Fuzzy trust management method and device for traffic information sharing based on blockchain

Abstract

The present invention discloses a fuzzy trust management method and device for sharing traffic information based on blockchain, comprising the following steps: 1. Building a behavior blockchain in the RSU network to store user behavior data. 2. Based on the on-chain data, the RSU interacting with the vehicle uses an intuitive fuzzy set to fuzzify the behavior and calculate the fuzzy trust level of the connected vehicle. 3. The RSU collects traffic information and performs the following information authenticity consensus process: 1) Selecting a consensus vehicle. 2) Granting vehicle consensus capabilities. 3) Information voting. 4) Behavior block generation. 5) Information release. 4. The RSU sets information feedback rules and uses fuzzy negative review reasoning to obtain the semantic label of the information. According to the semantic label, the RSU manages the information. The advantage of this method is that fuzzy intelligent technology is used in the management of the blockchain, which reduces the risk of leakage of vehicle privacy behavior, significantly reduces the calculation and communication costs of the interaction between the RSU and the vehicle, and has more real-time management.

Description

Technical Field

[0001] The present invention relates to the fields of vehicle networking, blockchain technology, fuzzy theory, and trust management. Specifically, it relates to a fuzzy trust management method and device for sharing traffic information based on blockchain. Background Art

[0002] Traffic conditions are constantly changing, and road conditions are complex and unpredictable. This creates a gap in information flow between vehicles traveling on a given road over a period of time. Failure to communicate road conditions to subsequent vehicles can lead to even greater congestion and traffic problems. Therefore, ensuring the authenticity of this information while sharing real-time traffic information with subsequent vehicles within a certain timeframe is a crucial component of achieving smooth traffic flow in smart cities.

[0003] On the other hand, with the widespread adoption of the Internet of Vehicles (IoV), vehicle privacy and information security are receiving increasing attention. Vehicles are vulnerable to eavesdropping attacks when sharing information and authenticating access to RSUs. Furthermore, the data transmitted contains sensitive behavioral data, making it susceptible to inference attacks. Therefore, protecting communication data is crucial.

[0004] Furthermore, while existing blockchain methods can share information by electing vehicles to participate in consensus, these consensus methods struggle to select vehicles within a local network of vehicles that understand road conditions, making it difficult to ensure the validity of the information. Furthermore, the computational cost of commonly used consensus algorithms is significant, making it impossible to guarantee the real-time nature of the information ultimately uploaded to the platform.

[0005] In response to the above problems, in the Internet of Vehicles environment, how to design appropriate blockchain and fuzzy trust solutions to reduce the complexity, computational latency and participation of consensus, and ensure the privacy and security of Internet of Vehicles vehicles in information services will be an issue that needs to be solved urgently. Summary of the Invention

[0006] The purpose of the present invention is to address the shortcomings of the existing technology and propose a fuzzy trust management method and device for traffic information sharing based on blockchain.

[0007] The object of the present invention is achieved through the following technical solutions: In the first aspect, the present invention provides a fuzzy trust management method for traffic information sharing based on blockchain, comprising the following steps:

[0008] (1) Vehicle authentication: Establish a behavior blockchain. When a user's vehicle is connected to the Internet of Vehicles, the RSU interacting with the vehicle verifies whether the behavior block stored in the connected vehicle is consistent with the behavior blockchain. If not, the vehicle will not be allowed to access the network.

[0009] (2) Fuzzy trust level and trust value calculation: Based on the user traffic information sharing behavior and consensus behavior stored on the chain, the RSU interacting with the vehicle will fuzzify the behavior of the connected vehicle, perform fuzzy reasoning based on the vehicle's trust behavior evaluation and malicious behavior evaluation, determine the fuzzy trust level and trust value of the connected vehicle, and issue a trust level certification;

[0010] (3) Information sharing: The RSU interacting with the vehicle sets three semantic thresholds and a 3-minute time limit to receive semantically similar information. If the time limit is reached or the number of semantically repeated information reaches the threshold, the following information authenticity consensus is initiated for all received shared information: 1) Select several more credible consensus vehicles based on the fuzzy trust level; 2) RSU assigns different voting weights and consensus completion thresholds to vehicles with different fuzzy trust levels; 3) Each vehicle participates in the information consensus according to its own voting weight; 4) Information sharing and consensus voting behaviors are recorded to form a block; 5) Information that is recognized as credible by consensus will be published by RSU in the RSU network, and untrustworthy information will be discarded;

[0011] (4) Information feedback: Set information feedback rules and use fuzzy negative review reasoning to obtain the semantic label of the information; based on the semantic label, the RSU network reminds the information sharer or forces the information to be removed from the shelves; the demander or sharer independently initiates the information update consensus based on the prompt, and updates or cancels the road condition information.

[0012] Furthermore, in step (1), the detailed steps of behavior blockchain construction and vehicle authentication are as follows:

[0013] (1-1) Establish a behavior blockchain in the network composed of RSUs to store users’ road condition information sharing behaviors and consensus behaviors.

[0014] (1-2) When a user's vehicle is connected to the Internet of Vehicles, the RSU interacting with the vehicle verifies whether the behavior stored in the connected vehicle is consistent with the behavior data of the vehicle on the behavior chain, and issues an authentication certificate.

[0015] Furthermore, in step (2), the detailed steps of calculating the fuzzy trust level and trust value are as follows:

[0016] (2-1) Constructing a fuzzy trust model: The RSU first fuzzifies the historical behaviors of the vehicle and then centrally evaluates the trust level of the vehicle's historical behaviors, that is, creating the following five behavior models to form the fuzzy trust model U.

[0017] U={Sh,Co,Rm,Cf,Rp}

[0018] There are two types of behaviors that enhance vehicle trust:

[0019] 1) Information Sharing Sh: After sharing information, if the information is verified by other vehicles, the RSU will publish the information, and the update of the information is also considered as information sharing;

[0020] 2) Information consensus Co: Participate in information consensus and confirm the information shared by other vehicles;

[0021] There are three types of behaviors that reduce vehicle trust:

[0022] 1) Information removal (Rm): After a negative review of an information, it is recorded as an action after semantic reasoning and consensus on information removal.

[0023] 2) Information voting error Cf: The vehicle makes an error in judging the correctness of the information during the information voting process;

[0024] 3) Information duplication Rp: For a certain road condition, different vehicles repeatedly publish information. If this occurs in the RSU, it will be recorded and the information will be removed after being reported by users.

