Trust management method, device and electronic equipment for industrial Internet
Through the combination of random forest algorithm and sliding time window, the problem of delay and computational complexity of trust management methods in the industrial Internet is solved, and the accuracy and efficiency of trust evaluation are improved. It is suitable for the Internet of Vehicles with high latency sensitivity Scene.
Patent Information
- Application Number
- CN202210189963.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-28
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2042-02-28
AI Technical Summary
The traditional cryptography-based trust management method in the industrial Internet has problems with high latency sensitivity and computing complexity, resulting in large resource losses and time delays, making it difficult to adapt to the dynamic network topology of the Internet of Vehicles.
The random forest algorithm is used to evaluate the trust of the motion nodes in the motion cluster, encrypt the decision results through the elliptic curve and hash function, and weighted summing is used to achieve distributed trust management in the industrial Internet.
Without increasing the trust evaluation bias, the impact of decision tree training failure in the random forest algorithm is reduced, the accuracy of trust evaluation of industrial Internet network motion nodes is improved, and the computational complexity and resource loss are reduced.
Smart Images

Figure CN114844641B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of industrial Internet communication security technology, and in particular to a trust management method, device and electronic equipment for the industrial Internet. Background Art
[0002] The highly dynamic network topology, unfamiliar relationships between logistics vehicles, exposed communication connections, etc. in the Industrial Internet inevitably pose security threats to the Industrial Internet. Appropriate trust management methods can identify malicious nodes, resist malicious attacks, and ensure the stability of the communication process, thereby improving the driving conditions of moving nodes and ultimately improving the security of moving nodes in the Industrial Internet scenario.
[0003] In view of the extremely high latency sensitivity of the Industrial Internet, traditional cryptography-based trust management methods in distributed trust management face serious problems such as high computational complexity, large resource loss, and long latency. Summary of the invention
[0004] In view of this, the purpose of this application is to propose a trust management method, device and electronic equipment for the industrial Internet to solve or partially solve the above-mentioned technical problems.
[0005] Based on the above purpose, the first aspect of the present application provides a trust management method for industrial Internet, the method comprising:
[0006] The first motion nodes in the industrial Internet are clustered according to the proximity principle to obtain motion clusters;
[0007] Obtaining a trust evidence set of the first motion node;
[0008] Based on the trust evidence set, a decision tree pre-constructed using a random forest algorithm is used to perform a trust evaluation on the motion cluster to obtain a decision result corresponding to the motion cluster;
[0009] Encrypting the decision result using an elliptic curve and a hash function to obtain an encrypted decision result;
[0010] Sending the encryption decision result to a base station node through a second motion node in the motion cluster;
[0011] The base station node uses a sliding time window to perform weighted summation on the encryption decision results to obtain the final trust value of the first motion node, thereby completing the motion trust management in the industrial Internet.
[0012] Based on the same inventive concept, the second aspect of the present application proposes a trust management device for an industrial Internet, comprising:
[0013] The clustering module is configured to cluster the first motion nodes in the industrial Internet according to the proximity principle to obtain motion clusters;
[0014] A trust evidence acquisition module, configured to acquire a trust evidence set of the first motion node;
[0015] A trust evaluation module is configured to perform a trust evaluation on the motion cluster based on the trust evidence set by using a decision tree pre-constructed by a random forest algorithm to obtain a decision result corresponding to the motion cluster;
[0016] An information encryption module is configured to encrypt the decision result using an elliptic curve and a hash function to obtain an encrypted decision result;
[0017] An information sending module, configured to send the encryption decision result to a base station node through a second motion node in the motion cluster;
[0018] The trust value calculation module is configured to perform weighted summation on the encryption decision results through the base station node using a sliding time window to obtain the final trust value of the first motion node, thereby completing motion trust management in the industrial Internet.
[0019] Based on the same inventive concept, the third aspect of the present application proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.
