Hardware-assisted certificateless signcryption authentication method and system for cross-domain unmanned aerial vehicle network

Abstract

The application discloses a hardware-assisted certificateless signcryption authentication method and system for cross-domain unmanned aerial vehicle networks, and belongs to the technical field of network space security, unmanned aerial vehicle communication security and password authentication. The application anchors a versioned accumulator root of a consortium chain, and a ground station synchronously forms a local trusted cache to realize non-message-by-message on-chain query member state verification. An unmanned aerial vehicle reconstructs a certificateless complete private key on demand by using a PUF and a TEE, and long-term software storage is avoided. A member witness participates in signcryption binding independently of a private key, and private key structure is not changed during revocation and update. Batch verification is performed by using a random aggregation coefficient to prevent forgery offsetting. The application can reduce authentication delay, resist key leakage after physical capture, keep key structure stable, and improve the reliability of batch verification in a high-concurrency scenario.

Description

Technical Field

[0001] This invention belongs to the fields of cyberspace security, UAV communication security, and cryptographic authentication technology. Specifically, it relates to a hardware-assisted certificateless signature authentication method for cross-domain UAV networks, and more particularly to a cross-domain UAV authentication and secure communication method that combines consortium blockchain state anchoring, dynamic accumulator member verification, and hardware-assisted key reconstruction. Background Technology

[0002] Unmanned aerial vehicle (UAV) networks have been widely applied in scenarios such as military reconnaissance, emergency rescue, border monitoring, patrol inspection, and air-to-ground communication services. In these applications, UAVs typically need to connect to ground stations in different management domains to achieve cross-domain communication, data transmission, and authentication. However, UAVs are characterized by resource constraints, open deployment environments, and dynamically changing network topologies. Their communication processes largely rely on open wireless channels, making them vulnerable to attacks such as eavesdropping, tampering, replay attacks, spoofing, and unauthorized access. Signature encryption technology can simultaneously perform signature and encryption functions in the same logical process, reducing the cost of secure communication for resource-constrained devices to some extent. However, in cross-domain UAV networks, the authentication object includes not only the communication message itself but also the membership status of the sending UAV in the cross-management domain environment. Therefore, ordinary signature encryption mechanisms alone cannot meet the requirements for verifiability and revocable authentication of membership status in cross-domain UAV networks.

[0003] To address these challenges, various cross-domain authentication and revocation management schemes have been proposed in existing technologies. For example, member status can be confirmed through blockchain ledger queries, centralized state servers, or revocation status synchronization mechanisms. However, if each authentication requires access to the blockchain ledger or remote state server, the authentication process will be affected by external query latency, making it difficult to meet the rapid processing requirements of high-concurrency drone access scenarios. If the ground station relies solely on local revocation status, it must ensure the trustworthiness of the local status, version consistency, and the binding relationship between member status and signature messages; otherwise, unauthorized access may occur. Furthermore, existing identity-based cryptographic schemes suffer from key escrow issues, and the method of storing complete private keys in software is insufficient to resist the risk of key leakage after a drone is physically captured. In high-concurrency authentication scenarios, traditional deterministic batch aggregation methods may also be vulnerable to attackers exploiting algebraic cancellation relationships between forged ciphertexts to bypass overall batch verification.

[0004] To address the aforementioned issues, existing research has proposed cross-domain drone anonymous authentication methods based on Physically Unclonable Functions (PUFs), such as the paper "PUF-Based Secure and Efficient Anonymous Authentication Protocol for UAV Towards Cross-Domain Environments". This scheme targets cross-domain drone identity authentication and session key negotiation, utilizing PUF responses to protect key authentication parameters stored locally on the drone. It also reduces authentication bottlenecks, privacy leaks, and the risk of verifier theft by ensuring the control center does not participate in real-time authentication and the ground station does not store drone information. However, this scheme is essentially still an anonymous authentication and key negotiation protocol, not a signature-based mechanism for business messages. Its PUF is primarily used to protect locally stored information; it does not introduce a Trusted Execution Environment (TEE) to reconstruct and use the drone's complete private key on demand within the protected environment, nor does it address the integrated binding of member status, revocation status, and cross-domain signature messages.

[0005] Another existing technology proposes using RSA accumulators for certificateless signature revocation, such as Chinese patent application CN202410420397.3. This type of method embeds the evidence value of elements in the accumulator into a portion of the user's private key and uses broadcast updates of credentials to achieve dynamic revocation for unauthorized users. While this method can achieve user revocation in signature authentication scenarios, its technical logic remains centered on signature authentication and does not provide confidentiality protection for business messages under a cryptographic mechanism. Furthermore, because the accumulator evidence value is embedded in a portion of the user's private key, revocation status updates can easily affect the user's private key or credential update process, making it unsuitable for direct application in high-frequency cross-domain cryptographic communication scenarios involving unmanned aerial vehicles (UAVs).

[0006] In summary, while existing technologies involve various methods such as PUF-assisted cross-domain authentication, accumulator-assisted revocation, blockchain state query, and batch verification, they still struggle to simultaneously address the following key technical issues in cross-domain drone networks: Efficiency and reliability of member status verification are difficult to balance: querying the ledger or remote service for each message results in high authentication latency; relying solely on local status makes it difficult to guarantee the reliability and version consistency of the status.

[0007] Reversing state updates affects private key stability: The existing accumulator reversal mechanism embeds evidence into the private key, which means that reversal updates require frequent changes to the user's private key, making it unsuitable for high-frequency signature and encryption scenarios.