[0025] (2-2) Fuzzy behavior evaluation calculation: After the vehicle interacts with the RSU, the RSU constructs the intuitive fuzzy number Y based on the historical behavior of the connected vehicle: (Sh (μ,v) ,Co (μ,v) ,Rm (μ,v) ,Cf (μ,v) ,Rp (μ,v) ), μ and v represent the membership and non-membership of the behavior respectively. The five behaviors defined in step (2-1) are composed of five intuitionistic fuzzy sets, that is, the number of intuitionistic fuzzy sets n = 5; the intuitionistic fuzzy matrix I of vehicle behavior Y It is expressed as the following formula:

[0026] I Y =[(μ Sh ,v Sh ),(μ Co ,v Co )],[(μ Rm ,v Rm ),(μ Cf ,v Cf ),(μ Rp ,v Rp )] n ,n=5

[0027] Re-define the behavioral factor weight matrix for behavioral evaluation matrix operations; define the composite operation method of the fuzzy matrix For the weight matrix X and fuzzy matrix R nm The calculation is as follows:

[0028]

[0029] Among them, R nmn and m are the row and column sizes of the matrix, X j Represents the data of the jth column in the weight matrix, R ij Represents the data at the i-th row and j-th column of the fuzzy matrix;

[0030] Set the trust behavior weight set A = [0.4, 0.6], from left to right are the behavior weights of information sharing and information consensus. 2×5 Perform matrix operations to obtain the evaluation vector C of trust behavior, which is as follows:

[0031]

[0032] The malicious behavior weight matrix B = [0.3, 0.4, 0.3], from left to right, is the weight of information removal, information confirmation error, and information duplication; it is combined with the malicious behavior fuzzy membership matrix R 3×5 Perform matrix operations to obtain the evaluation vector E of malicious behavior. The formula is as follows:

[0033]

[0034] The trust behavior evaluation vector C is composed of five matrix calculation values ​​C = [c1, c2, c3, c4, c5]; the matrix calculation results of C correspond to the five trust levels of "not positive", "general", "relatively positive", "positive", and "enthusiastic" respectively; the malicious behavior evaluation vector E is set to five fuzzy calculation results E = [e1, e2, e3, e4, e5], corresponding to the evaluation labels "fewer", "general", "more", "many", and "very many"; the RSU makes a trust assessment based on the membership of the trust evaluation vector using the maximum membership principle; if the membership calculation results are the same, the non-membership is compared, and the one with the smaller non-membership is taken to assign the current vehicle the fuzzy behavior evaluation.

[0035] (2-3) Fuzzy reasoning and trust value calculation: Fuzzy control rules are set according to the trust behavior evaluation and malicious behavior evaluation, and fuzzy reasoning is performed on the fuzzy behavior evaluation.

[0036] The fuzzy membership function of the trust level is established using the trapezoidal membership function, and its domain X cre is the vehicle trust value [0,100]; RSU performs multi-label fuzzy reasoning calculation through the activated fuzzy control rules to obtain the fuzzy trust domain U(x): U(x)=c i ∧e j ∧X(a), where ∧ is the smallest value and X(a) is the fuzzy membership function. Defuzzify the trust region to obtain the vehicle trust value X.

[0037] Furthermore, in step (3), the detailed steps of information sharing are as follows:

[0038] (3-1) Information collection: After the sharer uploads the traffic information, the RSU sets three semantic thresholds and a 3-minute time limit based on the traffic information released by the sharer to receive information with similar semantics;

[0039] (3-2) Consensus vehicle selection: Select other vehicles near the road to form a temporary local vehicle network to achieve information authenticity consensus. When selecting a consensus vehicle, if the fuzzy trust level of the vehicles near the road is the same, compare their trust values ​​X to select the vehicle with the highest trust.

[0040] RSU counts the total number of behavior chain blocks n generated within ten minutes bc , the total number of vehicles connected to the local vehicle network within ten minutes n c And the average number of vehicle network nodes n within ten minutes v ; According to the Poisson distribution, calculate the upper limit k of the consensus vehicle selection, that is, calculate the value p(k) of the maximum probability k under the Poisson distribution:

[0041]

[0042] Based on the number of vehicles connected to the local vehicle network, a number of vehicles is selected that does not exceed the maximum probability k in the Poisson distribution, and the maximum number of consensus vehicles selected does not exceed 8. The above vehicle selection is carried out in the following way: the RSU uses a sliding window method to select vehicles for interaction, which sets four windows with a time window limit of 30s, and gradually sends consensus information until the number of vehicle nodes willing to participate in the consensus vote reaches 8;

[0043] (3-3) Vehicle Consensus Capability Assignment: When initiating consensus, RSU assigns different voting capabilities and consensus completion thresholds to vehicles of different fuzzy trust levels; the details are as follows:

[0044] 1) Set corresponding voting weights for different fuzzy trust levels. When voting for information on other vehicles, the vehicle's voting weight participates in the road condition information voting; the fuzzy trust level "untrustworthy" has a weight of 1, "not very trustworthy" has a weight of 2, "somewhat trustworthy" has a weight of 3, "trustworthy" has a weight of 4, and "very trustworthy" has a weight of 5;

[0045] 2) “Highly trustworthy” vehicles can directly release road condition information without information consensus;

[0046] 3) For information released by vehicles with “relatively credible” and “credible” trust values, the approval weight of voting vehicles must accumulate to 3 to achieve information consensus;

[0047] 4) Information released by “untrusted” and “less trustworthy” vehicles requires the approval weight of voting vehicles to accumulate to 5 to achieve information consensus;

[0048] (3-4) Information authenticity consensus: The temporary local vehicle network selected in (3-2) reaches information authenticity consensus according to the following steps:

[0049] 1) Information voting: Each vehicle participates in information consensus according to its own voting weight; information voting only considers the information recognition among consensus vehicles; the voting weight of the vehicle node that recognizes the authenticity of the information is Cre con , obtain the weight of the information consent node; the voting process formula is as follows:

[0050] G=∑Cre Con

[0051] Where G is the consensus success threshold of the information sharing vehicles; if the consensus vote is successful, the vehicle nodes that do not recognize the information are traced back; if the vote does not reach the consensus success threshold, it means that the information release has failed, and the nodes that recognize the information are traced back;

[0052] 2) Behavior block generation: All vehicles send digital signatures Sig j , verifying the pseudonymous identity of the actor; the RSU network records information sharing and consensus voting behaviors, forming a block for subsequent tracing and query; three key information are recorded on the chain: 1) the address of the vehicle participating in the information release; 2) whether the information release is successful; 3) the address of the vehicle participating in the information voting; in accordance with the needs of trust behavior, the established block structure is shown in the following formula:

[0053] BC cre =ID d ,preHash,nonce,time,bool j ,Sig j |Body

[0054] Among them, ID d It is the id of the block, indicating that the block is the dth block in the chain. preHash is the hash value of the previous block; nonce is a random number; time is the timestamp of the creation time; bool j Indicates whether the sharing behavior can be successfully released after the authentication of surrounding vehicles. If the consensus is successful, bool j If the item is set to 1, the consensus fails, bool j Item is set to 0; Sig j It is the pseudonymous signature of the vehicle that released the information; the Body records the digital signatures and fuzzy trust levels of the vehicles involved in this information authentication, including all vehicles that approve and disapprove the information;

[0055] 3) Information Release and Block Storage: RSU will publish information that is recognized as credible on the network, and unreliable information will be discarded. Shared vehicles and authenticated vehicles must store the block locally.

[0056] Furthermore, in step (4), the steps of information feedback are as follows:

[0057] (4-1) Information evaluation: Users evaluate information sharing, which is divided into positive and negative reviews. RSU collects the number of negative reviews of the information;

[0058] (4-2) Setting fuzzy rules to remove information: If the published information exceeds the time window t or the number of demanding users exceeds the threshold d, the RSU performs fuzzy negative review reasoning on the information and stores its semantic label on the RSU side.

[0059] The specific process of fuzzy negative review reasoning is as follows: according to the trust level and the number of people who need information, a fuzzy negative review rule is established to calculate the fuzzy semantic label of the information; there are two input variables, one is the number of people who need information, and the other is the trust negative review ratio; here, a fuzzy membership function is established for the number of people who need information and the trust negative review ratio. The domain X is the number of people who request a certain information, and its semantic labels are FN 1 "less", FN 2 "less", FN 3 "More", FN 4 "More"; the domain X of the trust-to-criticism ratio is the trust value N of the critic cre Trust value N with all demanders total The ratio of X = N cre / N total , establish semantic labels as FD 1 "Low", FD 2 "Lower", FD 3 Higher, FD 4 "high";

[0060] The following fuzzy inference rules are established based on the number of people who need information and the ratio of negative reviews:

[0061] 1. If the number of people who need information is small and the trust-to-negative ratio is low than the semantic tag has a small number of negative reviews;

[0062] 2. If the number of people who need information is large and the trust-to-negative ratio is low, than the semantic tag has more negative reviews;

[0063] 3. If the number of people who need information is small and the trust-to-negative ratio is high than the semantic tag has more negative reviews;

[0064] 4. If the number of people who need information is small and the trust-to-negative ratio is low, then the semantic tags are mostly negative;

[0065] 5. If the number of people who need information is small and the trust-to-negative ratio is high than the semantic tag is negative;

[0066] 6. If the number of people who need information is large and the trust-to-negative ratio is high than the semantic tag is negative;

[0067] After time t or the number of demanding users exceeds the threshold d, the semantic label of a certain road condition information is calculated; four linguistic variables "a", "b", "c", and "d" are defined, corresponding to "a few negative reviews", "many negative reviews", "mostly negative reviews", and "a flood of negative reviews"; and different management measures are set for different levels of negative reviews, specifically: "a few negative reviews" means no reminder, "many negative reviews" means reminder to modify, "mostly negative reviews" means reminder to the demander, and "a flood of negative reviews" means consensus to remove the information.

[0068] (4-3) Information Update, Expiration, and Delisting: After a period of time, traffic information is no longer real-time, and the demander or sharer initiates a consensus to update the information and confirm the expiration of the information. When the fuzzy reasoning obtains the semantic label "a flood of negative reviews", the RSU forcibly initiates a consensus to remove the information. The RSU reaches consensus according to the following steps:

[0069] 1) Witness selection: The method is the same as step (3-2) consensus vehicle selection;

[0070] 2) Consensus authority granting: The method is the same as step (3-3) vehicle consensus capability granting;

[0071] 3) Information voting: The method is the same as the information voting in steps (3-4);

[0072] 4) Behavior block generation: The method is the same as the behavior block generation in steps (3-4);

[0073] 5) Information update, invalidation or removal: Information will be modified, invalidated, deleted or removed from shelves according to vehicle behavior.

[0074] In a second aspect, the present invention also provides a fuzzy trust management device for traffic information sharing based on blockchain, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it implements the fuzzy trust management method for traffic information sharing based on blockchain.

[0075] In a third aspect, the present invention also provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the fuzzy trust management method for sharing traffic information based on blockchain is implemented.

[0076] Beneficial effects of the present invention:

[0077] (1) The trust calculation method based on fuzzy reasoning and the consensus voting method based on fuzzy trust levels reduce the risk of leakage of vehicle privacy behavior, shorten the consensus time of vehicle networks, and significantly reduce the network computing cost and communication cost of RSU and vehicle authentication.

[0078] (2) The vehicle node selection method based on Poisson distribution and fuzzy trust level effectively solves the problem of effective node selection, reduces the communication cost between RSU and vehicles, and improves the reliability of traffic consensus information.

[0079] (3) Real-time fuzzy reasoning based on user feedback makes the feedback of road condition information more in line with the needs of users and real-time Internet of Vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0081] Figure 1 This is a flowchart of a fuzzy trust management method for traffic information sharing based on blockchain;

[0082] Figure 2 Modeling graphs for the set of fuzzy membership functions of trust-sharing behavior;

[0083] Figure 3 Calculate vehicle fuzzy behavior level case for fuzzy matrix;

[0084] Figure 4 It is a case study of trust synthesis of fuzzy trust reasoning and reasoning results;

[0085] Figure 5 This is a structural diagram of a fuzzy trust management device for sharing traffic information based on blockchain in the present invention. DETAILED DESCRIPTION

[0086] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0087] Reference Figure 1 ,The fuzzy trust management method for traffic information sharing based on blockchain,includes the following steps:

[0088] (1) Behavior chain construction and vehicle authentication: A trust management blockchain (hereinafter referred to as the behavior chain) is established in the RSU network to store users' road condition information sharing behaviors and consensus behaviors. When a user's vehicle is connected to the Internet of Vehicles, the RSU uses a zero-knowledge proof method to verify whether the behavior stored in the connected vehicle is consistent with that on the behavior chain and issues an authentication certificate.