[0020] From the above, it can be seen that the trust management method, device and electronic device for the industrial Internet provided by the present application use a random forest algorithm to classify the motion nodes in each motion cluster to obtain high-trust motion nodes and low-trust motion nodes, which can limit the overfitting problem of distributed trust management decision learning without increasing the trust assessment bias, and use a sliding time window to store the classification results of the motion nodes, thereby reducing the impact of decision tree training failures in the random forest algorithm, thereby improving the accuracy of trust assessment of motion nodes in the industrial Internet network. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0022] Figure 1 A flow chart of a trust management method for the industrial Internet according to an embodiment of the present application;
[0023] Figure 2 A schematic diagram of the process of motion cluster classification according to an embodiment of the present application;
[0024] Figure 3 A flow chart of trust evaluation according to an embodiment of the present application;
[0025] Figure 4 A schematic diagram of a process for constructing a decision tree according to an embodiment of the present application;
[0026] Figure 5 A schematic diagram of a process of applying a sliding time window in an embodiment of the present application;
[0027] Figure 6 A schematic diagram of the structure of a trust management device for the industrial Internet according to an embodiment of the present application;
[0028] Figure 7 A schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0029] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.
[0030] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0031] As described in the background technology, in recent years, driven by the Internet of Things technology, the industrial Internet + transportation environment has enabled rapid development in various fields, which has brought certain opportunities and challenges to the development of logistics and transportation companies. Logistics and transportation companies need to strengthen the effective application of Internet technology, improve the management of transportation vehicles, and solve problems such as transportation safety and traffic congestion. With the rapid development of technologies such as wireless access and core networks, the Internet of Vehicles built on the industrial Internet is being widely deployed. The ultimate goal of the Internet of Vehicles is to improve the safety, intelligence and efficiency of the transportation system by exchanging key information between vehicles and other transportation entities. In the Internet of Vehicles, vehicles are equipped with intelligent devices such as wireless sensors and OBU (On board Unit), which have powerful communication, storage and computing capabilities. In addition, the Internet of Vehicles can also realize the integration of intelligent transportation system management and dynamic information services, and can also reduce the number of traffic accidents and alleviate traffic congestion. However, the highly dynamic network topology of the Internet of Vehicles, the unfamiliar relationship between vehicles, and the exposed communication connections inevitably pose a security threat to the Internet of Vehicles. On the one hand, the information received by the vehicle may deviate from the real information due to the failure of intelligent devices such as sensors; on the other hand, malicious vehicles will inject false information into the network to deceive honest vehicles in the network and obtain their own illegal interests. As one of the most common types of vehicles in the Internet of Vehicles scenario, transport vehicles will also be attacked by malicious nodes, which has a significant impact on whether transport vehicles can safely and accurately carry out delivery according to the established plan.
[0032] An appropriate trust management mechanism can identify malicious nodes, resist malicious attacks, and ensure the stability of the communication process, thereby improving driving conditions and ultimately improving the safety of transportation vehicles in industrial Internet scenarios. An appropriate trust sharing mechanism can accurately share the trust information of vehicle nodes in the network to ensure the global consistency of node trust information. For systems with extremely high latency sensitivity in Internet of Vehicles, traditional cryptography-based trust management and sharing mechanisms face serious problems such as high computational complexity, large resource loss, and long latency. Therefore, many studies focus on non-cryptography-based trust management mechanisms. With the development of machine learning, trust management mechanisms based on machine learning have become a hot topic in current research. For traditional trust sharing mechanisms, a central server is used to share vehicle trust information. For dynamic network structures, fixed servers will limit system efficiency and scalability.
[0033] In summary, it is of great research significance to establish efficient trust management for industrial Internet logistics and transportation scenarios. Random forest is an integrated learning method with the advantages of high accuracy, introduction of randomness, not easy to overfit, good noise resistance, fast training speed, etc. It is suitable for building a trust management mechanism for industrial Internet logistics and transportation vehicles.
[0034] The concept of "trust management" was first proposed in 1996. Trust management is an important part of network service security. Trust management mechanisms can be considered from the aspects of identity authentication, attack detection and mitigation, confidentiality, privacy, trust and reputation. The core of trust management is to formulate a suitable trust evaluation mechanism based on precise rules. Malicious nodes are identified by calculating trust values, and other nodes in the network select trusted nodes to interact with.
[0035] According to the framework of the trust management mechanism, it can be divided into two categories: centralized management and distributed management. For centralized trust management, a trusted entity in the network is required to execute the trust management mechanism, and all information is stored again in the central server. When a vehicle needs to exchange information with other vehicles, it must communicate with the central server. The centralized trust management mechanism has good stability but poor scalability and cannot adapt to the highly dynamic network topology of the Internet of Vehicles. In addition, it also faces the problem of single point failure. Distributed trust management is more suitable for the Internet of Vehicles. At present, the research on distributed trust management mechanism mainly adopts the following underlying technologies: encryption technology, recommendation technology, fuzzy logic technology, game theory technology and machine learning technology.