[0008] Key escrow and physical capture risks coexist: Identity-based cryptography suffers from centralized key escrow issues; software storing complete private keys can easily lead to key leakage after physical capture.

[0009] Batch verification carries the risk of forgery cancellation: traditional deterministic batch aggregation methods are easily bypassed by attackers who can exploit the algebraic cancellation relationship between forged ciphertexts to bypass overall verification.

[0010] Therefore, existing technologies have not effectively solved the problems of verifying member state and signature messages in the same authentication process, and revoking state updates without changing the complete private key structure of the UAV. To address this, it is necessary to propose an authentication method for cross-domain UAV networks, enabling ground stations to complete trusted member state verification without accessing the ledger message by message, allowing UAVs to avoid storing complete private keys for extended periods, and achieving reliable signature authentication under high concurrency conditions. Summary of the Invention

[0011] In view of this, the present invention provides a hardware-assisted certificate-free signature authentication method and system for cross-domain unmanned aerial vehicle networks to solve the problems in the background art.

[0012] To achieve the above objectives, the present invention adopts the following technical solution: On the one hand, this invention discloses a hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle (UAV) networks, including the following specific steps: The consortium blockchain ledger is deployed, the private key generation centers of each management domain are registered to the consortium blockchain ledger, and the dynamic accumulator is initialized, and the root of the initial accumulator is anchored to the consortium blockchain ledger; The drone uses a physically unclonable function to generate a hardware-bound secret and calculate its public component within a trusted execution environment. It then submits its pseudo-identity and public component to the private key generation center of the management domain. The private key generation center generates a partial private key for the drone, adds it to a dynamic accumulator, and updates its state to obtain a version number, accumulator root, and member witness. Subsequently, the drone's registration information and versioned accumulator root are anchored to the consortium blockchain, and a partial private key, member witness, and state version number are distributed to the drone. The drone only stores the physically unclonable function challenge value, a partial private key, member witness, and version number, deleting temporary intermediate values. The ground station generates its own secret using a true random number generator, calculates the ground station's public component, and submits the ground station's pseudo-identity and public component to the private key generation center of its management domain; the private key generation center generates a partial private key for the ground station and uploads the ground station's registration information to the consortium blockchain ledger; the ground station calculates the complete private key and stores it in the hardware security module; The private key generation center of each management domain submits the accumulator root and state version number updated due to drone registration or revocation to the consortium blockchain ledger for anchoring; the ground station synchronizes and verifies the latest state from the consortium blockchain ledger to form a trusted local state cache with trusted anchoring. The UAV acquires the public parameters of the target ground station and pre-calculates the target ground station's public key; during the online signing phase, the UAV calls a physically unclonable function within a trusted execution environment to reconstruct the complete private key, uses the reconstructed complete private key to generate a signing authentication value, and outputs a cross-domain signing message; The ground station receives the cross-domain signature message and checks the timestamp; it calls the ground station's complete private key in the hardware security module to decrypt and restore the authentication payload; it verifies the validity of the drone's membership in the current state; after the membership verification is successful, it calculates the drone's public key based on the restored authentication payload and verifies the correctness of the signature equation; if the verification is successful, it accepts the business message.

[0013] This method anchors the accumulator root and state version number in the consortium blockchain, enabling the ground station to directly verify the membership of cross-domain drones based on a local trusted state cache, avoiding the high latency issues caused by accessing the blockchain ledger message by message. Simultaneously, it employs certificateless key generation and a PUF / TEE hardware-assisted mechanism, ensuring that the drone's complete private key is composed of a partial private key and a hardware-derived secret, which can be reconstructed as needed. This eliminates the key custody risks of the private key generation center and prevents the long-term exposure of software keys due to physical capture. The ground station's private key is stored in the HSM, ensuring the security of the signature authentication process. This technical solution achieves efficient, reliable, and physically attack-resistant cross-domain signature authentication.

[0014] Preferably, it also includes: randomized batch verification, whereby the ground station independently generates random aggregation coefficients for each candidate ciphertext that has passed member status verification, and executes the randomized batch verification equation; if the equation is true, the batch is accepted, otherwise invalid ciphertexts are recursively divided, located, and isolated.

[0015] This method introduces randomized batch verification on the candidate ciphertext set after state filtering. By generating random aggregation coefficients independently for each ciphertext, it disrupts the algebraic cancellation relationship between forged ciphertexts pre-constructed by attackers, which significantly improves the reliability of batch authentication in high-concurrency access scenarios. At the same time, it reduces the computational overhead of verification by message through aggregation verification.

[0016] Preferably, the steps for the UAV to reconstruct the complete private key by calling the physically unclonable function within the trusted execution environment are as follows: the physically unclonable function module is called again to obtain the physically unclonable function response; a derived secret is calculated based on the response; and the complete private key is reconstructed by combining the stored partial private key with preset algebraic operations; the complete private key is deleted from the trusted execution environment immediately after the signing is completed.

[0017] In this method, the drone only calls the PUF to reconstruct the complete private key within the TEE at the signing time, and deletes it immediately after use. The complete private key is not stored in non-volatile memory in software form for a long time. Combined with the unique hardware binding of the PUF device, even if the drone is physically captured, the attacker cannot obtain the usable complete private key, effectively resisting physical capture attacks and offline key extraction, and improving the terminal's resistance to damage.