[0089] (2) Fuzzy trust level and trust value calculation: Based on the user behavior stored on the chain, RSU uses intuitionistic fuzzy sets to fuzzify the behavior and calculate the fuzzy trust level and trust value of the connected vehicle.

[0090] (3) Pseudonym registration and setting: After the vehicle is connected to the Internet of Vehicles, it obtains the linked pseudonym from the RSU and can apply to change the pseudonym at regular intervals. Each time it interacts with the RSU, the RSU searches the block for all pseudonym participation behaviors corresponding to the real address based on the user's pseudonym change record.

[0091] (4) Information Sharing: The RSU interacting with the vehicle sets a threshold for the number of duplicate messages. If the number of shared messages with semantic duplication reaches the threshold, the following information authenticity consensus is initiated for all received shared messages: 1) Select several more trustworthy consensus vehicles based on the fuzzy trust level; 2) The RSU assigns different voting weights and consensus completion thresholds to vehicles with different fuzzy trust levels; 3) Each vehicle participates in the information consensus based on its own voting weight; 4) The information sharing and consensus voting behaviors are recorded and formed into a block; 5) The RSU publishes the information that is recognized as trustworthy in the RSU network, and the untrustworthy information is discarded.

[0092] (5) Information feedback: Set information feedback rules and use fuzzy negative review reasoning to obtain the semantic label of the information; based on the semantic label, the RSU network reminds the information sharer or forces the information to be removed from the shelves; the demander or sharer independently initiates the information update consensus based on the prompt, and updates or cancels the road condition information.

[0093] In step (1), the detailed steps of behavior chain construction and vehicle access authentication are as follows:

[0094] (1-1) A trust management blockchain (hereinafter referred to as the behavior chain) is established in the RSU network to store users' road condition information sharing behaviors and consensus behaviors.

[0095] (1-2) When a user's vehicle is connected to the Internet of Vehicles, the RSU interacts with the vehicle, uses zero-knowledge proof to verify whether the currently stored behavior of the connected vehicle is consistent with the behavior chain, and issues an authentication certificate.

[0096] In step (2), the detailed steps of calculating the fuzzy trust level and trust value are as follows:

[0097] (2-1) Constructing a fuzzy trust model: The fuzzy trust model is composed of a variety of fuzzy behaviors. According to the evaluation dimension of the blockchain vehicle network on the vehicle trust level, the following five behavior models are created to form the trust factor set U.

[0098] U={Sh,Co,Rm,Cf,Rp}

[0099] There are two types of behaviors that enhance vehicle trust:

[0100] 1) Information Sharing Sh: After sharing information, the information is verified by other nodes and the information is successfully released. The update of information by other vehicles is also considered as information sharing.

[0101] 2) Information consensus Co: Participate in information consensus and confirm the information shared by others.

[0102] There are three types of behaviors that reduce vehicle trust:

[0103] 1) Information removal (Rm): After a negative review of an information, it is recorded as an action after semantic reasoning and consensus on information removal.

[0104] 2) Information voting error Cf: The vehicle makes an error in judging the correctness of the information during the information voting process.

[0105] 3) Information duplication Rp: For a certain road condition, a vehicle repeatedly publishes information. If the information already exists in the RSU network, the behavior is recorded and the information is removed.

[0106] (2-2) Establish fuzzy behavior model: Based on the stored user behavior, the vehicle behavior is modeled with a trapezoidal fuzzy membership function. The domain of the fuzzy set is the ratio of a certain vehicle's behavior to the total vehicle behavior, that is, the domain X is N action / N total .

[0107] According to the membership function and non-membership function of trust sharing behavior, the following five intuitionistic fuzzy sets are constructed: FA 1 , FA 2 , FA 3 , FA 4 , FA 5 , the domain is the proportion of sharing behavior in the total number of behaviors of the vehicle; according to these five fuzzy sets, a certain behavior data of the vehicle is represented as (μ, v), where μ is the membership value, v is the non-membership value, and μ+v≤1.

[0108] For example, taking the membership function of trust sharing behavior as an example, the following membership function is constructed:

[0109]

[0110]

[0111]

[0112] where v FA It is the segmentation point of each trapezoidal membership function.

[0113] Non-membership function v A Determined by the membership function, the formula is as follows:

[0114]

[0115] Then establish the hesitation level of behavioral assessment:

[0116] π A =1-μ A -v A

[0117] The above trapezoidal membership function, non-membership function and hesitation function are used to model information sharing behavior. and x Are the upper and lower limits of the domain, which represent the maximum percentage of the vehicle behavior in the chain. The intersection of the two sub-functions is the endpoint of the fuzzy partition interval. The calculation method is defined as v i 、v i+1 With v j 、v j+1 is the horizontal coordinate of the segment point of a certain oblique line segment of the trapezoidal membership function and the top and bottom of the trapezoid, b j with b i is the intercept of two intersecting oblique line segments. By dividing the intersection point, the intersection of each fuzzy set can form fuzzy intervals of different lengths, namely:

[0118] (2-3) Fuzzy behavior evaluation calculation: After the vehicle interacts with the RSU, the RSU constructs the intuitive fuzzy number Y based on the historical behavior of the connected vehicle: (Sh (μ,v) ,Co (μ,v) ,Rm (μ,v) ,Cf (μ,v) ,Rp (μ,v) ), μ and v represent the membership and non-membership of the behavior respectively. The five behaviors defined in step (2-1) are composed of five intuitionistic fuzzy sets. That is, the intuitionistic fuzzy matrix of vehicle behavior can be expressed as the following formula:

[0119] I Y =[(μ Sh ,v Sh ),(μ Co ,vCo )],[(μ Rm ,v Rm ),(μ Cf ,v Cf ),(μ Rp ,v Rp )] n ,n=5

[0120] Then define the behavior factor weight matrix for behavior evaluation matrix operation. Define the composite operation method of fuzzy matrix For the weight matrix X and fuzzy matrix R nm Calculate, R nm The n and m are the row and column sizes of the matrix, as shown below:

[0121]

[0122] Among them, R nm n and m are the row and column sizes of the matrix, X j Represents the data of the jth column in the weight matrix, R ij Represents the data at the i-th row and j-th column of the fuzzy matrix;

[0123] Set the trust behavior weight set A = [0.4, 0.6] and the vehicle fuzzy behavior membership matrix R 2×5 Perform matrix operations to obtain the evaluation vector C of trust behavior, which is as follows:

[0124]

[0125] Malicious behavior weight matrix B = [0.3, 0.4, 0.3], from left to right, represents the weights of information removal, information confirmation error, and information duplication. The matrix calculates the malicious behavior evaluation vector E, and the formula is as follows:

[0126]

[0127] Figure 3 As a specific implementation case of the above-mentioned fuzzy behavior evaluation calculation, the fuzzy matrix calculation is performed through the intuitionistic fuzzy set to obtain the fuzzy behavior level of the vehicle's trustworthy behavior and malicious behavior.

