A security verification method, system, device, and storage medium for edge nodes.

Abstract

This invention relates to the field of network security technology, specifically to a method, system, device, and storage medium for security verification of edge nodes. The method first verifies the reputation value of edge nodes that wish to participate in processing tasks published by users, determining whether the edge node's reputation value has changed and deciding whether the edge node should execute the task. Then, based on the edge node's edge location data, public key, and private key, it verifies the node information to determine whether the edge node's information has been tampered with, deciding whether the edge node should execute the task. Finally, without requiring users and edge nodes to upload location privacy and real identities, it performs a matching index overlap check to determine whether the edge node is within the service range, selecting a secure edge node to execute the task. This method not only performs secure edge node verification but also protects the identity and location privacy of edge nodes, saving unnecessary overhead.

Description

Technical Field

[0001] This invention relates to the field of network security technology, specifically to a security verification method, system, device, and storage medium for edge nodes. Background Technology

[0002] Edge nodes are widely distributed near the terminal layer, which can coordinate and manage idle computing resources in the network, effectively solving most new challenges that cloud servers cannot solve alone, such as limited bandwidth, high latency and poor connectivity.

[0003] However, edge nodes are vulnerable to exploitation by attackers. Attackers can use malicious edge nodes to obtain users' private information by providing services, and may even collude with other maliciously compromised nodes, posing a great risk to the security of the entire network.

[0004] Existing methods often perform security checks on edge nodes by considering their location, identity, and transmitted data. Such security checks not only have poor performance but also incur high computational costs. Summary of the Invention

[0005] The purpose of this invention is to provide a security verification method, system, device, and storage medium for edge nodes.

[0006] The technical solution of this invention is as follows:

[0007] A security verification method for edge nodes includes the following operations:

[0008] S1. After receiving the task pending signal, the edge node performs a reputation value check on the edge node based on the edge node's reputation value and the standard reputation value to determine whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, proceed to S2.

[0009] S2. Obtain the edge position data of the edge node, insert the edge position data into the filter data, and perform one-dimensional vector mapping processing to obtain the first mapping filter data; based on the characteristic that the value in the first mapping filter data is 0 or 1, sum or subtract the first mapping filter data from the filter random number to obtain the second mapping filter data and the third mapping filter data; the second mapping filter data and the third mapping filter data are respectively processed by index transformation to obtain the first edge index and the second edge index; the first edge index and the second edge index form the edge pairing index; based on the edge pairing index of the edge node, the edge initial public key and the edge initial private key, perform node information verification on the edge node to determine whether the edge node information has been tampered with; if it has been tampered with, the edge node does not perform the task; if it has not been tampered with, proceed to S3.

[0010] S3. Based on the edge pairing index of the edge node, the user pairing index of the user who published the task to be processed, and the user location data, perform a pairing index overlap test on the edge node to determine whether the edge node is within the service range; if it is within the service range, the edge node executes the task; if it is not within the service range, the edge node does not execute the task.

[0011] The credit rating verification operation in S1 can be obtained through the following formula (1):

[0012] Formula (1):

[0013] If formula (1) holds true, the reputation value of the edge node remains unchanged; if formula (1) does not hold true, the reputation value of the edge node changes; in formula (1), H C Here, hash is the hash reputation value, hash() is the hash function, g is the exponential function generator, and D, D1, and D2 are the first hash security parameter, the second hash security parameter, and the third hash security parameter, respectively.

[0014] Cm′ v =g rv ·b r mod(p), Cm v =g rv ·b r mod(p),

[0015] rv′ is the reputation value of the edge node, rv is the standard reputation value, r′ is the prime number of the edge node, r is the prime number of the edge node, b is the reputation value learning parameter, mod() is the modulo operation, and p is the safe prime number.

[0016] The specific operations for obtaining the second and third mapping filter data in S2 are as follows: if the value at the current position in the first mapping filter data is 0, then the values ​​at the corresponding positions in the second and third mapping filter data are 0; if the value at the current position in the first mapping filter data is 1, then the values ​​at the corresponding positions in the second and third mapping filter data are respectively the sum of the current position value and the filtered random number, and the difference between the current position value and the filtered random number.