[0018] Preferably, the cross-domain signature message is generated as follows: the business message, the drone pseudo-identity, the drone public component, the management domain identifier, the member witness, and the status version number are concatenated in a preset order to form an authentication payload; the pre-calculated temporary public value and the authentication payload are used to input the signature binding hash; the drone uses the reconstructed complete private key within the trusted execution environment to generate a signature authentication value and outputs the cross-domain signature message.

[0019] This method concatenates the business message, pseudo-identity, public component, domain identifier, member witness, and status version number into an authentication payload and participates in the signature-cryptography binding hash calculation, making the member status information inseparable from the signature-cryptography message. Attackers cannot replace, delete, or replay the member witness and version number without breaking the signature-cryptography verification equation, thus ensuring the strong binding relationship between the member status and the message content and preventing revocation bypass attacks.

[0020] Preferably, the specific steps for the ground station to verify the validity of the drone's membership in the current state are as follows: If the state version number carried in the ciphertext is equal to the current state version number of the ground station, then directly verify whether the member witness corresponds to the current accumulator root; If the state version number carried in the ciphertext is less than the current state version number of the ground station, then the accumulator leaf of the UAV is determined based on the verified auxiliary update information to see if it has been revoked between the two versions. If it has not been revoked, then it is confirmed that it still has valid membership. If the state version number carried in the ciphertext is greater than the current state version number of the ground station, the ground station will temporarily suspend processing the ciphertext and perform state synchronization.

[0021] Based on the difference between the state version number carried in the ciphertext and the current version of the ground station, this method adopts a processing strategy of direct verification, auxiliary update information judgment, or delayed synchronization, which enables the ground station to correctly process the old or new version member witnesses without querying the blockchain message by message. For old version drones that have not been revoked, it can assist them in updating to the current version member witnesses, ensuring the continuity and state consistency of cross-domain communication.

[0022] Preferably, the specific steps for verifying the correctness of the signature equation are as follows: The ground station calculates the drone's public key based on the recovered drone pseudo-identity, the drone's public components, and the identifier of its management domain, combined with the domain public key of the management domain. Input the recovered business message, drone identity information, member witness, status version number, temporary public value obtained from designing, and timestamp into the hash function to recalculate the hash value of the signature binding; Verify whether the temporary public value recovered from the decryption is equal to the product of the decryption authentication value and the elliptic curve generator, minus the product of the decryption binding hash value and the UAV public key.

[0023] This method ensures that business messages, identity information, member witnesses, status version numbers, and timestamps have not been tampered with by recovering the drone's public key, recalculating the signature binding hash, and verifying the signature equation on the elliptic curve. When the equation holds true, it proves the authenticity, integrity, and non-repudiation of the message, while implicitly confirming the member witness and identity binding, thus achieving the integration of signature verification and member status verification.

[0024] Preferably, the randomized batch verification step is performed as follows: Suppose that the candidate ciphertext set contains ciphertexts that have passed member status verification. For each candidate ciphertext, the ground station calculates its corresponding temporary public value, signature authentication value, signature binding hash value, and drone public key. A true random number generator independently generates random aggregation coefficients for each candidate ciphertext, and the random aggregation coefficients are uniformly selected from a set of positive integers of a preset size. Verify the weighted aggregation equation: Is the sum of the products of the random aggregation coefficients of all candidate ciphertexts and their temporary public values ​​equal to the sum of the products of the random aggregation coefficients of all candidate ciphertexts and their signature authentication values ​​multiplied by the elliptic curve generator, minus the sum of the triple products of the random aggregation coefficients of all candidate ciphertexts, the signature binding hash value, and the drone public key? If the equation is true, the batch is accepted; otherwise, invalid ciphertext is located and isolated using a recursive partitioning method. The isolated invalid ciphertext is not included in the set of valid messages.

[0025] This method generates random aggregation coefficients independently for each ciphertext in the candidate ciphertext set that has passed member state verification, and verifies the weighted aggregation equation. Since the aggregation coefficients are truly random and independent, attackers cannot pre-construct forged ciphertext pairs that can cancel each other out under fixed aggregation. When verification fails, invalid ciphertexts are located and isolated by recursive partitioning, which ensures both the high efficiency of batch processing and avoids the security risk of forged ciphertexts bypassing batch verification.

[0026] Preferably, the consortium blockchain ledger is used to anchor globally public parameters, public parameters of each management domain, entity pseudo-identity registration records, and versioned accumulator roots, and does not participate in the real-time authentication of each signature message; the accumulator root in the local trusted state cache of the ground station must be consistent with the accumulator root of the corresponding version on the chain, and member witnesses must be able to verify the accumulator root.

[0027] This method explicitly positions the consortium blockchain as a low-frequency trusted state anchor point, used only to anchor the accumulator root and version number, and does not participate in the real-time authentication of each signature message; the local cache of the ground station must be strictly consistent with the corresponding version accumulator root on the chain, and member witnesses must be able to verify the root, thereby reducing authentication latency while ensuring the trustworthiness and version consistency of the local state and preventing historical state replay attacks.