[0128] The trust behavior evaluation vector C consists of five matrix calculations: C = [c1, c2, c3, c4, c5]. The matrix calculation results of C correspond to the five trust levels: "unenthusiastic," "general," "relatively proactive," "active," and "enthusiastic." The malicious behavior evaluation vector E is composed of five fuzzy calculation results: E = [e1, e2, e3, e4, e5], corresponding to the evaluation labels "few," "general," "more," "many," and "very many." The RSU uses the maximum membership principle to make a trust assessment based on the membership data values ​​of the trust evaluation vector. If the membership calculation results are the same, the non-membership is compared and the one with the smaller non-membership is selected to assign the current vehicle the fuzzy behavior evaluation.

[0129] (2-4) Fuzzy reasoning and trust value calculation: According to the trust behavior evaluation and malicious behavior evaluation, the fuzzy control rules in Table 1 are set, and fuzzy reasoning is performed on the fuzzy behavior evaluation.

[0130] Table 1 Fuzzy trust level reasoning table

[0131]

[0132] Use the trapezoidal membership function (the membership function type is consistent with the vehicle behavior) to establish the fuzzy membership function of the trust level, the domain X cre is the vehicle trust value [0,100]. RSU performs fuzzy reasoning through the activated fuzzy control rules, defines the fuzzy trust function, and calculates the trust value by taking the union. The fusion result formula of fuzzy reasoning is as follows:

[0133]

[0134] Take the fuzzy membership function of the evaluated reasoning label a and the label weight of the fuzzy behavior, that is, calculate the trust reasoning result U(x) of multiple behaviors: U(x) = c i ∧e j ∧X(a). This formula calculates the minimum of the two types of behavior weights in the fuzzy trust membership function curve to obtain the union of the fuzzy trust function U(x). Based on this, the area center method is used to obtain the defuzzified trust value X.

[0135] Figure 4 This is a specific case of performing fuzzy reasoning through fuzzy behavior evaluation to obtain the union of fuzzy trust functions U(x).

[0136] In step (3), the detailed steps for pseudonym registration and setting are as follows:

[0137] (3-1) Pseudonym Registration and Setup: When a vehicle connects to the IoV, it receives a temporary pseudonym from the RSU, which is used for a specific IoV activity. This pseudonym is valid for 10 minutes, and if it expires, a new pseudonym must be requested. Connecting to the IoV using a pseudonym ensures the security of the vehicle's location and privacy.

[0138] In step (4), the detailed steps of information sharing are as follows:

[0139] (4-1) Information Collection: After the sharer uploads the road condition information, the RSU interacting with the vehicle sets three semantic thresholds and a 3-minute time limit to receive information with similar semantics.

[0140] (4-2) Consensus vehicle selection: Select other vehicles near the road to form a temporary network to reach a consensus on the authenticity of the information. When selecting a consensus vehicle, if the fuzzy trust level of nearby vehicles is the same, their trust values ​​X are compared to select the vehicle with the highest trust.

[0141] RSU counts the total number of behavior chain blocks n generated within ten minutes bc , the total number of vehicles connected to the local vehicle network within ten minutes n c And the average number of vehicle network nodes n within ten minutes v According to the Poisson distribution, the upper limit k of consensus vehicle selection is calculated, that is, the value of the maximum probability k under the Poisson distribution is calculated:

[0142]

[0143] Based on the number of vehicles connected to the local vehicle network, a number of vehicles is selected that does not exceed the maximum probability k in the Poisson distribution, and the maximum number of consensus vehicles selected does not exceed 8. The above vehicle selection is carried out in the following way: the RSU uses a sliding window method to select vehicles for interaction. It sets four windows with a time window limit of 20 seconds and gradually sends consensus information until the number of vehicle nodes willing to participate in the consensus vote reaches 8.

[0144] (4-3) Vehicle consensus capability assignment: When initiating consensus, RSU assigns different voting capabilities and consensus completion thresholds to vehicles with different fuzzy trust levels. The details are as follows:

[0145] 1) Set voting weights for different fuzzy trust levels. When voting for information about other vehicles, the vehicle's voting weight participates in the road condition information voting. The fuzzy trust level "untrustworthy" has a weight of 1, "not very trustworthy" has a weight of 2, "somewhat trustworthy" has a weight of 3, "trustworthy" has a weight of 4, and "very trustworthy" has a weight of 5.

[0146] 2) “Highly trustworthy” vehicles can directly release road condition information without reaching an information consensus.

[0147] 3) For information released by vehicles with “relatively credible” and “credible” trust values, the approval weight of voting vehicles must accumulate to 3 to achieve information consensus.

[0148] 4) For information released by “untrusted” and “less trustworthy” vehicles, the approval weight of voting vehicles must accumulate to 5 to achieve information consensus.

[0149] (4-4) Information authenticity consensus: The temporary vehicle consensus network selected in (4-2) performs information authenticity consensus according to the following steps:

[0150] 1. Information voting. Each vehicle participates in information consensus according to its own voting weight. Information voting only considers the information recognition among consensus vehicles. The voting weight of the vehicle node that recognizes the authenticity of the information is Cre con , get the weight of the information agreeing node. The voting process formula is as follows:

[0151] G=∑Cre Con

[0152] Where G is the consensus success threshold for information-sharing vehicles. If the consensus vote is successful, the nodes that did not approve the information are traced back. If the vote does not reach the consensus success threshold, the information release fails, and the nodes that approved the information are traced back.