[0017] The node information verification operation in S2 can be achieved through formula (2):

[0018] Formula (2):

[0019] If formula (2) holds true, the edge node information has not been tampered with; if formula (2) does not hold true, the edge node information has been tampered with; in formula (2), g is the exponential function generator, r3 is the third edge learning prime number, E1, E2, E3, and E4 are the first edge reputation parameter, the second edge reputation parameter, the third edge reputation parameter, and the fourth edge reputation parameter, respectively, and PI′ EN ,PI″ EN These are the first edge index and the second edge index, respectively. This is the initial private key for the edge node. This is the initial public key for the edge node.

[0020] The operation of the paired index overlap test in S3 can be implemented by formula (3):

[0021] Formula (3):

[0022] If C in formula (3) exceeds the overlap threshold, the edge node is within the service range; if C in formula (3) does not exceed the overlap threshold, the edge node is not within the service range; in formula (3), C is the edge overlap degree, PI′ EN ,PI″ EN These are the first edge index and the second edge index, respectively, PI′ EN ,PI″ EN These are the first user index and the second user index, respectively. BFSru is the user's first mapping filter data, and || is the modulo operation.

[0023] After an edge node completes its current round of tasks, and the number of times the edge node has executed tasks exceeds the execution threshold, the edge node's reputation value is updated based on its historical positive and negative feedback data. The updated reputation value is then used to perform the next round of security checks on the edge node.

[0024] After S1, the process also includes determining whether the reputation value of the edge node exceeds the reputation value threshold; if it does, execute S2; otherwise, the edge node does not execute this task.

[0025] A security verification system for edge nodes, comprising:

[0026] The reputation value verification module is used to verify the reputation value of an edge node after it receives a task pending signal. Based on the edge node's reputation value and the standard reputation value, the module determines whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, the node information verification module is executed.

[0027] The node information verification module is used to acquire edge position data of edge nodes, insert the edge position data into filter data, and perform one-dimensional vector mapping processing to obtain first mapping filter data. Based on the characteristic that the values ​​in the first mapping filter data are 0 or 1, the first mapping filter data is summed or subtracted from the filtered random number to obtain second and third mapping filter data. The second and third mapping filter data are respectively processed by index transformation to obtain a first edge index and a second edge index. The first edge index and the second edge index form an edge pairing index. Based on the edge pairing index, public key, and private key of the edge node, the node information of the edge node is verified to determine whether the edge node information has been tampered with. If it has been tampered with, the edge node does not perform the task; if it has not been tampered with, the index overlap verification module is executed.

[0028] The index overlap verification module is used to perform an index overlap verification on edge nodes based on edge pairing indexes, user pairing indexes of users who publish tasks to be processed, and user location data, to determine whether the edge node is within the service range; if it is within the service range, the edge node executes the task; if it is not within the service range, the edge node does not execute the task.

[0029] A security inspection device for edge nodes includes a processor and a memory, wherein the processor implements the aforementioned security inspection method for edge nodes when executing a computer program stored in the memory.

[0030] A computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned security verification method for edge nodes.

[0031] The beneficial effects of this invention are as follows:

[0032] This invention provides a security verification method for edge nodes. First, for edge nodes intending to participate in task processing published by users, a reputation value verification is performed to determine if the edge node's reputation value has changed, thus deciding whether the edge node should execute the task. Then, based on the edge node's edge location data, public key, and private key, node information verification is performed to determine if the edge node's information has been tampered with, again deciding whether the edge node should execute the task. Finally, without requiring users and edge nodes to upload location privacy and real identities, a pairing index overlap verification is performed to determine if the edge node is within the service range, selecting a secure edge node to execute the task. This method not only performs secure edge node verification but also protects the identity and location privacy of edge nodes, saving unnecessary overhead. Detailed Implementation

[0033] This embodiment provides a security verification method for edge nodes, including the following operations:

[0034] S1. After receiving the task pending signal, the edge node performs a reputation value check on the edge node based on the edge node's reputation value and the standard reputation value to determine whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, proceed to S2.

[0035] S2. Obtain edge position data of edge nodes, insert the edge position data into filter data, and perform one-dimensional vector mapping processing to obtain first mapping filter data; based on the characteristic that the values ​​in the first mapping filter data are 0 or 1, sum or subtract the first mapping filter data from the filter random number to obtain second mapping filter data and third mapping filter data; the second mapping filter data and third mapping filter data are respectively processed by index transformation to obtain first edge index and second edge index; the first edge index and the second edge index form an edge pairing index;

[0036] Based on the edge pairing index, initial public key, and initial private key of the edge node, the node information of the edge node is verified to determine whether the edge node information has been tampered with. If it has been tampered with, the edge node does not execute the task; if it has not been tampered with, S3 is executed.