[0036] Vehicles travel at high speeds according to a predetermined route. It is not enough to establish a suitable trust management mechanism to ensure the security of the Internet of Vehicles. How to prove the consistency of "trust" is another issue worthy of attention. At present, most researchers use a central system controller to share the trust value of vehicles. However, the location of the central controller is generally stationary. As the distance between the vehicle and the controller increases, the effectiveness of trust sharing decreases significantly, which limits the scalability of the Internet of Vehicles.
[0037] Considering the decentralization, immutability, transparency and fault tolerance of blockchain, many researchers use blockchain technology to implement trust management and sharing mechanisms. However, it must be noted that the mining cost of running the consensus mechanism is very expensive and requires a lot of computing and storage resources, which limits the application of blockchain.
[0038] For the centralized trust management mechanism, the central server is fixed and needs to manage the entire system. When a node needs to communicate with other nodes, it must first establish a connection with the central server to obtain communication permissions. Although it has good stability and accuracy, it is not suitable for dynamic vehicle networking scenarios. It has poor scalability and cannot adapt to the highly dynamic network topology of the Internet of Vehicles. It also faces single point failure problems. Once the central server fails or is hacked, the entire system will be paralyzed. Distributed trust management mechanisms can solve the above problems. Cryptography-based encryption technology is often the first line of defense for communication systems to resist attacks. However, traditional cryptography-based trust management and sharing mechanisms often require powerful computing power and face serious problems such as high computational complexity, large resource loss, and long latency. It is a challenge to apply it to the Internet of Vehicles scenario. Non-cryptography-based trust management mechanisms have certain advantages in this regard. For the traditional trust sharing mechanism, it uses a central controller to globally unify the trust value of the vehicle to achieve global unification of trust information. It also faces problems such as poor scalability, high latency, and single point failure, and will generate a lot of signaling overhead. With the development of blockchain technology, considering its characteristics of decentralization, immutability, transparency and fault tolerance, many scholars have applied it to the trust sharing mechanism. However, the mining cost of running the consensus mechanism is very expensive and requires a lot of computing and storage resources, which limits the application of blockchain.
[0039] An embodiment of the present application provides a trust management method for the industrial Internet, which can be executed on a client that installs a corresponding motion node, or on a corresponding server.
[0040] like Figure 1 As shown, the method of this embodiment includes:
[0041] Step 101, clustering the first motion nodes in the industrial Internet according to the proximity principle to obtain motion clusters.
[0042] In this step, the first motion node can be an unmanned vehicle, aircraft or other means of transportation, or a vehicle, aircraft or other means of transportation equipped with an automatic driving system. The preferred first motion node of this application is an unmanned logistics vehicle. The logistics vehicle can be equipped with intelligent devices such as wireless sensors, has communication, storage and computing capabilities, can freely communicate, can be equipped with GPS (Global Positioning System), and has a known starting position and end point.
[0043] Step 102: Obtain a trust evidence set of the first motion node.
[0044] In this step, the trust evidence may be a trust indicator generated according to the number of messages sent by the first moving node, the message value, and the communication link used to send the messages.
[0045] Step 103 , based on the trust evidence set, a decision tree pre-constructed using a random forest algorithm is used to perform a trust evaluation on the motion cluster to obtain a decision result corresponding to the motion cluster.
[0046] In this step, the first motion node in the motion cluster uses the decision tree in the random forest algorithm to make a decision on whether the first motion node belongs to the high trust category based on the trust evidence set, and then obtains the high trust category and the low trust category in the motion cluster according to the decision result, which can limit the overfitting problem of distributed trust management decision learning without increasing the trust evaluation bias.
[0047] Step 104: encrypt the decision result using an elliptic curve and a hash function to obtain an encrypted decision result.
[0048] In this step, since the communication of the industrial Internet is completed through an open wireless channel, there are many attacks on the wireless channel. The trust evaluation method based on random forest can better protect the communication process of the first motion node, but the communication process between the first motion node and the second motion node, and the second motion node and the base station node lacks a protection mechanism. Once the above communication process is attacked by an attacker, high-aggregation information such as decision results will be directly leaked, seriously threatening the security of the industrial Internet. Considering the limitations of the first motion node's computing, storage, and communication capabilities and its sensitivity to high latency, the decision results are encrypted using a lightweight encryption mechanism of elliptic curve encryption and encrypted hash functions.