[0153] 2. Behavior block generation. All vehicles send digital signatures Sig j , verifying the pseudonymous identity of the actor. After receiving the relevant data, the elected edge node in the RSU network records the information sharing and consensus voting behavior, forming a block for subsequent tracing and querying. Three key pieces of information are recorded on the chain: 1) The address of the vehicle participating in the information release; 2) Whether the information release was successful; and 3) The address of the vehicle participating in the information vote. Based on the requirements of trust behavior, the established block structure is as follows:

[0154] BC cre =ID d ,preHash,nonce,time,bool j ,Sig j |Body

[0155] Among them, ID d It is the block ID, indicating that the block is the dth block in the chain. preHash is the hash value of the previous block. nonce is a random number. time is the timestamp of the creation time. bool j Indicates whether the sharing behavior can be successfully published on the platform after being authenticated by surrounding vehicles, a bool in the information publishing block j If the item is set to 1, it means the consensus is successful, but the bool of the block where the information fails to be published is jThe item is set to 0. Sig j This is the pseudonymous signature of the vehicle issuing the information. The body contains the digital signatures and fuzzy trust levels of all vehicles involved in the information authentication, including all vehicles that approved and rejected the information.

[0156] 3) Information Release and Block Storage. Information deemed credible by consensus is published by the RSU on the network, while unreliable information is discarded. Shared and authenticated vehicles must store the block locally.

[0157] In step (5), the steps of information feedback are as follows:

[0158] (5-1) Information evaluation: Users evaluate the information shared, which is divided into two types: positive and negative. RSU collects the number of negative reviews of the information.

[0159] (5-2) Setting fuzzy rules to remove information: If the published information exceeds the time window t or the number of users in need exceeds the threshold d, RSU performs fuzzy negative review reasoning on the information and stores its semantic label at the edge.

[0160] The specific process of fuzzy negative review reasoning is as follows: Based on the trust level and the number of people who need information, a fuzzy negative review rule is established, and the fuzzy semantic label of the information is calculated. There are two input variables: the number of people who need information and the trust negative review ratio. Here, a fuzzy membership function is established for the number of people who need information and the trust negative review ratio. The domain X is the number of people who request a certain information, and its semantic labels are FN 1 "less", FN 2 "less", FN 3 "More", FN 4 "More". The domain X of the trust-to-criticism ratio is the trust value N of the critic. cre Trust value N with all demanders total The ratio of X = N cre / N total , establish semantic labels as FD 1 "Low", FD 2 "Lower", FD 3 Higher, FD 4 "high".

[0161] The following fuzzy inference rules are established based on the demand and negative review ratio:

[0162] 1. If the number of people who need information is small and the trust-to-negative ratio is low than the semantic tag has a small number of negative reviews;

[0163] 2. If the number of people who need information is large and the trust-to-negative ratio is low, than the semantic tag has more negative reviews;

[0164] 3. If the number of people who need information is small and the trust-to-negative ratio is high than the semantic tag has more negative reviews;

[0165] 4. If the number of people who need information is small and the trust-to-negative ratio is low, then the semantic tags are mostly negative;

[0166] 5. If the number of people who need information is small and the trust-to-negative ratio is high than the semantic tag is negative;

[0167] 6. If the number of people who need information is large and the trust-to-negative ratio is high than the semantic tag is negative reviews.

[0168] After time t or when the number of demanding users exceeds a threshold d, calculate the semantic label for a piece of road condition information. Define four linguistic variables, "a," "b," "c," and "d," and set different management measures for different levels of negative reviews, as shown in the table below.

[0169] Table 2 Fuzzy negative review management mechanism

[0170] Language variables Semantic Tags Platform behavior mechanism a A few negative reviews No reminder b More negative reviews Make modification reminders c Mostly negative reviews Remind the demander d A flood of negative reviews Consensus on Information Removal

[0171] (5-3) Information Update, Expiration, and Delisting: After a period of time, traffic information is no longer real-time. The demander or sharer can initiate an update, confirm the expiration of the information, or, when fuzzy reasoning reaches the semantic label "a flood of negative reviews," force the RSU to initiate a consensus to remove the information. The RSU, the information demander, will select a vehicle node from the traffic information and reach an information update consensus to update or cancel the traffic information. This consensus is achieved through the following steps.

[0172] 1) Witness selection. The method is the same as step (4-2) consensus vehicle selection.

[0173] 2) Consensus authority granting: The method is the same as step (4-3) consensus authority granting.

[0174] 3) Information voting: The method is the same as step (4-4) 1. Information voting.

[0175] 4) Generate behavior blocks. The method is the same as step (4-4) 2. Generate behavior blocks.

[0176] 5) Information update, invalidation or removal: Modify, invalidate, delete or remove information in accordance with the agreed purpose.

[0177] Corresponding to the aforementioned embodiment of a fuzzy trust management method for sharing traffic information based on blockchain, the present invention also provides an embodiment of a fuzzy trust management device for sharing traffic information based on blockchain.

[0178] See also Figure 5An embodiment of the present invention provides a fuzzy trust management device for traffic information sharing based on blockchain, including a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it is used to implement a fuzzy trust management method for traffic information sharing based on blockchain in the above embodiment.

[0179] The embodiment of the fuzzy trust management device for sharing traffic information based on blockchain provided by the present invention can be applied to any device with data processing capabilities, and the device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory for execution. From the hardware level, if Figure 5 As shown, this is a hardware structure diagram of a fuzzy trust management device for sharing traffic information based on blockchain provided by the present invention, where any device with data processing capability is located. Figure 5 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0180] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0181] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present invention. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0182] An embodiment of the present invention also provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, a fuzzy trust management method for traffic information sharing based on blockchain in the above embodiment is implemented.

