Multi-UAV System Mutual Authentication Method Based on Data Distribution Service

Abstract

The present invention belongs to the technical field of mutual authentication of multi-unmanned systems. The present invention discloses a method for mutual authentication of multi-unmanned systems based on data distribution service. The method includes: when receiving a connection request from an agent to be authenticated, broadcasting a request for obtaining security parameters within the communication range; receiving t -1 security parameters; calculating a virtual public key based on t -1 security parameters and the security parameters of the current agent; when the virtual public key is the same as the group public key, sending a communication key to the agent to be authenticated and adding the agent to the communication list; The present invention calculates the virtual public key based on the security parameters, compares the virtual public key with the group public key, thereby authenticating the identity of the agent to be authenticated, and solves the problem that the prior art cannot complete the agent identity authentication in the case of network interruption or offline scenario, providing an important technical guarantee for the application of multi-unmanned systems in multiple fields.

Description

Technical Field

[0001] The present invention belongs to the technical field of mutual authentication of multi-unmanned systems, and particularly relates to a method for mutual authentication of multi-unmanned systems based on data distribution service. Background Art

[0002] In a multi-unmanned system, it usually includes a ground control station and one or more groups of agents. The agents can be robots, drones or unmanned vehicles. When an agent far from the ground control station collaborates with surrounding agents, in order to confirm whether the surrounding agents are in the task group, it is necessary to authenticate their identities.

[0003] The authentication system based on a centralized certificate authority (CA) is widely adopted. This method issues digital certificates for each agent through a central server, and requires the agent to submit a certificate verification request to the ground control station during communication to confirm the identity legitimacy. For example, each agent in the multi-unmanned system needs to obtain a certificate from the ground control station in advance, and interact with the ground control station in real time during the execution of tasks to complete the identity verification.

[0004] However, in a complex natural environment, due to the inconsistent moving speeds of the ground control station and each agent, it is impossible to ensure the continuity and stability of the communication network between the agent and the ground control station during the task. Therefore, the existing method can only perform identity authentication when the agent can communicate with the ground control station, and cannot complete the identity authentication between agents in the case of network interruption or offline scenarios. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for mutual authentication of multi-unmanned systems based on data distribution service to complete the identity authentication between agents when the agent and the ground control station cannot communicate.

[0006] The present invention adopts the following technical solutions:

[0007] A method for mutual authentication of multi-unmanned systems based on data distribution service, comprising the following steps:

[0008] When receiving a connection request from an agent to be authenticated, broadcast a request for obtaining security parameters within the communication range;

[0009] Receive t -1 security parameters; wherein, t -1 security parameters include the security parameter of the agent to be authenticated and at least t -2 security parameters of authenticated agents within the communication range;

[0010] Based on t-1 security parameter and the security parameters of the current agent to calculate the virtual public key; where the security parameters include the pseudonym, public key, and expiration time;

[0011] When the virtual public key is the same as the group public key, send the communication key to the agent to be authenticated and add the agent to the communication list; where the group public key is the key shared by all authenticated agents and is used to generate the public key. t Indicates the first number of agents required to recover the group public key.

[0012] Furthermore, before broadcasting the acquisition request for the security parameters within the communication range, it includes:

[0013] Send an acquisition request for the expiration time to the agent to be authenticated.

[0014] Receive the expiration time sent by the agent to be authenticated, compare the expiration time with the current time; check whether the pseudonym of the agent to be authenticated is in the cancellation information table.

[0015] If the expiration time is later than the current time and the pseudonym is not in the cancellation information table, then continue to execute.

[0016] Furthermore, if the expiration time is earlier than the current time or the pseudonym is in the cancellation information table, then terminate the execution.

[0017] Furthermore, the specific method for generating the group public key is:

[0018] Select elliptic curve parameters based on the SM2 elliptic curve public key cryptography algorithm, and the elliptic curve parameters include the generator.

[0019] Set the first number of agents required to recover the group public key based on the Shamir threshold secret sharing scheme t ;

[0020] According to the first number t , randomly construct t -1 polynomials;

[0021] Select t The constant term in the -1 polynomial as the group private key, and use the group private key and the generator to construct the group public key.

[0022] Furthermore, the specific method for generating the public key is:

[0023] Select a random number;

[0024] Use the random number as the pseudonym of the agent, and substitute the random number as a variable into t -1 polynomial to calculate the threshold key of the agent.