[0028] On the other hand, this invention discloses a hardware-assisted certificate-free signature authentication system for cross-domain unmanned aerial vehicle (UAV) networks, which, when applied to the above method, is characterized by including: The decentralized trust layer includes a consortium blockchain ledger and private key generation centers for multiple management domains; the consortium blockchain ledger is used to anchor global public parameters, public parameters of each management domain, entity pseudo-identity registration records, and versioned accumulator roots; the private key generation centers are responsible for the registration of drones and ground stations within their respective management domains, partial private key generation, drone member status updates, and revocation status maintenance. The edge authentication layer includes a ground station, a trusted state cache, a batch authentication module, a true random number generator, and a hardware security module. The ground station is used to receive cross-domain signature messages and perform designing and member state verification. The trusted state cache is used to store local states anchored to the consortium blockchain. The batch authentication module is used to perform randomized batch verification. The hardware security module is used to protect the ground station's long-term private key. The drone layer includes a drone, a physically unclonable function module, a trusted execution environment, a member witness, and a pre-computed tuple; the physically unclonable function module is used to generate hardware secrets bound to the drone device; the trusted execution environment is used to reconstruct the drone's complete private key on demand and perform signature cryptography in a protected execution environment; the member witness is used to prove that the drone has valid membership in the corresponding state version.

[0029] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a hardware-assisted certificate-free signature authentication method and system for cross-domain drone networks. Compared with existing cross-domain drone authentication methods, the present invention has the following beneficial effects: 1. This invention reduces the reliance on message-by-message on-chain queries for cross-domain member state verification and improves the reliability of member state verification. Existing methods easily rely on blockchain ledger queries or remote state services when confirming drone membership and revocation status, making the authentication path susceptible to external query latency. This invention anchors the versioned accumulator root in the consortium blockchain ledger, and the ground station obtains and verifies the local trusted state cache during the state synchronization phase. In the single-signal message authentication phase, the ground station completes local verification based on the verified local accumulator state and the member witness carried by the drone, without needing to access the consortium blockchain ledger for each authentication message. Testing shows that the average latency for reading queries from the consortium blockchain ledger is approximately [missing information]. When the effective member size At that time, member witness verification only requires This is significantly lower than the latency of on-chain queries per message. Therefore, this invention can shift the high-frequency authentication process from on-chain queries to local trusted state verification at the ground station, reducing authentication latency.

[0030] 2. This invention enables verifiable revocation state management with minimal on-chain state overhead. Instead of relying on a local revocation list maintained by the ground station, this invention uses a versioned accumulator root as a trusted on-chain anchor, ensuring that the member states cached by the ground station are constrained by the corresponding version accumulator root in the consortium blockchain ledger. The drone carries member witness information and a state version number in its signature message. After restoring the authentication payload, the ground station can determine whether the drone still has valid membership based on the current version or verified auxiliary update information. Testing shows that only the accumulator root and version number need to be maintained on-chain, with a size of [size missing]. ;when At that time, each message carries member witnesses containing The number of hash values ​​introduced, and the computational cost of member state verification, are... The corresponding communication overhead is Therefore, this invention can exchange quantifiable member witness overhead for locally verifiable member state and avoid including consortium blockchain access in every message authentication path.

[0031] 3. This invention maintains the stability of the complete private key structure of the drone while achieving revocable authentication. Existing revocation methods that embed the accumulator evidence value into the user's partial private key can easily cause revocation state updates to affect the user's private key or credential update process. This invention separates the member witness from the certificateless private key structure, allowing it to participate as an independent member state proof in the signature payload and signature binding hash. Therefore, changes in the revocation state only cause updates to the accumulator root, state version number, and member witness, without altering the computational structure of the drone's partial private key, the PUF derived secret, and the on-demand reconstruction of the complete private key within the TEE. Thus, this invention ensures revocation verifiability while avoiding changes to the drone's complete private key structure during revocation updates, reducing key maintenance complexity in cross-domain high-frequency authentication scenarios.

[0032] 4. This invention reduces the risks of key escrow and the risk of complete private key leakage after device capture. In existing identity-based authentication or signature methods, user private keys are usually directly generated by a private key generation center, posing a key escrow risk. This invention employs a certificate-free signature method. The complete private key of the UAV is composed of a partial private key issued by the private key generation center and a hardware-derived secret on the UAV side. The private key generation center cannot obtain the complete private key of the UAV. Simultaneously, the UAV side uses PUF to generate a secret component bound to the device and reconstructs the complete private key as needed within TEE. The complete private key is not permanently stored in non-volatile memory in software form; the relevant temporary secret is deleted after signature completion. The ground station stores and uses the long-term private key through HSM, enabling designification-related private key operations to be completed in a hardware-isolated environment. Therefore, this invention can simultaneously reduce the risks of centralized key escrow and the risk of long-term private key exposure on the terminal side.

[0033] 5. This invention improves the anti-forgery cancellation capability in high-concurrency batch authentication. In traditional deterministic batch verification, attackers may construct forged ciphertexts with algebraic cancellation relationships, causing invalid ciphertexts to cancel verification errors in the overall aggregation equation, thus leading to the acceptance of invalid batches. This invention independently generates random aggregation coefficients for each candidate ciphertext before batch verification, making it impossible for attackers to pre-construct stable cancellation relationships under a fixed aggregation structure. Theoretical analysis shows that if each After the batch is fixed, from the set If each batch is selected independently and uniformly, the probability that a batch containing invalid ciphertext will pass verification is at most 1 / 3. In one implementation, take The upper bound of the false acceptance probability of invalid batches is Tests have shown that the maximum batch size is... In the scenario described above, the randomized batch decryption time of this invention is... Compared to sequential verification baseline, it reduces approximately In the direction Inject the constructed offset ciphertext pair into each valid request and perform... Under conditions of independent trials, the randomized aggregation mechanism of this invention detected all injection pairs. Therefore, this invention can balance batch processing efficiency and batch verification reliability in high-concurrency authentication.