[0037] S3. Based on the edge pairing index of the edge node, the user pairing index of the user who published the task to be processed, and the user location data, perform a pairing index overlap test on the edge node to determine whether the edge node is within the service range; if it is within the service range, the edge node executes the task; if it is not within the service range, the edge node does not execute the task.

[0038] This embodiment provides a security verification method for edge nodes, applied in a service system consisting of a user (RU), an edge node (EN), a cloud server (CS), and a reputation center (RC). In the service system, the user (RU) uploads a task to be processed to the cloud server (CS). The cloud server (CS) sends a task processing signal to the edge node (EN). Upon receiving the signal, the edge node (EN) sends relevant information to the reputation center (RC) via the cloud server (CS) for calculations including, but not limited to, reputation value verification, node information verification, and matching index overlap verification. The reputation center (RC) feeds back the calculation results to the cloud server (CS) for verification. Based on the verification results, the cloud server (CS) controls the edge node (EN)'s ability to execute tasks. When the edge node (EN) completes the task and meets the requirements, it sends the execution result to the user (RU) via the cloud server (CS) and sends a task execution report to the reputation center (RC) for updating the reputation value of the edge node (EN).

[0039] This embodiment provides a security verification method for edge nodes, the specific scheme of which is as follows:

[0040] S1. After receiving the task pending signal, the edge node performs a reputation value check on the edge node based on the edge node reputation value and the standard reputation value to determine whether the edge node reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, proceed to S2.

[0041] After receiving the signal that a task is pending, the edge node sends its own relevant information to the reputation center through the cloud server for parameter calculation in the reputation value verification. The reputation center then sends the parameter calculation results back to the cloud server for parameter verification to determine whether the edge node's reputation value has changed and whether the edge node is secure.

[0042] The credit rating verification process can be obtained using the following formula (1):

[0043] Formula (1):

[0044] If formula (1) holds true, that is, both sides of the equation hold true, then the reputation value of the edge node has not changed, the reputation value rv′ uploaded by the edge node itself is equal to the standard reputation value rv assigned to the edge node by the reputation center, the edge node is reliable, and S2 is executed; if formula (1) does not hold true, then the reputation value of the edge node has changed, the edge node is unreliable, and the edge node does not execute this task.

[0045] In formula (1), H C H is the hash reputation value. c = hash(Cm1||Cm2), where hash() is the hash function, and Cm1 = g ω ·b η ·mod(p) is the first initial credit value, Cm2 = g ω ·b η ·mod(p) is the second initial credit value, g is the exponential function generator, and D = ω + H c ·rv′、D1=η+H c ·r′、D2=η′+H c ·r represents the first hash security parameter, the second hash security parameter, and the third hash security parameter, respectively; ω, η, and η′ represent the compensation prime numbers for the first hash security parameter, the second hash security parameter, and the third hash security parameter, respectively; rv represents the edge node reputation value, rv represents the standard reputation value provided by the reputation center, r′ represents the edge node prime number, and r represents the edge node prime number.

[0046] Among them, Cm′ v =g rv ·b rmod(p), Cm v = gr v·b r mod(p), rv′ is the reputation value of the edge node, rv is the standard reputation value, r′ is the prime number of the edge node, r is the prime number of the edge node, b is the reputation value learning parameter, mod() is the modulo operation, and p is the safe prime number.

[0047] The calculation process for each parameter is implemented through the reputation center, and the process of determining whether both sides of the equation are true is implemented by the cloud server.

[0048] To further enhance the security of edge node verification, after S1, the cloud server also determines whether the edge node's reputation value exceeds a reputation value threshold; if it does, S2 is executed; otherwise, the edge node does not execute this task. The reputation value threshold can be set by the user who issued the task request.

[0049] S2. Obtain the edge position data of the edge node, insert the edge position data into the filter data, and perform one-dimensional vector mapping processing to obtain the first mapping filter data. Based on the characteristic that the value in the first mapping filter data is 0 or 1, sum or subtract the first mapping filter data from the filter random number to obtain the second mapping filter data and the third mapping filter data. The second mapping filter data and the third mapping filter data are respectively processed by index transformation to obtain the first edge index and the second edge index. The first edge index and the second edge index form the edge pairing index. Based on the edge pairing index of the edge node, the edge initial public key, and the edge initial private key, perform node information verification on the edge node to determine whether the edge node information has been tampered with. If it has been tampered with, the edge node does not perform the task; if it has not been tampered with, proceed to S3.