[0025] Randomly generate a time fence private key;

[0026] Construct a time fence public key using the time fence private key and the generator;

[0027] Calculate the private key using the threshold key, expiration time, and time fence private key, expressed as:

[0028] ,

[0029] where, SK represents the private key, y represents the threshold key, T represents the expiration time of the agent, SKT represents the time fence private key;

[0030] Construct the public key using the private key and the generator.

[0031] Further, before sending a request to the agent to be authenticated for the expiration time, include: t Calculate the virtual public key based on

[0032] -1 security parameters and the security parameters of the current agent, including: t Calculate the Lagrange coefficients using

[0033] ,

[0034] where, represents the Lagrange coefficient of the agent with the pseudonym i , i represents the pseudonym of any agent, j represents the pseudonyms of other agents except i ;

[0035] Calculate the virtual public key using the Lagrange coefficients, expressed as:

[0036] ,

[0037] where, A represents the virtual public key, PK i represents the public key of the agent with the pseudonym i , T i represents the expiration time of the agent with the pseudonym i , PKT represents the time fence public key.

[0038] Further, before sending a request to the agent to be authenticated for the expiration time, include:

[0039] Send a request to the agent to be authenticated to obtain the pseudonym;

[0040] Receive the kana and broadcast a request to obtain the communication key of the agent to be authenticated corresponding to the kana within the communication range;

[0041] If the communication key is received, add the agent to be authenticated to the communication list.

[0042] Further, if the communication key is not received, continue to execute.

[0043] Further, if the expiration time is later than the current time and the kana is not in the cancellation information table, execute the following steps:

[0044] Send a request to obtain the public key and the first information to the agent to be authenticated; the first information is encrypted by the private key of the agent to be authenticated; the first information includes the current time and the signature of the current time;

[0045] Receive the public key and the first information;

[0046] Verify the signature of the current time using the public key to obtain the second information;

[0047] Calculate the transmission time using the current time;

[0048] If the second information is the same as the current time and the transmission time is less than the maximum allowable transmission time, continue to execute.

[0049] Further, if the second information is not the same as the current time or the transmission time is greater than or equal to the maximum allowable transmission time, terminate the execution.

[0050] The beneficial effects of the present invention are as follows: The present invention calculates the virtual public key based on the security parameters, compares the virtual public key with the group public key, thereby authenticating the identity of the agent to be authenticated. Through a certain number of security parameters, the group public key is restored to achieve the identity authentication between agents, enabling the agents to independently complete the identity authentication in the case of disconnection from the ground control station, enabling the multi-unmanned system to maintain a reliable cooperation relationship in a complex task environment, greatly improving the system survivability and task continuity, and providing an important technical guarantee for the application of the multi-unmanned system in multiple fields. Brief Description of the Drawings

[0051] Figure 1 It is a flowchart of the method of the present invention. Detailed Embodiments

[0052] The present invention will be described in detail below in conjunction with the drawings and specific embodiments.

[0053] To solve the problem of local area identity authentication in the case of network interruption or disconnection between the agent and the ground control station, it is necessary to consider the characteristics of this environment: there is no communication between the local area where the agent is located and the central area where the ground control station is located; there is mutual distrust among a limited number of agents in the local area and the agent to be authenticated; the computing power of the agent itself is limited; and the number of agents in the multi-unmanned system is variable.

[0054] The application scenario of multi-unmanned system collaborative operation consists of one or more ground control stations and multiple types of agents. The ground control station and the agents conduct one-to-many one-way communication through the Data Distribution Service (DDS). Based on DDS, a communication model is constructed. Therefore, the ground control station and the agents respectively establish topics and register as publishers of these topics to publish information. When an agent enters the communication coverage range of other agents or the ground control station, the agent automatically joins the topics of other agents or the ground control station and becomes a subscriber for information sharing.

[0055] All multi-unmanned systems use the second-generation Robot Operating System (ROS2), support the DDS communication middleware, and can modify DDS to ensure the operation of the system. Among them, the ground control station, as the network center of the multi-unmanned system, is used to remotely control and monitor the agents during task execution, and can provide information such as the status of the agents, transmitted data, and route planning in real time. The ground control station, as the ground infrastructure, is regarded as the central unit with the highest computing power in the present invention, has the starting and ending permissions of the task, assumes the role of the group administrator, and has functions such as storing the group key (group public key and group private key) and the agent key (private key and public key); the agent, as a node of the unmanned system, is regarded as a communication node with relatively low computing power in the present invention. In a network with a ground control station, the agent exists as a group member, and its identity authentication is completed by the ground control station. In a network without a ground control station, the agent can be part of the authentication agency to authenticate other agents.