[0034] 6. This invention reduces the cryptographic computation burden on drones during their online phase. By combining offline pre-computation with online signature encryption, the drone can perform time-consuming dot multiplication and addition operations during idle periods and cache pre-computed tuples. This reduces the computational burden during the real-time communication phase. During the online phase, the UAV primarily performs lightweight operations such as PUF calls, private key reconstruction within the TEE, hashing, and XOR operations. Testing showed that the relevant cryptographic operations were implemented on NIST P-256 elliptic curves, running on an ARMv8 platform and Ubuntu 22.04 system, using the MIRACL library and running GCC. O2 compilation, including hash operations According to the calculation process of this invention, the main cryptographic computation overhead of the UAV online signature phase is: This reduces the computational burden during the real-time communication phase of the drone.

[0035] In summary, this invention enables cross-domain drone authentication to complete trusted member state verification without accessing the consortium blockchain ledger message by means of consortium blockchain state anchoring, dynamic accumulator member verification, PUF / TEE-assisted certificateless private key reconstruction, HSM protection of ground station private keys, and randomized batch verification. This improves the reliability of high-concurrency batch authentication while avoiding long-term software storage of complete private keys and changes to the private key structure due to revocation and updates. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0037] Figure 1 The system framework diagram provided for this invention; Figure 2 The flowchart for entity registration, member status update and on-chain anchoring provided by this invention; Figure 3 The flowchart for cross-domain signature authentication and random weighted batch verification provided by this invention is shown. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] This invention discloses a hardware-assisted certificate-free signature authentication system for cross-domain unmanned aerial vehicle (UAV) networks, such as... Figure 1 As shown, the system of the present invention includes a decentralized trust layer, an edge authentication layer, and a drone layer.

[0040] The decentralized trust layer comprises a consortium blockchain ledger and multiple private key generation centers for different management domains, such as Private Key Generation Center A and Private Key Generation Center B. The consortium blockchain ledger anchors globally public parameters, public parameters for each management domain, entity pseudo-identity registration records, and versioned accumulator roots, providing a unified and trusted state basis for cross-domain authentication. Each private key generation center is responsible for the registration of drones and ground stations within its respective management domain, partial private key generation, drone member state updates, and revocation state maintenance. The consortium blockchain does not participate in the real-time processing of every authentication message; instead, it serves as a trusted state anchor point for low-frequency updates, constraining the trustworthiness and version consistency of the ground station's local cached state.

[0041] The edge authentication layer comprises a ground station, a trusted state cache, a batch authentication module, a True Random Number Generator (TRNG), and a Hardware Security Module (HSM). The ground station, acting as a cross-domain access and authentication node, is responsible for receiving signature messages from drones in other domains, restoring the authentication payload, verifying drone member status, and performing single or batch verifications based on the business scenario. The trusted state cache stores the local state anchored to the consortium blockchain, including the current accumulator root, state version number, pseudo-identity registration records, and auxiliary update information. The batch authentication module performs randomized batch verification of multiple candidate ciphertexts in high-concurrency scenarios. The HSM protects the ground station's long-term private key, ensuring that related private key operations are performed in a hardware-isolated environment.

[0042] The drone layer comprises the drone, a PUF module, a Trusted Execution Environment (TEE), member witnesses, pre-computed tuples, and a partial private key on the drone side. The PUF module generates hardware secrets bound to the drone device; the TEE reconstructs the drone's complete private key on demand within the Protected Execution Environment and performs sensitive computations during the signature process; the member witness proves the drone's valid membership in the corresponding state version; and the pre-computed tuples reduce computational burden during real-time communication. The drone does not permanently store the complete private key; it only stores the PUF challenge, a partial private key, member witnesses, and the state version number. The complete private key is reconstructed by the TEE on demand when signature is required and deleted after use.

[0043] In another embodiment, a hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle (UAV) networks is disclosed, with the following specific steps: Step 1, system input security parameters Select the elliptic curve parameters. Let... For prime numbers of order 1 Additive cyclic group, For the group The generator.

[0044] Step S2, the system defines the hash function. and accumulator hash function .in, Used to map PUF responses to secret scalars Used to bind entity pseudo-identities, public components, and management domain identifiers. Used to generate a ciphertext mask. Used to generate signature-key binding hash Used to generate dynamic accumulator leaves.

[0045]

[0046] Step S3, each management domain The private key generation center selects the domain master secret. And calculate the domain public key. .

[0047] Step S4, each management domain will Register with the consortium blockchain ledger to form a globally public set of parameters required for cross-domain authentication.

[0048] Step S5: The system initializes the dynamic accumulator and generates the initial accumulator root. And anchor the initial state to the consortium blockchain ledger. .

[0049] For ease of explanation, the following description uses cross-domain authentication between two management domains as an example. (Drone) Belongs to the management domain Private key generation center Management, ground station Belongs to the management domain Private key generation center Management. The description uses two management domains as an example, but the present invention is not limited to the case of two management domains. When there are multiple management domains, multiple drones and multiple ground stations, each entity completes registration, status synchronization, cross-domain signature and authentication verification in the same way.

[0050] To further optimize the above technical solution, it also includes: drone registration methods, such as... Figure 2As shown; Step S6, Drone Select or receive PUF challenge value And generate a response through the PUF module. Calculate derived secrets within the TEE and calculate the publicly available components of the drone. To the management domain Submit registration information to the private key generation center ,in To create a false identity for the drone.