[0183] The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0184] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any modifications and changes made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A fuzzy trust management method for traffic information sharing based on blockchain, characterized in that: The steps include: (1) Vehicle authentication: Establish a behavior blockchain. When a user's vehicle is connected to the Internet of Vehicles, the RSU interacting with the vehicle verifies whether the behavior block stored in the connected vehicle is consistent with the behavior blockchain. If not, the vehicle will not be allowed to access the network. (2) Fuzzy trust level and trust value calculation: Based on the user traffic information sharing behavior and consensus behavior stored on the chain, the RSU interacting with the vehicle will fuzzify the behavior of the connected vehicle, perform fuzzy reasoning based on the vehicle's trust behavior evaluation and malicious behavior evaluation, determine the fuzzy trust level and trust value of the connected vehicle, and issue a trust level certification certificate; the detailed steps of fuzzy trust level and trust value calculation are as follows: (2-1) Constructing a fuzzy trust model: RSU first fuzzifies each historical behavior of the vehicle, and then centrally evaluates the trust level of the vehicle's historical behavior, that is, creates the following five behavior models to form a fuzzy trust model U; U={Sh,Co,Rm,Cf,Rp} There are two behaviors that enhance vehicle trust: 1) Information Sharing Sh: After sharing information, if the information is verified by other vehicles, the RSU will publish the information, and the update of the information will also be regarded as information sharing; 2) Information consensus Co: Participate in information consensus and confirm the information shared by other vehicles; There are three behaviors that reduce vehicle trust: 1) Information removal Rm: After the information is negatively reviewed, it is recorded as an action after semantic reasoning and consensus on information removal; 2) Information voting error Cf: The vehicle makes a mistake in judging the correctness of the information during the information voting process; 3) Information duplication Rp: For a certain road condition, different vehicles repeatedly publish information; if it exists in the RSU, after being reported by the user, the behavior will be recorded and the information will be removed; (2-2) Fuzzy behavior evaluation calculation: After the vehicle interacts with the RSU, the RSU constructs an intuitive fuzzy number Y based on the historical behavior of the connected vehicle: (Sh (μ,v) ,Co (μ,v) ,Rm (μ,v) ,Cf (μ,v) ,Rp (μ,v) ), μ and v represent the membership and non-membership of the behavior respectively. The five behaviors defined in step (2-1) are composed of five intuitionistic fuzzy sets, that is, the number of intuitionistic fuzzy sets n = 5; the intuitionistic fuzzy matrix I of vehicle behavior Y It is expressed as the following formula: I Y =[(μ Sh ,v Sh ),(m Co ,v Co )],[(m Rm ,v Rm ),(m Cf ,v Cf ),(m Rp ,v Rp )] n ,n=5 Redefine the behavior factor weight matrix for behavior evaluation matrix operations; define the composite operation method of the fuzzy matrix For the weight matrix X and fuzzy matrix R nm The calculation is as follows: Among them, R nm n and m are the row and column sizes of the matrix, X j Represents the data of the jth column in the weight matrix, R ij Represents the data of the fuzzy matrix at the i-th row and j-th column; Set the trust behavior weight set A = [0.4, 0.6], from left to right are the behavior weights of information sharing and information consensus; compare it with the trust behavior fuzzy membership matrix R 2×5 Perform matrix operations to obtain the evaluation vector C of trust behavior, the formula is as follows: Malicious behavior weight matrix B = [0.3, 0.4, 0.3], from left to right are the behavior weights of information removal, information confirmation error, and information duplication; compare it with the malicious behavior fuzzy membership matrix R 3×5 Perform matrix operations to obtain the evaluation vector E of malicious behavior. The formula is as follows: The trust behavior evaluation vector C is composed of five matrix calculation values ​​C = [c1, c2, c3, c4, c5]; the matrix calculation results of C correspond to the five trust levels of "not positive", "general", "relatively positive", "positive", and "enthusiastic" respectively; the malicious behavior evaluation vector E is set to five fuzzy calculation results E = [e1, e2, e3, e4, e5], corresponding to the evaluation labels "less", "general", "more", "many", and "very many"; RSU makes a trust assessment based on the membership of the trust evaluation vector and the maximum membership principle; if the membership calculation results are the same, the non-membership is compared, and the one with the smaller non-membership is taken to assign the fuzzy behavior evaluation to the current vehicle; (2-3) Fuzzy reasoning and trust value calculation: Fuzzy control rules are set according to the trust behavior evaluation and malicious behavior evaluation, and fuzzy reasoning is performed on the fuzzy behavior evaluation; The fuzzy membership function of the trust level is established using the trapezoidal membership function, and its domain X cre is the vehicle trust value [0,100]; RSU performs multi-label fuzzy reasoning calculation through the activated fuzzy control rules to obtain the fuzzy trust domain U(x): U(x)=c i ∧e j ∧X(a), where ∧ is the smallest value and X(a) is the fuzzy membership function; defuzzify the trust region to obtain the vehicle trust value X; (3) Information sharing: The RSU interacting with the vehicle sets three semantic thresholds and a 3-minute time limit to receive information with similar semantics. If the time reaches the time limit or the number of semantically repeated information reaches the threshold value, the following information authenticity consensus is initiated for all received shared information: 1) Select a number of more credible consensus vehicles based on the fuzzy trust level; 2) RSU assigns different voting weights and consensus completion thresholds to vehicles with different fuzzy trust levels; 3) Each vehicle participates in information consensus according to its own voting weight; 4) Information sharing and consensus voting behaviors are recorded to form a block; 5) For information that is recognized as credible by consensus, the RSU will publish it in the RSU network, and untrustworthy information will be discarded; (4) Information feedback: Set information feedback rules and use fuzzy negative review reasoning to obtain the semantic label of the information; based on the semantic label, the RSU network reminds the information sharer or forces the information to be removed; the demander or sharer independently initiates information update consensus based on the prompt, and updates or cancels the road condition information.

2. According to claim 1, a fuzzy trust management method for traffic information sharing based on blockchain is characterized in that: In step (1), the detailed steps of behavior blockchain construction and vehicle authentication are as follows: (1-1) Establish a behavior blockchain in the network composed of RSUs to store users’ road condition information sharing behaviors and consensus behaviors; (1-2) When a user's vehicle is connected to the Internet of Vehicles, the RSU interacting with the vehicle verifies whether the behavior stored in the connected vehicle is consistent with the behavior data of the vehicle on the behavior chain, and issues an authentication certificate.