[0056] Therefore, the present invention realizes efficient communication and dynamic networking. The publish-subscribe mode of DDS supports the autonomous discovery and real-time communication of agents. Combining its decentralized characteristics, it enhances the scalability and real-time performance of the system in a dynamic environment.

[0057] At the same time, the present invention combines the SM2 elliptic curve public key cryptography algorithm and the Shamir threshold secret sharing scheme for identity authentication between unmanned systems, and lightweightly solves the problem that the agent and the ground control station can only perform identity authentication when maintaining a communication connection, and cannot complete the identity authentication between agents in the case of network interruption or offline scenarios.

[0058] A mutual authentication method for multi-unmanned systems based on data distribution service, as Figure 1 shown, includes the following steps:

[0059] S110. When receiving a connection request from an agent to be authenticated, broadcast a request for obtaining security parameters within the communication range.

[0060] S120. Receive t -1 security parameters; among them, t -1 security parameters include the security parameters of the agent to be authenticated and at least t -2 security parameters of the authenticated agents within the communication range.

[0061] S130. Calculate a virtual public key based on t -1 security parameters and the security parameters of the current agent; among them, the security parameters include pseudonyms, public keys, and expiration times.

[0062] S140. When the virtual public key is the same as the group public key, send a communication key to the agent to be authenticated and add the agent to the communication list; among them, the group public key is a key shared by all authenticated agents and is used to generate a public key, t indicating the first quantity of agents required to recover the group public key.

[0063] The present invention calculates a virtual public key based on security parameters, compares the virtual public key with the group public key, thereby authenticating the identity of the agent to be authenticated, realizes the identity authentication between agents by recovering the group public key through a certain number of security parameters, enables the agent to autonomously complete identity authentication in the case of disconnection from the ground control station, enables the multi-unmanned system to maintain a reliable cooperation relationship in a complex task environment, greatly improves the system survivability and task continuity, and provides an important technical guarantee for the application of the multi-unmanned system in multiple fields.

[0064] Before broadcasting a request for obtaining security parameters within the communication range, it includes: sending a request for obtaining the expiration time to the agent to be authenticated; receiving the expiration time sent by the agent to be authenticated, comparing the expiration time with the current time; checking whether the pseudonym of the agent to be authenticated is in the cancellation information table; if the expiration time is later than the current time and the pseudonym is not in the cancellation information table, then continue to execute.

[0065] If the expiration time is earlier than the current time or the pseudonym is in the cancellation information table, then terminate the execution.

[0066] The present invention introduces the check of the expiration time, which improves the timeliness and security of authentication. By pre-verifying the expiration time, directly intercepts the communication requests of agents with expired public keys, reduces the resource consumption caused by invalid calculations, and blocks the replay attacks launched using expired public keys.

[0067] Before sending a request to the agent to be authenticated for obtaining the expiration time, it includes: sending a request to the agent to be authenticated for obtaining a pseudonym; receiving the pseudonym, and broadcasting a request to the agent to be authenticated within the communication range for obtaining the communication key corresponding to the pseudonym; if the communication key is received, adding the agent to be authenticated to the communication list.

[0068] If the communication key is not received, continue to execute.

[0069] If the expiration time is later than the current time and the pseudonym is not in the cancellation information table, perform the following steps: sending a request to the agent to be authenticated for obtaining the public key and the first information; the first information is encrypted by the private key of the agent to be authenticated; the first information includes the current time and the signature of the current time; receiving the public key and the first information; verifying the signature of the current time using the public key to obtain the second information; calculating the transmission time using the current time; if the second information is the same as the current time and the transmission time is less than the maximum allowed transmission time, continue to execute.

[0070] If the second information is not the same as the current time or the transmission time is greater than or equal to the maximum allowed transmission time, terminate the execution.

[0071] The present invention calculates the transmission time using the current time and compares it with the maximum allowed transmission time. It can effectively prevent replay attacks. When the transmission time is greater than or equal to the maximum allowed transmission time, it means that the attacker has performed a replay attack on the multi-unmanned system.