[0051] Step S7, Private Key Generation Center Calculation According to the domain lord's secret Generate a partial private key for the drone .

[0052] Step S8, Define the drone public key .because The verifier can also calculate using publicly available information. .

[0053]

[0054] Step S9: The private key generation center calculates the leaf node of the accumulator corresponding to the drone. Add it to the dynamic accumulator, update the member set state, and obtain the new state version number. accumulator root Witnessed by members .

[0055] Step S10, the private key generation center will transfer the drone registration information. and versioned accumulator root Submit to the consortium blockchain ledger for anchoring.

[0056] Step S11: The private key generation center sends a portion of the private key, member witness information, and status version number to the drone. The drone stores the PUF challenge, a portion of the private key, member witnesses, and the state version number in non-volatile memory. And delete the PUF response and related temporary intermediate values.

[0057] Furthermore, in this invention, the member witness is not part of the drone's partial private key, nor does it participate in the generation of the drone's complete private key. The member witness is only used to prove that the drone belongs to a valid member set under a specific state version. During the registration phase, the private key generation center issues a partial private key, member witness, and state version number to the drone. The partial private key is used for subsequent reconstruction of the complete private key without a certificate, while the member witness and state version number are used for subsequent member state verification. The two are logically independent in terms of storage and use. Therefore, when a member is revoked or the accumulator state is updated, only the accumulator root, state version number, and related member witness or auxiliary update information need to be updated. It is not necessary to change the drone's partial private key, nor is it necessary to change the complete private key composed of the PUF-derived secret and the partial private key.

[0058] Step S12: When the drone needs to generate a signature message, it calls the PUF module again within the TEE to obtain a response. Calculate derived secrets And reconstruct the complete private key as needed. The complete private key obtained from the reconstruction is only used temporarily within the TEE and is deleted after signing is completed.

[0059] To further optimize the above technical solution, the following is also included: a ground station registration method; Step S13, Ground Station Generate its own secret using a true random number generator (TRNG). And calculate the publicly available components of the ground station. .

[0060] Step S14, the ground station sends a signal to its management domain. The private key generation center submits the ground station's fake identity and the ground station's public components. .

[0061] Step S15: The private key generation center calculates the identity binding hash. And generate a partial private key for the ground station. Ground station registration information Uploaded to the blockchain.

[0062] Step S16, the ground station calculates the complete private key. It is then stored in the HSM, and subsequent decryption-related private key operations are performed under the protection of the HSM.

[0063] Step S17, Define the ground station public key Similarly, the verification method can calculate using publicly available parameters. .

[0064]

[0065] To further optimize the above technical solutions, the following methods are also included: state anchoring and synchronization methods, such as... Figure 2 As shown; In step S18, the private key generation center of each management domain is responsible for the registration, revocation, and status update of drones in its domain. When a drone is registered, the private key generation center adds its accumulator leaf node to the valid member set; when an illegitimate drone is revoked, the private key generation center removes its accumulator leaf node from the valid member set.

[0066] During the aforementioned member state update process, the dynamic accumulator only maintains the member state set of the drone and does not change the drone's certificateless key material. After the accumulator leaf corresponding to the revoked drone is removed from the valid member set, the new accumulator root and state version number are committed to the consortium blockchain ledger. Drones that have not been revoked can obtain new version member witnesses based on auxiliary update information, while their partial private keys, PUF challenges, and the complete private keys reconstructed on demand by the TEE remain unchanged.

[0067] Step S19: After each registration or revocation causes a change in the accumulator state, the private key generation center generates a new state version number and the accumulator root key. Submit to the consortium blockchain ledger.

[0068] In step S20, after the accumulator root in the consortium blockchain ledger is updated due to drone registration or revocation, the ground station synchronizes the latest state version and auxiliary update information, verifying the consistency between the local trusted state cache and the on-chain anchored state. If the accumulator root in the local cache is inconsistent with the corresponding version of the accumulator root on the chain, the ground station performs state synchronization or refuses authentication based on the cached state. During the single-signature message authentication phase, the ground station does not access the consortium blockchain ledger for each message, but instead completes member state verification based on the verified local trusted state cache.

[0069] Step S21: When the ground station receives a drone message carrying the witness of an older version member, if the version number carried in the ciphertext is less than the current version number of the ground station... The ground station determines the corresponding accumulator leaf for the UAV based on the verified local trusted state cache and auxiliary update information. Has it been revoked? If revoked, reject the encrypted message; if not revoked, verify whether the drone still has valid membership in its current state. If so, the ground station calculates the membership information using locally stored accumulator information. Return the encrypted message to the drone for subsequent communication; otherwise, reject the encrypted message.

[0070] Through the above processing, the consortium blockchain ledger only serves the functions of state trust anchoring and version consistency constraints, and does not participate in the real-time authentication of each signature message. The accumulator root in the local trusted state cache of the ground station must be consistent with the accumulator root of the corresponding version on the chain. The member witness must be able to verify the accumulator root, and the version number of the member witness must also be consistent with the version information recovered from the signature message. Therefore, even if an attacker obtains a member witness of a certain historical version, they cannot remove it from the original signature message and use it in a new cross-domain authentication process.

[0071] To further optimize the above technical solutions, it also includes: cross-domain signature methods, such as... Figure 3 As shown; Step S22: The UAV acquires publicly available parameters of the target ground station. Pre-compute the target ground station public key .