[0049] Step 105: Send the encryption decision result to the base station node through the second motion node in the motion cluster.
[0050] In this step, the second motion node can be an unmanned vehicle, aircraft or other means of transportation equipped with a communication route, or a vehicle, aircraft or other means of transportation equipped with a communication route and an automatic driving system. The preferred second motion node of this application is a bus equipped with a communication route. The base station node can be a roadside fixed device for accessing the second motion node, and the preferred RSU (Roadside Unit) of this application. After lightweight encryption, the communication process between the second motion node and the base station node ensures a high degree of trust between the motion cluster and the base station node in the industrial Internet under low latency conditions.
[0051] Step 106, using the base station node to perform weighted summation on the encryption decision results using a sliding time window to obtain a final trust value of the first motion node, thereby completing motion trust management in the industrial Internet.
[0052] In this step, a sliding time window is used to store the classification results of the moving nodes, which reduces the impact of decision tree training failure in the random forest algorithm and thus improves the accuracy of trust assessment of moving nodes in industrial Internet networks.
[0053] Through the above scheme, the random forest algorithm is used to classify the motion nodes in each motion cluster to obtain high-trust motion nodes and low-trust motion nodes. This can limit the overfitting problem of distributed trust management decision learning without increasing the trust assessment bias. The sliding time window is used to store the classification results of the motion nodes, which reduces the impact of decision tree training failure in the random forest algorithm, thereby improving the accuracy of trust assessment of motion nodes in industrial Internet networks.
[0054] In some embodiments, Figure 2 As shown, step 101 specifically includes:
[0055] Step 1011, determining the second motion node in the industrial Internet through the communication link of the base station node;
[0056] Step 1012, calculating the Euclidean distance between the first motion node and the second motion node, and taking all the first motion nodes whose Euclidean distance is less than a predetermined value as the initial motion cluster of the second motion node;
[0057] Step 1013, obtaining the driving directions of all the first motion nodes in the initial motion cluster;
[0058] Step 1014, taking the first motion nodes with the same driving direction in the initial motion cluster to obtain a first initial motion cluster;
[0059] Step 1015, obtaining the communication range of the second moving node;
[0060] Step 1016: Take the first motion nodes in the first initial motion cluster that are smaller than the communication range to obtain a second initial motion cluster, and use the second initial motion cluster as the motion cluster.
[0061] In the above scheme, the mobility of the moving nodes in the industrial Internet leads to dynamic changes in its network topology, which will inevitably increase the difficulty of communication system management. Therefore, a dynamic clustering mechanism is proposed to solve the above problem. Among them, there is a second moving node in each moving cluster as the cluster head node to manage the first moving node in the moving cluster. The first moving node in the network will be divided into several small clusters according to the current location, destination location, driving direction and communication range of the second moving node. The information related to the position of the first moving node is calculated according to formulas (1) to (3):
[0062] X i (t) = (x i ,y i ) (1)
[0063]
[0064]
[0065] Where X i (t) is the position of node i at the current time t, is the speed (v) of node i at time t+Δt passing through t i,x ,v i,y ) calculated estimated position, d ij Define the Euclidean distance between vehicle nodes i and j.
[0066] In this embodiment, the driving direction of the first motion node is taken into consideration to maintain the relative stability of the motion cluster. i Definition: Only the first motion nodes with the same driving direction at the current moment can be classified into the same motion cluster. The driving direction of the first motion node in the motion cluster is calculated according to formula (4):
[0067]
[0068] where α ij is a binary decision variable, if and only if the first motion nodes i and j have the same driving direction α ij =1, otherwise α ij = 0. Finally, the first moving node with the same driving direction will continue to perform clustering according to the communication range of the nearby second moving node, and calculate the communication range of the second moving node in the moving cluster according to formula (5):
[0069]
[0070] Where R w is the communication range of the second moving node w, β wiis a binary decision variable about the first moving node i, used to determine whether the first moving node i is classified into cluster w. If and only if the second moving node has the same driving direction as the first moving node and is within the communication range of the current second moving node, β wi =1, otherwise β wi =0.
[0071] Through the above scheme, the first motion node in the industrial Internet is divided into motion clusters. The first motion node in the motion cluster communicates with the base station node through the second motion node, realizing the distributed trust management of the industrial Internet and providing a data basis for the subsequent application of the random forest algorithm.