3. According to a fuzzy trust management method for traffic information sharing based on blockchain as described in claim 1, it is characterized in that: In step (3), the detailed steps of information sharing are as follows: (3-1) Information collection: After the sharer uploads the traffic information, the RSU sets three semantic thresholds and a 3-minute time limit to receive information with similar semantics based on the traffic information released by the sharer; (3-2) Consensus vehicle selection: Select other vehicles near the road to form a temporary local vehicle network to reach a consensus on the authenticity of the information; when selecting a consensus vehicle, if the fuzzy trust levels of the vehicles near the road are the same, compare their trust values ​​X to select a high-trust vehicle; RSU counts the total number of behavior chain blocks n generated within ten minutes bc , the total number of vehicles connected to the local vehicle network within ten minutes n c And the average number of vehicle network nodes n within ten minutes v ; According to the Poisson distribution, calculate the consensus vehicle selection upper limit k, that is, calculate the value p(k) of the maximum probability k under the Poisson distribution: According to the vehicles connected to the local vehicle network, the number of vehicles that does not exceed the maximum value k of the probability in the Poisson distribution is selected, and the maximum number of consensus vehicles selected does not exceed 8; the above selection of vehicles is carried out in the following way: RSU uses a sliding window method to select vehicles for interaction, which is set with 4 windows with a 30s time window limit, and gradually sends consensus information until the number of vehicle nodes willing to participate in the consensus vote reaches 8; (3-3) Vehicle consensus capability granting: When initiating consensus, RSU grants vehicles with different fuzzy trust levels different voting capabilities and consensus completion thresholds; the details are as follows: 1) Set corresponding voting weights for different fuzzy trust levels. When voting for other vehicles, the voting weight of the vehicle is used to participate in the road condition information voting; the fuzzy trust level "untrustworthy" has a weight of 1, "not very trustworthy" has a weight of 2, "somewhat trustworthy" has a weight of 3, "trustworthy" has a weight of 4, and "very trustworthy" has a weight of 5; 2) "Very trustworthy" vehicles can directly release road condition information without information consensus; 3) For information released by vehicles with "relatively credible" and "credible" trust values, the approval weight of the voting vehicles must accumulate to 3 to complete the information consensus; 4) For information released by "untrusted" and "less trustworthy" vehicles, the approval weight of the voting vehicles must accumulate to 5 to complete the information consensus; (3-4) Information authenticity consensus: The temporary local vehicle network selected in (3-2) reaches information authenticity consensus according to the following steps: 1) Information voting: Each vehicle participates in information consensus according to its own voting weight; information voting behavior only considers the information recognition among consensus vehicles; the voting weight of the vehicle node that recognizes the authenticity of the information is Cre con , obtain the weight of the information consent node; the voting process formula is as follows: G=∑Cre Con Where G is the consensus success threshold of information sharing vehicles; if the consensus vote is successful, the vehicle nodes that do not recognize the information are traced back; if the vote does not reach the consensus success threshold, it means that the information release has failed, and the nodes that recognize the information are traced back; 2) Behavior block generation: All vehicles send digital signatures Sig j , verify the pseudonymous identity of the actor; the RSU network records the information sharing and consensus voting behaviors to form a block for subsequent tracing and query; three key information are recorded on the chain: 1) the address of the vehicle participating in the information release; 2) whether the information release is successful; 3) the address of the vehicle participating in the information voting; according to the needs of trust behavior, the established block structure is shown in the following formula: BC cre =<ID d ,preHash,nonce,time,bool j ,Sig j |Body Among them, ID d is the id of the block, indicating that the block is the dth block of the chain; preHash is the hash value of the previous block; nonce is a random number; time is the timestamp of the creation time; bool j Indicates whether the sharing behavior can be successfully released after being authenticated by surrounding vehicles. If the consensus is successful, bool j If the item is set to 1, when consensus fails, bool j Item is set to 0; Sig j It is the pseudonymous signature of the vehicle that released the information; the Body records the digital signatures and fuzzy trust levels of the vehicles involved in this information authentication, including all vehicles that approve and oppose the information; 3) Information release and block storage: RSU will release information that is recognized as reliable in the network, and unreliable information will be discarded; shared vehicles and authenticated vehicles must store the block locally.

4. According to claim 3, a fuzzy trust management method for traffic information sharing based on blockchain is characterized in that: In step (4), the steps of information feedback are as follows: (4-1) Information evaluation: Users evaluate information sharing, which is divided into positive and negative reviews. RSU collects the number of negative reviews of the information; (4-2) Setting fuzzy rules to remove information: If the published information exceeds the time window t or the number of demand users exceeds the threshold d, the RSU performs fuzzy negative review reasoning on the information and stores its semantic label on the RSU end; The specific process of the above fuzzy negative review reasoning is as follows: according to the trust level and the number of people who need information, a fuzzy negative review rule is established to calculate the fuzzy semantic label of the information; there are two input variables, one is the number of people who need information, and the other is the trust negative review ratio; here, a fuzzy membership function is established for the number of people who need information and the trust negative review ratio. The domain X is the number of people who request a certain information, and its semantic labels are FN 1 "Less", FN 2 "Less", FN 3 "More", FN 4 "More"; the domain X of the trust-to-negative ratio is the trust value N of the negative reviewer cre Trust value with all demanders N total The ratio of X = N cre / N total , establish the semantic label as FD 1 "Low", FD 2 "Lower", FD 3 "Higher", FD 4 "high"; The following fuzzy reasoning rules are established according to the number of people who need information and the ratio of negative reviews:

1. If the number of people who need information is small and the trust negative review ratio is low than the semantic tag is a small number of negative reviews; 2. If the number of people who need information is large and the trust-to-negative ratio is low than the semantic tag has more negative reviews; 3.If the number of people who need information is small and the trust-to-negative ratio is higher than the semantic tag has more negative reviews; 4. If the number of people who need information is small and the ratio of trust to negative reviews is lower than the semantic tags are mostly negative reviews; 5. If the number of people who need information is small and the trust-to-negative ratio is higher than the semantic tag is negative; 6. If the number of people who need information is large and the trust-to-negative ratio is high than the semantic tag is negative reviews; After time t or the number of users who need it exceeds the threshold d, the semantic label of a certain road condition information is calculated; four language variables "a", "b", "c", and "d" are defined to correspond to "a few negative reviews", "many negative reviews", "mostly negative reviews", and "a lot of negative reviews"; and different management measures are set for different levels of negative reviews, specifically: "a few negative reviews" means no reminder, "many negative reviews" means reminder to modify, "mostly negative reviews" means reminder to the demander, and "a lot of negative reviews" means consensus to remove the information; (4-3) Information update, invalidation and removal: After a period of time, traffic information is no longer real-time, and the demander or sharer initiates the consensus to update the information and confirm the invalidation of the information; when the fuzzy reasoning obtains the semantic label "a flood of negative reviews", the RSU forcibly initiates the consensus to remove the information from the shelves; RSU consensus is achieved through the following steps: 1) Witness selection: The method is the same as step (3-2) consensus vehicle selection; 2) Consensus authority granting: The method is the same as step (3-3) vehicle consensus capability granting; 3) Information voting: The method is the same as the information voting in step (3-4); 4) Behavior block generation: The method is the same as the behavior block generation in step (3-4); 5) Information update, invalidation or removal: Information will be modified, invalidated, deleted or removed according to vehicle behavior.

5. A fuzzy trust management device for traffic information sharing based on blockchain, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, a fuzzy trust management method for sharing traffic information based on blockchain as described in any one of claims 1 to 4 is implemented.

6. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by the processor, a fuzzy trust management method for sharing traffic information based on blockchain as described in any one of claims 1 to 4 is implemented.

Images