimated intersection size exceeds a preset threshold.
4. The method according to claim 3, characterized in that, The threshold determination step specifically includes: The output length of the hash function is determined based on a preset intersection size threshold. Multiple independent verifications are performed. In each round, the server randomly selects a hash function and a target value and broadcasts them to all client nodes. Each client node calculates a subset of samples that satisfy the hash collision on its local machine, encrypts them, and returns them. The server aggregates and decrypts the encrypted results, and counts the number of rounds in which samples exist at the same location across all client nodes; If the number of successful rounds exceeds a preset decision threshold, the intersection size is determined to exceed the intersection size threshold, triggering a complete alignment process.
5. The method according to claim 1, characterized in that, The number theory transformation-accelerated ring fault-tolerant learning encryption system supports homomorphic addition operations on ciphertext. The central server aggregates encrypted samples from each client by performing homomorphic addition operations on the ciphertext domain and aggregating the encrypted samples from each client into a single aggregated ciphertext. In the encoding operation, random numbers from the first or second numerical interval are generated by a lattice-based pseudo-random function based on the ring-tolerant learning hypothesis.
6. A lightweight, fault-tolerant private sample alignment system for low-altitude wireless networks, characterized in that, include: Multiple client nodes are deployed on a mobile platform in a low-altitude wireless network. Each client node holds a local private sample set. The client nodes are used to perform differential noise-based encoding operations and use a ring fault-tolerant learning encryption scheme to encrypt the encoded data and generate encrypted samples. The central server is used to receive encrypted samples uploaded by each client node, aggregate the encrypted samples, solicit key shares of no less than a preset threshold value from the currently online client nodes to reconstruct the aggregated key, and perform decryption and noise filtering to extract candidate information from the intersection of samples. Specifically, the client node and the central server distribute key shares based on the Fiat-Shamir verifiable secret sharing mechanism to establish a cryptographic system that supports threshold decryption; the differential noise-based encoding operation is as follows: for each sample identifier bit in the sample set, if the sample exists, it is encoded as a random number from a first numerical range; if the sample does not exist, it is encoded as a random number from a second numerical range, wherein the interval between the first numerical range and the second numerical range is greater than the noise amplitude of the ring fault-tolerant learning encryption scheme.
7. The lightweight fault-tolerant private sample alignment system for low-altitude wireless networks as described in claim 6, characterized in that, Before performing the aggregation and decryption of encrypted samples, the central server also performs a threshold judgment operation: based on the Goldwasser-Sipser protocol, it probabilistically estimates the size of the sample intersection, and only when the estimated intersection size exceeds a preset threshold is the aggregation and decryption operation of encrypted samples triggered.
8. The lightweight fault-tolerant private sample alignment system for low-altitude wireless networks as described in claim 7, characterized in that, The threshold determination operation specifically includes: The output length l of the hash function is determined based on the preset intersection size threshold T; Multiple independent verifications are performed. In each round, the server randomly selects a hash function and a target value and broadcasts them to all client nodes. Each client node calculates a subset of samples that satisfy the hash collision on its local machine, encrypts them, and returns them. The server aggregates and decrypts the encrypted results, and counts the number of rounds in which samples exist at the same location across all client nodes; If the number of successful rounds exceeds the preset decision threshold, the intersection size is determined to exceed T.
9. The lightweight fault-tolerant private sample alignment system for low-altitude wireless networks as described in claim 6, characterized in that, The communication between the client node and the central server is adapted to the dynamic nature and bandwidth constraints of the low-altitude wireless network. When the low-altitude wireless network link status is lower than the preset quality standard, the client node will upload the encrypted samples in batches.
10. The lightweight fault-tolerant private sample alignment system for low-altitude wireless networks as described in claim 6, characterized in that, The client node integrates an embedded privacy computing module, which includes: The number theory transformation acceleration unit is used to perform multiplication operations on the polynomial ring; A ring-tolerant learning cryptographic coprocessor is used to perform key generation, encryption, decryption, and homomorphic addition operations; The fault-tolerant key management unit is used to securely store and process key shares based on verifiable secret sharing.