[0072] In some embodiments, step 101 specifically further includes:
[0073] In response to determining that the same first motion node belongs to different second initial motion clusters at the same time, the communication ranges of the second motion nodes in the second initial motion cluster including the same first motion node are sorted, and the first motion node is divided into the second initial motion cluster corresponding to the larger communication range.
[0074] In the above scheme, since the first motion nodes are randomly distributed and move independently, the first motion nodes may face the situation of satisfying two clusters at the same time. The motion cluster division situation when the first motion node satisfies two motion clusters at the same time is calculated according to formula (6):
[0075]
[0076] Among them, α wj is a binary decision variable, if and only if the first motion node i and the second motion node w have the same driving direction α wj =1, otherwise α wj = 0, the driving direction of the first motion node i is d wj Definition, R w is the communication range of the second motion node w, and the variable of the second motion node in another motion cluster is represented by the subscript (w+1).
[0077] In some embodiments, Figure 3 As shown, step 103 specifically includes:
[0078] Step 1031, obtaining a trust attribute set in the trust evidence set;
[0079] Step 1032, arranging the trust attribute set according to the decision tree to obtain an arrangement result;
[0080] Step 1033: divide the first motion node into high-trust motion nodes and low-trust motion nodes based on the arrangement result, use the classification result as the decision result, and complete the trust evaluation.
[0081] In the above scheme, after collecting trust evidence, this embodiment uses the random forest algorithm to evaluate the reliability of the node. Random forest is an ensemble learning algorithm that uses the sampling technique with replacement to extract a random sample set from the original sample set to construct a single decision. Using the random feature subspace, a split node is selected at each node of the decision tree for splitting. Finally, these decision trees are combined to generate the final classification result through aggregation.
[0082] Specifically, the first motion node can be divided into several motion clusters, the total number of motion clusters is P, and Num is the total number of nodes in different motion clusters. w For the motion cluster w, each node independently trains CART (Classification And Regression Tree) as a decision tree. w Represents the set of decision trees in motion cluster w. S1 w and F1 w From S and T1 respectively w The training data and attribute sets extracted from .
[0083] Num={num1,num2,...,num P}
[0084]
[0085]
[0086]
[0087] Note S1 w The N samples in are randomly selected from the training set S with replacement, so each tree has a different training set, and S1 w Some samples of will appear multiple times, m is determined by the size of the attribute set F.
[0088] Once the first motion node V x If an accident e is found, it will send information about the accident to other first motion nodes in the same motion cluster. e Then the other first motion nodes V in the motion cluster i ,i=1,2,...,x-1,x+1,...num w Independent execution based on F i wThe decision-making process divides the first motion node into a high-trust motion node and a low-trust motion node.
[0089] Through the above scheme, the random forest algorithm is used to classify the first motion node in each motion cluster to obtain high-trust motion nodes and low-trust motion nodes, which can limit the overfitting problem of distributed trust management decision learning without increasing the trust evaluation bias.
[0090] In some embodiments, Figure 4 As shown, the decision tree construction process in step 103 includes:
[0091] Step 301, obtaining a training set by using a resampling method based on the trust evidence set;
[0092] Step 302, randomly selecting the trust evidence set in the training set as an attribute set;
[0093] Step 303, calculating the Gini coefficient of each attribute in the attribute set;
[0094] Step 304: Select attributes from the attribute set based on the size of the Gini coefficient and construct the decision tree in a tree structure.
[0095] In the above scheme, CART is a binary tree, and each split will produce two child nodes. CART trees are divided into classification trees and regression trees. Classification trees are mainly for target scalars as classification variables. In this embodiment, the classification variables are selected as high-trust motion nodes and low-trust motion nodes. If it is a classification tree, a split attribute that can minimize the GINI value of the node after splitting will be selected. CART, like other decision tree algorithms, needs to be pruned to prevent the algorithm from overfitting and thus ensure the generalization performance of the algorithm.
[0096] The calculation formula of the Gini coefficient of each attribute in the attribute set is shown in formula (7):
[0097]
[0098] Among them, Gini(p) is the Gini coefficient of the attribute, p k is the probability that the first motion node corresponding to the attribute belongs to the kth classification, m is the number of attributes, k=2 in this embodiment, when k=1, the first motion node is a high-trust motion node, when k=2, the second motion node is classified as a low-trust motion node.
[0099] Through the above scheme, the constructed decision tree can be used as a part of the random forest algorithm to provide a classification tool for the trust evaluation of the first motion node in the motion cluster.