[0050] Once the reputation value of an edge node remains unchanged, the reputation center will calculate the parameters in the edge pairing index verification process based on the edge node data sent by the edge node through the cloud server, and send the parameter calculation results to the cloud server. The cloud server will then determine whether the edge node information has been tampered with, and control the execution capabilities of the edge node based on the determination result.

[0051] First, a hash function is used that combines multiple random functions. The edge position data SEN of the edge nodes is inserted into the filter data of the Bloom filter, and a one-dimensional vector mapping process is performed to obtain the first mapped filter data BF. SEN ={(x1, x2, ... x δ ) T |x i ∈{0,1}}. The data at each position in the first mapping filter data is either 0 or 1.

[0052] Then, based on the numerical features at each position in the first mapped filter data, the second and third mapped filter data are obtained. The operation for obtaining the second and third mapped filter data is as follows: if the value at the current position in the first mapped filter data is 0, then the value at the corresponding position in the second and third mapped filter data is 0; that is, for... If the value at the i-th position (the current position) is 0, x i =0, then If the value at the current position in the first mapping filter data is 1, that is, x i =1, then the values ​​at corresponding positions in the second and third mapping filter data are respectively the sum of the current position value and the filter random number, that is... And the difference between the current position value and the filtered random number, i.e. r is the prime number of edge nodes. Therefore, the second mapping filter data is obtained. Third mapping filter data

[0053] Next, the data from the second and third mapping filters are subjected to index transformation to obtain the edge pairing index. The formula for the index transformation is: PI′ EN ,PI″ EN Let M1 and M2 be the first edge index and the second edge index, respectively. The first edge index and the second edge index form the edge pairing index. M1 and M2 are the first invertible matrix and the second invertible matrix, respectively.

[0054] Finally, node information verification is performed, which can be achieved using formula (2):

[0055] Formula (2):

[0056] If formula (2) holds true, i.e., both sides of the equation hold true, then the edge node information has not been tampered with, and S3 is executed; if formula (2) does not hold true, then the edge node information has been tampered with, and the edge node does not execute the task; in formula (2), g is the exponential function generator, r3 is the third edge learning prime number, E1, E2, E3, and E4 are the first edge reputation parameter, the second edge reputation parameter, the third edge reputation parameter, and the fourth edge reputation parameter, respectively, and PI′ EN ,PI″ EN These are the first edge index and the second edge index, respectively. This is the initial private key for the edge node, which can be set according to requirements. is the initial public key for the edge node, and g is the exponential function generator.

[0057] The process of calculating each parameter is implemented through a reputation center, while the process of determining whether both sides of the equation are true is implemented by a cloud server.

[0058] S3. Based on the edge pairing index of the edge node, the user pairing index of the user who published the task to be processed, and the user location data, perform a pairing index overlap test on the edge node to determine whether the edge node is within the service range; if it is within the service range, the edge node executes the task; if it is not within the service range, the edge node does not execute the task.

[0059] Once the node information of the edge node has not been tampered with, the reputation center will perform parameter calculations in the matching index overlap test based on the edge node data sent by the edge node through the cloud server, and send the parameter calculation results to the cloud server. The cloud server will then compare the edge overlap of the edge node with the overlap threshold and control the execution capability of the edge node based on the judgment result.

[0060] The operation of the paired index overlap test can be performed using formula (3):

[0061] Formula (3):

[0062] If the edge overlap degree C in formula (3) exceeds the overlap degree threshold, the edge node is within the service range and executes the task; if C in formula (3) does not exceed the overlap degree threshold, the edge node is outside the service range and does not execute the task. C is the edge overlap degree, PI′ EN ,PI″ EN These are the first edge index and the second edge index, respectively, PI′ EN ,PI″ EN These are the first user index and the second user index, respectively. The acquisition method is similar to that of the first edge index and the second edge index mentioned above. BFSru is the user's first mapping filter data, and || is the modulo operation.

[0063] After the edge node passes the above test and completes the task for this round, and the number of times the edge node has executed the task exceeds the execution threshold, the reputation value of the edge node is updated based on the historical positive feedback data and historical negative feedback data of the edge node. The updated reputation value is used to execute the security test of the edge node in the next round.