[0072] The present invention adopts a dynamic key management mechanism to achieve double security guarantees: First, after each communication session ends, a new communication key is generated based on a cryptographically secure random number generation algorithm, and the communication key is distributed through the asymmetric encryption method of the SM2 elliptic curve public key cryptography algorithm to ensure that there is no derivable mathematical relationship between the communication keys, thus having forward security; Second, when the communication key is leaked or the agent exits the communication, the system will immediately start the key rotation mechanism to block potential security threats by re-negotiating the communication key, effectively guaranteeing backward security. This two-stage protection system not only guards against the risk of historical communication keys being cracked but also can promptly eliminate the continuous impact after the communication key is leaked.

[0073] At the same time, the present invention adopts a hierarchical filtering mechanism. Through a three-level authentication process of looking up the communication list, verifying the validity of the public key, and verifying the correctness of the group public key, it realizes the gradient elimination of risk agents, reduces unnecessary operations of the system, simplifies the authentication process, and makes the authentication process more lightweight.

[0074] The group public key is generated by the ground control station. The specific generation method is:

[0075] Select elliptic curve parameters based on the SM2 elliptic curve public key cryptography algorithm. The elliptic curve parameters include a generator.

[0076] Set the first number of agents required to recover the group public key based on the Shamir threshold secret sharing scheme. t ;

[0077] According to the first number t , randomly construct a t -1 degree polynomial.

[0078] Select t the constant term in the -1 degree polynomial as the group private key, and construct the group public key using the group private key and the generator.

[0079] The present invention combines the SM2 elliptic curve public key cryptography algorithm with the Shamir threshold secret sharing scheme, improving the security and recoverability of group key generation. The SM2 elliptic curve public key cryptography algorithm provides high-strength asymmetric encryption capabilities, and the Shamir threshold secret sharing scheme disperses the risk of group key leakage through the threshold mechanism. The combination of the two can not only ensure the anti-attack ability of the group key but also support the recovery of the group key when some agents fail.

[0080] The specific method for generating the public key is as follows:

[0081] Select a random number.

[0082] Use the random number as the pseudonym of the agent and substitute the random number as a variable into the t -1 degree polynomial to calculate the threshold key of the agent.

[0083] Randomly generate a time fence private key.

[0084] Use the time fence private key and the generator to construct the time fence public key.

[0085] Calculate the private key using the threshold key, expiration time, and time fence private key, expressed as:

[0086] ,

[0087] where SK represents the private key, y represents the threshold key, T represents the expiration time of the agent, SKT represents the time fence private key.

[0088] Use the private key and the generator to construct the public key.

[0089] The present invention introduces time-fence keys (time-fence private key and time-fence public key), enhancing the timeliness management of agent keys. By binding the expiration time to the agent key, it ensures that the agent key automatically expires after a preset time, preventing the risk of agent key leakage caused by long-term exposure. Meanwhile, it supports dynamic updates to adapt to the periodic changes of tasks.

[0090] During the authentication process between the ground control station and several agents, pseudonyms are used, which are randomly generated by the ground control station during registration. The same agent cancels its pseudonym after completing the current task and can register for a new one when conducting the next task, so identity attacks on the agent cannot be carried out.

[0091] The present invention dynamically generates agent keys, taking into account both the uniqueness and controllability of agent keys. By combining pseudonyms, expiration time, and threshold keys to generate agent keys, it not only ensures the uniqueness of agent keys but also controls the life cycle of agent keys through time-fence keys, avoiding the risk of illegal agents misusing them.

[0092] Based on t -1 security parameters and the security parameters of the current agent, calculating the virtual public key includes:

[0093] Using t -1 security parameters and the security parameters of the current agent to calculate the Lagrange coefficients, expressed as:

[0094] ,

[0095] Among them, represents the Lagrange coefficient of the agent with the pseudonym i , i represents the pseudonym of any agent, j represents the pseudonyms of other agents except i .

[0096] Using the Lagrange coefficients to calculate the virtual public key, expressed as:

[0097] ,

[0098] Among them, A represents the virtual public key, PK i represents the public key of the agent with the pseudonym i , T i represents the expiration time of the agent with the pseudonym i , PKT represents the time-fence public key.

[0099] The present invention realizes secret recovery based on the Shamir threshold secret sharing scheme, disperses risks, and enhances the fault tolerance ability. By using t security parameters, the secret can be recovered, and authentication can still be completed when the data of some agents is lost or damaged, improving the stability of the system in local failure scenarios.