[0100] In some embodiments, step 106 specifically includes:
[0101] The final trust value of the first motion node is calculated according to the following formula:
[0102]
[0103] in, is the final trust value, is the number of all the high-trust motion nodes in the decision result, is the number of the low-trust motion nodes, and δ is a predetermined positive integer.
[0104] In the above scheme, the second motion node combines the decision results of the first motion node in the motion cluster to obtain the decision result The encrypted data is uploaded to the base station node, which will perform a majority voting process to obtain the classification result of the first moving node x at the current moment. In the subsequent trust accumulation process, a sliding time window is used to perform weighted summation of the trust values from time t-(h-1) to time t, and V is obtained according to formula (8): x The final trust value of:
[0105]
[0106] in and are the number of high-trust motion node classification results and the number of low-trust motion node classification results in the decision result VR, respectively. δ is a positive integer penalty factor. The value of It rises sharply with the increase of , reflecting the characteristic of severe punishment for low-trust movement nodes.
[0107] Through the above scheme, the base station node calculates the final trust value of each first motion node in the motion cluster within a period of time through a sliding time window, reducing the impact of decision tree training failure in the random forest algorithm, thereby improving the accuracy of trust assessment of motion nodes in the industrial Internet network.
[0108] In some embodiments, Figure 5 As shown, step 106 specifically includes:
[0109] Step 1061, the base station node performs a majority voting process on the decision result to obtain a classification result;
[0110] Step 1062: Use a sliding time window to perform weighted summation on the first h classification results, where h is the predetermined number of time points in the sliding time window.
[0111] In the above scheme, the trust evaluation result V of the first motion node is x Expressed as where num w is the size of motion cluster w, and the encrypted decision result of the first motion node in motion cluster w is expressed as The base station node performs a majority voting process on the encryption decision results at h moments, and the classification result is expressed as Among them, t is the current time.
[0112] Through the above solution, a data basis is provided for subsequent base station nodes to calculate the final trust value of each first motion node in the motion cluster within a period of time.
[0113] In some embodiments, the trust evidence set in step 1031 includes: motion node trust evidence, data trust evidence, and link trust evidence.
[0114] In the above scheme, trust evidence is the basis of decision tree learning and plays a vital role in training the decision tree structure and testing the accuracy of the decision tree. This embodiment designs trust evidence from three aspects: nodes, data, and links to fully consider the reliability of the communication system.
[0115] 1) Based on node trust evidence: and Used to measure the selfishness and honesty of vehicle nodes respectively
[0116]
[0117]
[0118] Where H is the number of messages sent by neighboring nodes about accident m, is the number of messages sent by vehicle i to its neighbor nodes, and M is the total number of accidents on the road. is the total number of honest messages sent by vehicle i to its neighbors, is the total number of messages sent by vehicle i.
[0119] 2) Data-based trust evidence: Based on spatiotemporal correlation, the quality of interactive information can be measured by its correlation with other information on the same incident, as measured by TEV3 i definition:
[0120]
[0121] where v i m is the value of information sent by node i about event m.
[0122] 3) Trust evidence based on links: Link quality directly affects the accuracy of information exchange between vehicles. For example, in man-in-the-middle attacks, attackers intercept normal network traffic data by attacking communication links and perform data tampering and sniffing. Therefore, we define and TEV5 i Measuring link quality:
[0123]
[0124]
[0125] in and They represent link transmission delay and link usage frequency, respectively. delay (n i ) is vehicle n i With adjacent nodes n j The link transmission delay between use (n i ) is n i The link usage frequency.
[0126] Through the above scheme, a data foundation is provided for decision tree learning in the random forest algorithm.
[0127] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.
[0128] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0129] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a trust management device for the industrial Internet.
[0130] refer to Figure 6 , the trust management device of the industrial Internet comprises:
[0131] The clustering module 601 is configured to cluster the first motion nodes in the industrial Internet according to the proximity principle to obtain motion clusters;
[0132] A trust evidence acquisition module 602 is configured to acquire a trust evidence set of the first motion node;
[0133] A trust evaluation module 603 is configured to perform a trust evaluation on the motion cluster based on the trust evidence set by using a decision tree pre-constructed by a random forest algorithm to obtain a decision result corresponding to the motion cluster;
[0134] The information encryption module 604 is configured to encrypt the decision result using an elliptic curve and a hash function to obtain an encrypted decision result;
[0135] An information sending module 605 is configured to send the encryption decision result to a base station node through a second motion node in the motion cluster;
[0136] The trust value calculation module 606 is configured to perform weighted summation on the encryption decision results through the base station node using a sliding time window to obtain the final trust value of the first motion node, thereby completing motion trust management in the industrial Internet.