[0064] The updated reputation score can be obtained using the following formula:

[0065] θ t =R f ·θ t-1 +R g ·f t ,

[0066] βt =R f ·β t-1 +(1-R g )·μ t ,

[0067]

[0068]

[0069] θ t Let R be the normalized positive feedback number of the edge node in round t. f For the forgetting factor, θ t f is the standardized positive feedback number of the edge node in the (t-1)th round. t Let β be the sum of the number of positive feedbacks at the edge nodes in round t. t Let β be the standardized negative feedback number of the edge node in round t. t-1 Let μ be the standardized negative feedback number of the edge node in round t-1. t RV represents the total number of negative feedbacks at the edge nodes in round t, and RV represents the updated reputation value.

[0070] The parameter calculations in the aforementioned reputation value update process, as well as the operation of replacing the original standard reputation value of the edge node with the updated reputation value, are implemented through the reputation center.

[0071] Meanwhile, the operation of updating reputation values ​​can also be performed after the reputation values ​​of edge nodes change in S1, which gives edge nodes the opportunity to participate in the next round of task execution.

[0072] The number of positive feedbacks and the number of negative feedbacks from the aforementioned edge nodes represent the number of times the user approves or disapproves the task results after the cloud server sends the task results processed by the edge nodes to the user.

[0073] When there are several edge nodes executing a task, the cloud server sends the mode of the task results processed by the edge nodes, and the corresponding task result, to the user who needs it.

[0074] In addition, to prevent malicious users from obtaining malicious edge nodes and thus disrupting system stability, before executing the edge node security verification method of this embodiment, the requesting user, edge node, and cloud server must register with the reputation center. The registration process is as follows:

[0075] User registration is required. First, the user randomly selects an initial private key. And calculate the user's initial public key via secure channel Send to the reputation center The requester provides their identity information. Then, the reputation center receives the requester's information. Then, calculate the user's intermediate private key. T i The timestamp of the user's registration is α, and α is a random number generated by the reputation center. The reputation center will... Sent to the requesting user via a secure channel, δ,M1,M2,V,H k These are, respectively, the security parameters, the first invertible matrix, the second invertible matrix, and the random vector. When the requesting user receives... Then, (δ,M1,M2,V,H) k Store and put Split into the following rules

[0076]

[0077] Next, the user needs to calculate the first component of the user's intermediate private key. Second component of the user's intermediate private key

[0078] Finally, the user's security public key is obtained. Requires user's secure private key

[0079] Analogous to the above process, a secure public key and a secure private key for the edge node are generated, but the edge identity information of the edge node is provided by a pseudonym from the reputation center. To replace it.

[0080] You can select the cloud server security public key and cloud server security private key and register them with the reputation center according to your needs.

[0081] This embodiment provides a security verification method for edge nodes. First, for edge nodes wanting to participate in task processing published by users, a reputation value verification is performed to determine if the edge node's reputation value has changed, thus deciding whether the edge node should execute the task. Then, based on the edge node's edge location data, public key, and private key, node information verification is performed to determine if the edge node's information has been tampered with, again deciding whether the edge node should execute the task. Finally, without requiring users and edge nodes to upload location privacy and real identities, a pairing index overlap verification is performed to determine if the edge node is within the service range, selecting a secure edge node to execute the task. This method not only performs secure edge node verification but also protects the identity and location privacy of edge nodes, saving unnecessary overhead.

[0082] The security verification method for edge nodes provided in this embodiment can also achieve the following:

[0083] Identity privacy. All entities register with the reputation center, and during the service process, the privacy information of edge nodes remains anonymous to all entities except trusted entities.

[0084] Location privacy. For users and edge nodes, only they and trusted entities know their location privacy.

[0085] Unloading path privacy. The processing or unloading path of a task request is known only to the edge nodes involved in the computation and the reputation center. It is difficult for other entities to infer the relevant path information for task processing or unloading from the existing information.

[0086] Reputation value privacy. The reputation value of edge nodes is not exposed or linked.

[0087] Unlinkability. Given two task requests, no adversary can determine whether they are offloaded to the same edge node for processing; nor can an adversary infer the reputation value of the edge node participating in the task from its unchanging or slightly changing reputation value. That is, the processing or offloading path of the task request and the reputation value of the edge node are private and cannot be speculated by malicious nodes.