[0100] In summary, based on security parameters, the present invention designs a series of processes for finding the communication list, verifying the public key, and recovering the group public key in sequence to realize the identity authentication between agents, enabling the agents to autonomously complete the identity authentication in the case of disconnection from the ground control station, enabling the multi-unmanned system to still maintain a reliable cooperation relationship in a complex mission environment, greatly improving the anti-destruction ability and mission continuity of the system, and providing an important technical guarantee for the application of the multi-unmanned system in multiple fields.

Claims

1. A mutual authentication method for multi-unmanned systems based on data distribution service, characterized in that Including the following steps: When receiving a connection request from an agent to be authenticated, send a request for obtaining a pseudonym to the agent to be authenticated; Receive the pseudonym, and broadcast a request for obtaining the communication key of the agent to be authenticated corresponding to the pseudonym within the communication range; If the communication key is received, add the agent to be authenticated to the communication list; If the communication key is not received, continue to execute; Send a request for obtaining the expiration time to the agent to be authenticated; Receive the expiration time sent by the agent to be authenticated, compare the expiration time with the current time; check whether the pseudonym of the agent to be authenticated is in the cancellation information table; If the expiration time is earlier than the current time or the pseudonym is in the cancellation information table, terminate the execution; If the expiration time is later than the current time and the pseudonym is not in the cancellation information table, send a request for obtaining the public key and the first information to the agent to be authenticated; the first information is encrypted by the private key of the agent to be authenticated; the first information includes the current time and the signature of the current time; Receive the public key and the first information; Verify the signature of the current time using the public key to obtain the second information; Calculate the transmission time using the current time; If the second information is the same as the current time and the transmission time is less than the maximum allowable transmission time, continue to execute; If the second information is not the same as the current time or the transmission time is greater than or equal to the maximum allowable transmission time, terminate the execution; Broadcast a request for obtaining security parameters within the communication range; Receive t - One security parameter; wherein, t - One of the security parameters includes the security parameter of the agent to be authenticated and at least t - Two security parameters of the authenticated agents; Based on t - Calculate a virtual public key based on the one security parameter and the security parameter of the current agent; wherein, the security parameter includes a pseudonym, a public key, and an expiration time; When the virtual public key is the same as the group public key, send a communication key to the agent to be authenticated and add the agent to the communication list; wherein, the group public key is a key shared by all authenticated agents and is used to generate a public key, t represents the first number of agents required to recover the group public key.

2. The mutual authentication method for multi-unmanned systems based on data distribution service according to claim 1, wherein, The specific method for generating the group public key is as follows: Select elliptic curve parameters based on the SM2 elliptic curve public key cryptography algorithm, and the elliptic curve parameters include a generator; Set the first quantity of agents required to recover the group public key based on the Shamir threshold secret sharing scheme t ; According to the first quantity t , a random structure t -1 degree polynomial; Select the t constant term in the -1-degree polynomial as the group private key, and construct the group public key using the group private key and the generator.

3. The mutual authentication method for multi-unmanned systems based on data distribution service according to claim 2, wherein The specific method for generating the public key is as follows: Select a random number; Use the random number as the pseudonym of the agent, and substitute the random number as a variable into the t -1 degree polynomial to calculate the threshold key of the agent; Randomly generate a time fence private key; Construct a time fence public key using the time fence private key and the generator; Calculate the private key using the threshold key, the expiration time, and the time fence private key, expressed as: ， Among them, SK represents the private key, y represents the threshold key, T represents the expiration time of the agent, SKT represents the time fence private key; Construct a public key using the private key and the generator.

4. The mutual authentication method for multiple unmanned systems based on data distribution service according to claim 3, wherein Based on t Calculating a virtual public key based on the -1 security parameter and the security parameter of the current agent includes: Utilize t - Calculate the Lagrange coefficient using the safety parameter of the -1 of the said safety parameters and the safety parameter of the current agent, expressed as: ， Among them, I i represents the Lagrangian coefficient of the agent with the pseudonym i . i represents the pseudonym of any agent, j represents except i the pseudonyms of other agents. Calculate the virtual public key using the Lagrange coefficient, expressed as: ， Among them, A represents the virtual public key, PK i represents the public key of the agent with the alias i ; T i represents the expiration time of the agent with the alias i ; PKT represents the time fence public key.

Images