[0137] In some embodiments, the clustering module 601 is specifically configured as follows:
[0138] Determine the second motion node in the industrial Internet through the communication link of the base station node;
[0139] Calculating the Euclidean distance between the first motion node and the second motion node, and taking all the first motion nodes whose Euclidean distance is less than a predetermined value as the initial motion cluster of the second motion node;
[0140] Obtaining the driving directions of all the first motion nodes in the initial motion cluster;
[0141] Extracting the first motion nodes with the same driving direction in the initial motion cluster to obtain a first initial motion cluster;
[0142] Acquire the communication range of the second moving node;
[0143] The first motion nodes that are smaller than the communication range in the first initial motion cluster are extracted to obtain a second initial motion cluster, and the second initial motion cluster is used as the motion cluster.
[0144] In some embodiments, the clustering module 601 is further configured to:
[0145] In response to determining that the same first motion node belongs to different second initial motion clusters at the same time, the communication ranges of the second motion nodes in the second initial motion cluster including the same first motion node are sorted, and the first motion node is divided into the second initial motion cluster corresponding to the larger communication range.
[0146] In some embodiments, the trust assessment module 603 is specifically configured to:
[0147] Obtaining a trust attribute set in the trust evidence set;
[0148] Arranging the trust attribute set according to the decision tree to obtain an arrangement result;
[0149] The first motion nodes are divided into high-trust motion nodes and low-trust motion nodes based on the arrangement result, and the classification result is used as the decision result to complete the trust evaluation.
[0150] In some embodiments, the decision tree construction unit in the trust evaluation module 603 is configured to:
[0151] A training set is obtained by using a resampling method based on the trust evidence set;
[0152] Randomly selecting the trust evidence set in the training set as an attribute set;
[0153] Calculating the Gini coefficient of each attribute in the attribute set;
[0154] Based on the size of the Gini coefficient, attributes are selected from the attribute set to construct the decision tree in a tree structure.
[0155] In some embodiments, the trust value calculation module 606 is specifically configured to:
[0156] The final trust value of the first motion node is calculated according to the following formula:
[0157]
[0158] in, is the final trust value, is the number of all the high-trust motion nodes in the decision result, is the number of the low-trust motion nodes, and δ is a predetermined positive integer.
[0159] In some embodiments, the trust value calculation module 606 is further configured to:
[0160] The base station node performs a majority voting process on the decision result to obtain a classification result;
[0161] A sliding time window is used to perform weighted summation on the first h classification results, where h is the predetermined number of time points in the sliding time window.
[0162] In some embodiments, the trust evidence set in the trust evidence acquisition module 602 includes: motion node trust evidence, data trust evidence, and link trust evidence.
[0163] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.
[0164] The device of the above embodiment is used to implement the corresponding trust management method of the industrial Internet in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.
[0165] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the trust management method for the industrial Internet described in any of the above embodiments is implemented.
[0166] Figure 7 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.
[0167] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.
[0168] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.
[0169] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.
[0170] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).
[0171] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).
[0172] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.
[0173] The electronic device of the above-mentioned embodiment is used to implement the corresponding trust management method of the industrial Internet in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.
[0174] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the trust management method of the industrial Internet as described in any of the above embodiments.
[0175] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.
[0176] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the trust management method of the industrial Internet as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.
[0177] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.
[0178] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.
[0179] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.
[0180] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.