[0088] Confidentiality, integrity, and authentication. Information will not be tampered with or replaced during transmission; even if tampered with, the anomaly can be detected by the cloud server and reputation center. Simultaneously, the cloud server can verify the legitimacy of task requests, the reputation value of edge nodes, and their task matching degree.

[0089] Non-repudiation. If a user engages in malicious behavior, the reputation center can track their true location and identity; if an edge node fails to complete the computation task as required, the reputation center can adjust its reputation value to refuse its continued participation in the computation.

[0090] Unforgeability. Malicious edge nodes cannot forge information related to legitimate edge nodes and pass verification; malicious edge nodes cannot participate in computational tasks by forging reputation values.

[0091] In addition, this embodiment also provides a security inspection system for edge nodes, including:

[0092] The reputation value verification module is used to verify the reputation value of an edge node after it receives a task pending signal. Based on the edge node's reputation value and the standard reputation value, the module determines whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, the node information verification module is executed.

[0093] The node information verification module is used to acquire edge position data of edge nodes, insert the edge position data into filter data, and perform one-dimensional vector mapping processing to obtain first mapped filter data. Based on the characteristic that the values ​​in the first mapped filter data are 0 or 1, the first mapped filter data is summed or subtracted from the filtered random number to obtain second and third mapped filter data. The second and third mapped filter data are respectively processed by index transformation to obtain first edge index and second edge index. The first edge index and the second edge index form an edge pairing index. Based on the edge pairing index, public key, and private key of the edge node, the node information of the edge node is verified to determine whether the edge node information has been tampered with. If it has been tampered with, the edge node does not perform the task; if it has not been tampered with, the index overlap verification module is executed.

[0094] The index overlap verification module is used to perform an index overlap verification on edge nodes based on edge pairing indexes, user pairing indexes of users who publish tasks to be processed, and user location data, to determine whether the edge node is within the service range; if it is within the service range, the edge node executes the task; if it is not within the service range, the edge node does not execute the task.

[0095] This embodiment also provides a security inspection device for edge nodes, including a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the above-described security inspection method for edge nodes.

[0096] This embodiment also provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the above-described security verification method for edge nodes.

Claims

1. A security verification method for edge nodes, characterized in that, This includes the following operations: S1. After receiving the task pending signal, the edge node performs a reputation value check on the edge node based on the edge node's reputation value and the standard reputation value to determine whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, proceed to S2. S2. Obtain the edge position data of the edge nodes, insert the edge position data into the filter data, and perform one-dimensional vector mapping processing to obtain the first mapping filter data. Based on the characteristic that the values ​​in the first mapping filter data are 0 or 1, the first mapping filter data and the filter random number are summed or subtracted to obtain the second mapping filter data and the third mapping filter data. Specifically: if the value at the current position in the first mapping filter data is 0, then the value at the corresponding position in the second and third mapping filter data is 0; if the value at the current position in the first mapping filter data is 1, then the values ​​at the corresponding positions in the second and third mapping filter data are respectively the sum of the value at the current position and the filter random number, and the difference between the value at the current position and the filter random number. The second and third mapping filter data are respectively processed by index transformation to obtain the first edge index and the second edge index; the first edge index and the second edge index form the edge pairing index. Based on the edge pairing index, initial public key, and initial private key of the edge nodes, the node information of the edge nodes is verified to determine whether the edge node information has been tampered with; the node information verification operation can be implemented by formula (2): Official (2): , If formula (2) is true, then the edge node information has not been tampered with; if formula (2) is false, then the edge node information has been tampered with; in formula (2), For exponential function generator, Learning prime numbers for the third edge, These are the first edge reputation parameter, the second edge reputation parameter, the third edge reputation parameter, and the fourth edge reputation parameter, respectively. These are the first edge index and the second edge index, respectively. This is the initial private key for the edge node. The initial public key for the edge node; If the edge node is tampered with, it will not execute the task; if it is not tampered with, it will execute S3. S3. Based on the edge node pairing index, the user pairing index of the user who published the task to be processed, and the user location data, perform a pairing index overlap test on the edge nodes to determine whether the edge nodes are within the service range; the pairing index overlap test can be implemented by formula (3): Official (3): , If C in formula (3) exceeds the overlap threshold, the edge node is within the service range; if C in formula (3) does not exceed the overlap threshold, the edge node is not within the service range; in formula (3), C is the edge overlap degree. These are the first user index and the second user index, respectively. For the user's first mapping filter data, For modulo operation; If within the service range, the edge node executes the task; if outside the service range, the edge node does not execute the task.