[0072] Step S23: The UAV randomly selects a temporary scalar during its idle state. ,calculate and And cache pre-computed tuples To reduce real-time communication latency.

[0073] In step S24, during the online signature stage, the drone calls the PUF module in the TEE and receives a response. And reconstruct the complete private key of the drone. .

[0074] Step S25, the drone reads the current timestamp. Member testimonies and status version number and business messages Drone fake identity Domain Identifier The authentication payload is composed of a pre-calculated temporary public value and input into the signature binding hash. Subsequently, the drone uses the fully reconstructed private key within the TEE to generate the signature authentication value and outputs a cross-domain signature message.

[0075] After the calculation is completed, TEE removes the relevant temporary sensitive intermediate values.

[0076] To further optimize the above technical solution, it also includes: methods for decryption and member status verification, such as... Figure 3 As shown; Step S26: The ground station receives the cross-domain signature message. timestamp Used to detect replayed messages and verify timestamps. Check if it is within the allowed time window; if the timestamp is expired, duplicated, or abnormal, reject the encrypted message.

[0077] Step S27: The ground station retrieves its private key under HSM protection. ,calculate ,recover Parse to obtain business messages False identity Domain Identifier Member testimonies Status version number and timestamp If the authentication payload format is incomplete, or if the member witness, status version number, or management domain identifier is inconsistent with the ciphertext structure, the ciphertext will be rejected.

[0078] Step S28: The ground station verifies the drone member status. Based on the recovered drone pseudo-identity, the drone's public components, and the identifier of its management domain, the ground station calculates the accumulator leaf corresponding to the drone. And verify member witnesses by combining the accumulator root and state version number in the local trusted state cache. If Ground station directly verifies member witnesses Does it correspond to the current accumulator root? If the verification passes, the drone is confirmed to belong to the current set of valid members; otherwise, the ciphertext is rejected. The ground station makes a judgment based on the verified auxiliary update information. Is it in to If the membership is revoked, or if not, the drone is confirmed to still have valid membership, and a current version of the membership witness will be generated or updated based on the verified auxiliary update information for subsequent communication; otherwise, the encrypted message will be rejected. If so, it means that the ground station's local trusted state cache is lagging behind the state version carried in the ciphertext. The ground station will temporarily suspend processing the ciphertext and perform state synchronization.

[0079] Step S29: After the member status verification is successful, the ground station calculates the drone's public key based on the recovered drone pseudo-identity, the drone's public components, and the identifier of its management domain. ,calculate Subsequently, the ground station verified the signature equation. .in, and All of these are the member witness and status version number recovered from this encrypted message. If the equation holds true, it means that the member witness, status version number, cross-domain identity information, and business message are all consistent with this signed message, and the output message is executed. If the equation is not true, then the ciphertext is rejected.

[0080] To further optimize the above technical solution, it also includes: a randomized batch verification method, such as... Figure 3 As shown; In high-concurrency access scenarios of drone swarms, the ground station can perform randomized batch verification on multiple candidate ciphertexts that have passed member status verification.

[0081] In step S30, the ground station performs timestamp checks, demasking, and membership status verification on the received ciphertexts to obtain a set of candidate ciphertexts. .

[0082] Step S31, for each candidate ciphertext The ground station calculates the corresponding And a random aggregation coefficient is independently generated for each candidate ciphertext using a true random number generator (TRNG). In an embodiment of the present invention, take Subsequently, the ground station conducted randomized batch verification. If the above equation holds true, the ciphertext in the candidate set passes the batch verification, and the ground station outputs the valid message set; if the above equation does not hold true, it means that there is invalid ciphertext in the candidate set. The ground station uses a recursive partitioning method to locate and isolate invalid ciphertext. The isolated invalid ciphertext does not enter the valid message set; the verified subset continues to output the corresponding message.

[0083] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0084] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle (UAV) networks, characterized in that, The specific steps include the following: The consortium blockchain ledger is deployed, the private key generation centers of each management domain are registered to the consortium blockchain ledger, and the dynamic accumulator is initialized, and the root of the initial accumulator is anchored to the consortium blockchain ledger; The drone uses a physically unclonable function to generate a hardware-bound secret and calculate the drone's public component in a trusted execution environment. It then submits the drone's pseudo-identity and public component to the private key generation center of the management domain. The private key generation center generates a portion of the drone's private key, adds it to the dynamic accumulator, updates the status, and obtains the version number, accumulator root, and member witness. Then, the drone registration information and the versioned accumulator root are anchored to the consortium blockchain, and some private keys, member witnesses and state version numbers are issued to the drones. The drone only stores the physically unclonable function challenge value, part of the private key, member witnesses, and version number, and deletes temporary intermediate values; The ground station generates its own secret using a true random number generator, calculates the ground station's public component, and submits the ground station's pseudo-identity and public component to the private key generation center of its management domain; the private key generation center generates a partial private key for the ground station and uploads the ground station's registration information to the consortium blockchain ledger; the ground station calculates the complete private key and stores it in the hardware security module; The private key generation center of each management domain submits the accumulator root and state version number updated due to drone registration or revocation to the consortium blockchain ledger for anchoring; the ground station synchronizes and verifies the latest state from the consortium blockchain ledger to form a trusted local state cache with trusted anchoring. The drone acquires the public parameters of the target ground station and pre-computes the target ground station's public key; During the online signature phase, the drone calls a physically unclonable function within a trusted execution environment to reconstruct the complete private key, uses the reconstructed complete private key to generate a signature authentication value, and outputs a cross-domain signature message. The ground station receives the cross-domain signature message and checks the timestamp; The hardware security module calls the ground station's complete private key to decrypt and restore the authentication payload; verifies the validity of the UAV's membership in the current state; after the membership verification is successful, calculates the UAV's public key based on the restored authentication payload and verifies the correctness of the signature-cryptography equation; If the verification passes, the business message is accepted.

2. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, Also includes: Randomized batch verification: For multiple candidate ciphertexts that have passed member status verification, the ground station independently generates random aggregation coefficients for each candidate ciphertext and executes the randomized batch verification equation. If the equation is true, the batch is accepted; otherwise, the invalid ciphertext is recursively partitioned, located, and isolated.

3. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, The steps for a drone to reconstruct a complete private key by calling a physically unclonable function within a trusted execution environment are as follows: the physically unclonable function module is called again to obtain the physically unclonable function response; a derived secret is calculated based on the response; and the complete private key is reconstructed by combining the stored partial private key with preset algebraic operations; the complete private key is deleted from the trusted execution environment immediately after signing is completed.

4. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, The cross-domain signature message is generated by concatenating the business message, drone pseudo-identity, drone public component, management domain identifier, member witness, and status version number in a preset order to form the authentication payload. The pre-calculated temporary public value and the authentication payload are used together to input the signature-cryptography binding hash; The drone uses the fully reconstructed private key within the trusted execution environment to generate a signed authentication value and outputs a cross-domain signed message.

5. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, The specific steps for the ground station to verify the validity of the drone's membership in its current state are as follows: If the state version number carried in the ciphertext is equal to the current state version number of the ground station, then directly verify whether the member witness corresponds to the current accumulator root; If the state version number carried in the ciphertext is less than the current state version number of the ground station, then the accumulator leaf of the UAV is determined based on the verified auxiliary update information to see if it has been revoked between the two versions. If it has not been revoked, then it is confirmed that it still has valid membership. If the state version number carried in the ciphertext is greater than the current state version number of the ground station, the ground station will temporarily suspend processing the ciphertext and perform state synchronization.

6. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, The specific steps to verify the correctness of the signature equation are as follows: The ground station calculates the drone's public key based on the recovered drone pseudo-identity, the drone's public components, and the identifier of its management domain, combined with the domain public key of the management domain. Input the recovered business message, drone identity information, member witness, status version number, temporary public value obtained from designing, and timestamp into the hash function to recalculate the hash value of the signature binding; Verify whether the temporary public value recovered from the decryption is equal to the product of the decryption authentication value and the elliptic curve generator, minus the product of the decryption binding hash value and the UAV public key.

7. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1 or 2, characterized in that, The specific method of the randomized batch verification step is as follows: Suppose that the candidate ciphertext set contains ciphertexts that have passed member status verification. For each candidate ciphertext, the ground station calculates its corresponding temporary public value, signature authentication value, signature binding hash value, and drone public key. A true random number generator independently generates random aggregation coefficients for each candidate ciphertext, and the random aggregation coefficients are uniformly selected from a set of positive integers of a preset size. Verify the weighted aggregation equation: Is the sum of the products of the random aggregation coefficients of all candidate ciphertexts and their temporary public values ​​equal to the sum of the products of the random aggregation coefficients of all candidate ciphertexts and their signature authentication values ​​multiplied by the elliptic curve generator, minus the sum of the triple products of the random aggregation coefficients of all candidate ciphertexts, the signature binding hash value, and the drone public key? If the equation is true, the batch is accepted; otherwise, invalid ciphertext is located and isolated using a recursive partitioning method. The isolated invalid ciphertext is not included in the set of valid messages.

8. The hardware-assisted certificate-free signature authentication method for cross-domain unmanned aerial vehicle networks according to claim 1, characterized in that, The consortium blockchain ledger is used to anchor globally public parameters, public parameters of each management domain, entity pseudo-identity registration records, and versioned accumulator roots, but does not participate in the real-time authentication of each signature message; the accumulator root in the local trusted state cache of the ground station must be consistent with the accumulator root of the corresponding version on the chain, and member witnesses must be able to verify the accumulator root.

9. A hardware-assisted certificate-free signature authentication system for cross-domain unmanned aerial vehicle (UAV) networks, comprising the hardware-assisted certificate-free signature authentication method for cross-domain UAV networks as described in any one of claims 1-8, characterized in that, include: A decentralized trust layer, including a consortium blockchain ledger and a private key generation center for multiple management domains; The consortium blockchain ledger is used to anchor global public parameters, public parameters of each management domain, entity pseudo-identity registration records, and versioned accumulator roots; the private key generation center is responsible for the registration of drones and ground stations within its management domain, partial private key generation, drone member status updates, and revocation status maintenance. The edge authentication layer includes a ground station, a trusted state cache, a batch authentication module, a true random number generator, and a hardware security module. The ground station is used to receive cross-domain signature messages and perform designing and member state verification. The trusted state cache is used to store local states anchored to the consortium blockchain. The batch authentication module is used to perform randomized batch verification. The hardware security module is used to protect the ground station's long-term private key. The drone layer includes a drone, a physically unclonable function module, a trusted execution environment, a member witness, and a pre-computed tuple; the physically unclonable function module is used to generate hardware secrets bound to the drone device; the trusted execution environment is used to reconstruct the drone's complete private key on demand and perform signature cryptography in a protected execution environment; the member witness is used to prove that the drone has valid membership in the corresponding state version.

Images