Claims
1. A trust management method for industrial Internet, characterized in that: The method comprises: Clustering the first motion nodes in the industrial Internet according to the proximity principle to obtain motion clusters; wherein the first motion nodes are unmanned vehicles; Acquire a trust evidence set of the first motion node; wherein the trust evidence set includes: motion node trust evidence, data trust evidence and link trust evidence; Based on the trust evidence set, a decision tree pre-constructed using a random forest algorithm is used to perform a trust evaluation on the motion cluster to obtain a decision result corresponding to the motion cluster; Encrypting the decision result using an elliptic curve and a hash function to obtain an encrypted decision result; Sending the encryption decision result to the base station node through a second motion node in the motion cluster; wherein the second motion node is an unmanned vehicle equipped with a communication route; The base station node uses a sliding time window to perform weighted summation on the encryption decision results to obtain a final trust value of the first motion node, thereby completing motion trust management in the industrial Internet; The trust evaluation of the motion cluster is performed based on the trust evidence set by using a decision tree pre-constructed by a random forest algorithm to obtain a decision result corresponding to the motion cluster, including: Obtaining a trust attribute set in the trust evidence set; Arranging the trust attribute set according to the decision tree to obtain an arrangement result; Based on the arrangement result, the first motion node is divided into a high-trust motion node and a low-trust motion node, and the classification result is used as the decision result to complete the trust evaluation; The decision tree construction process includes: A training set is obtained by using a resampling method based on the trust evidence set; Randomly selecting the trust evidence set in the training set as an attribute set; Calculating the Gini coefficient of each attribute in the attribute set; Based on the size of the Gini coefficient, attributes are selected from the attribute set to construct the decision tree in a tree structure.
2. The method according to claim 1, characterized in that The first motion nodes in the industrial Internet are clustered according to the proximity principle, and the obtained motion clusters include: Determine the second motion node in the industrial Internet through the communication link of the base station node; Calculating the Euclidean distance between the first motion node and the second motion node, and taking all the first motion nodes whose Euclidean distance is less than a predetermined value as the initial motion cluster of the second motion node; Obtaining the driving directions of all the first motion nodes in the initial motion cluster; Extracting the first motion nodes with the same driving direction in the initial motion cluster to obtain a first initial motion cluster; Acquire the communication range of the second moving node; The first motion nodes that are smaller than the communication range in the first initial motion cluster are extracted to obtain a second initial motion cluster, and the second initial motion cluster is used as the motion cluster.
3. The method according to claim 2, characterized in that The clustering of the first motion nodes in the industrial Internet according to the proximity principle also includes: In response to determining that the same first motion node belongs to different second initial motion clusters at the same time, the communication ranges of the second motion nodes in the second initial motion cluster including the same first motion node are sorted, and the first motion node is divided into the second initial motion cluster corresponding to the larger communication range.
4. The method according to claim 1, characterized in that: The step of performing weighted summation of the decision results by the base station node using a sliding time window to obtain a final trust value of the first motion node includes: The final trust value of the first motion node is calculated according to the following formula: in, is the final trust value, is the number of all the high-trust motion nodes in the decision result, is the number of the low-trust motion nodes, and δ is a predetermined positive integer.
5. The method according to claim 1, characterized in that The step of performing weighted summation of the decision results by the base station node using a sliding time window also includes: The base station node performs a majority voting process on the decision result to obtain a classification result; A sliding time window is used to perform weighted summation on the first h classification results, where h is the predetermined number of time points in the sliding time window.
6. A trust management device for industrial Internet, characterized in that: include: The clustering module is configured to cluster the first motion nodes in the industrial Internet according to the proximity principle to obtain motion clusters; wherein the first motion nodes are unmanned vehicles; A trust evidence acquisition module is configured to acquire a trust evidence set of the first motion node; wherein the trust evidence set includes: motion node trust evidence, data trust evidence and link trust evidence; A trust evaluation module is configured to perform a trust evaluation on the motion cluster based on the trust evidence set by using a decision tree pre-constructed by a random forest algorithm to obtain a decision result corresponding to the motion cluster; An information encryption module is configured to encrypt the decision result using an elliptic curve and a hash function to obtain an encrypted decision result; An information sending module is configured to send the encryption decision result to a base station node through a second moving node in the moving cluster; wherein the second moving node is an unmanned vehicle equipped with a communication route; A trust value calculation module is configured to perform weighted summation of the encryption decision results by the base station node using a sliding time window to obtain a final trust value of the first motion node, thereby completing motion trust management in the industrial Internet; The trust assessment module is specifically configured as follows: Obtaining a trust attribute set in the trust evidence set; Arranging the trust attribute set according to the decision tree to obtain an arrangement result; Based on the arrangement result, the first motion node is divided into a high-trust motion node and a low-trust motion node, and the classification result is used as the decision result to complete the trust evaluation; The decision tree building unit in the trust assessment module is configured as follows: A training set is obtained by using a resampling method based on the trust evidence set; Randomly selecting the trust evidence set in the training set as an attribute set; Calculating the Gini coefficient of each attribute in the attribute set; Based on the size of the Gini coefficient, attributes are selected from the attribute set to construct the decision tree in a tree structure.
7. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method according to any one of claims 1 to 5 when executing the computer program.