2. The security verification method for edge nodes according to claim 1, characterized in that, The credit value verification operation in S1 can be obtained through the following formula (1): Official (1): , If formula (1) holds true, the reputation value of the edge node remains unchanged; if formula (1) does not hold true, the reputation value of the edge node changes. In formula (1), H C Here, is the hash reputation value, hash() is the hash function, g is the exponential function generator, and D, D1, and D2 are the first hash security parameter, the second hash security parameter, and the third hash security parameter, respectively. ， ， The reputation value of the edge node. As a standard reputation value, For the edge nodes, prime numbers Let b be the prime number of the edge node, b be the reputation value learning parameter, mod() be the modulo operation, and p be the safe prime number.

3. The security verification method for edge nodes according to claim 1, characterized in that, After an edge node completes its current round of tasks, and the number of times the edge node has executed tasks exceeds the execution threshold, the edge node's reputation value is updated based on its historical positive and negative feedback data. The updated reputation value is then used to perform the next round of security checks on the edge node.

4. The security verification method for edge nodes according to claim 1, characterized in that, After the operation of S1, it also includes determining whether the reputation value of the edge node exceeds the reputation value threshold; If the limit is exceeded, execute S2; if the limit is exceeded, the edge node will not execute this task.

5. A security inspection system for edge nodes, characterized in that, include: The reputation value verification module is used to verify the reputation value of an edge node after it receives a task pending signal. Based on the edge node's reputation value and the standard reputation value, the module determines whether the edge node's reputation value has changed. If it has changed, the edge node does not execute the task; if it has not changed, the node information verification module is executed. The node information verification module is used to obtain the edge position data of the edge nodes, insert the edge position data into the filter data, and perform one-dimensional vector mapping processing to obtain the first mapping filter data. Based on the characteristic that the values ​​in the first mapping filter data are 0 or 1, the first mapping filter data and the filter random number are summed or subtracted to obtain the second mapping filter data and the third mapping filter data. Specifically: if the value at the current position in the first mapping filter data is 0, then the value at the corresponding position in the second and third mapping filter data is 0; if the value at the current position in the first mapping filter data is 1, then the values ​​at the corresponding positions in the second and third mapping filter data are respectively the sum of the value at the current position and the filter random number, and the difference between the value at the current position and the filter random number. The second and third mapping filter data are processed by index transformation to obtain the first edge index and the second edge index; the first edge index and the second edge index form the edge pairing index; based on the edge pairing index, public key and private key of the edge node, the node information of the edge node is checked to determine whether the edge node information has been tampered with; the node information check operation can be implemented by formula (2): Formula (2): If formula (2) holds true, then the edge node information has not been tampered with; if formula (2) does not hold true, then the edge node information has been tampered with; in formula (2), g is an exponential function generator. Learning prime numbers for the third edge, These are the first edge reputation parameter, the second edge reputation parameter, the third edge reputation parameter, and the fourth edge reputation parameter, respectively. These are the first edge index and the second edge index, respectively. This is the initial private key for the edge node. The initial public key for the edge node; If the edge node is tampered with, it will not perform any tasks; if it is not tampered with, it will perform the index overlap check module. The index overlap verification module is used to perform index overlap verification on edge nodes based on edge pairing indexes, user pairing indexes of users who publish tasks to be processed, and user location data, to determine whether the edge nodes are within the service range. The operation of the paired index overlap test can be performed using formula (3): Formula (3): If C in formula (3) exceeds the overlap threshold, the edge node is within the service range; if C in formula (3) does not exceed the overlap threshold, the edge node is not within the service range; in formula (3), C is the edge overlap degree. These are the first user index and the second user index, respectively. For the user's first mapping filter data, For modulo operations; if within the service range, edge nodes execute tasks; if outside the service range, edge nodes do not execute tasks.

6. A security inspection device for edge nodes, characterized in that, It includes a processor and a memory, wherein the processor implements the security verification method for edge nodes as described in any one of claims 1-4 when executing a computer program stored in the memory.

7. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the security verification method for edge nodes as described in any one of claims 1-4.

Images