Internet of vehicles man-vehicle separation trust management method based on double-layer block chain

Abstract

The invention belongs to the field of trust management in Internet of Vehicles security, and relates to an Internet of Vehicles man-vehicle separation trust management method based on a double-layer block chain, which comprises the following steps: carrying out man-vehicle authentication on a driver and a vehicle, and adding the vehicle which is successfully authenticated into an Internet of Vehicles trust management system; the data request vehicle request information and the data providing vehicle provide information; after the data request vehicle receives the information, trust evaluation is carried out on the data providing vehicle, and a new reputation value is obtained; updating the vehicle reputation value block chain based on an AS-PBFT algorithm according to the new reputation value; establishing a driver recent trust table according to the updated vehicle reputation value block chain; evaluating the active state of the driver, calculating a new reputation value of the driver according to the active state of the driver and the recent trust table of the driver, and updating the driver reputation value block chain according to the new reputation value of the driver; according to the invention, the double-layer block chain is adopted to store trust information of people and vehicles, so that the accuracy of the trust degree of the system is improved.

Description

Technical Field

[0001] The present invention belongs to the field of trust management in Internet of Vehicles security, and in particular to a trust management method for vehicle-to-vehicle separation in Internet of Vehicles based on a double-layer blockchain. Background Art

[0002] In recent years, with the rapid development of economy and the accelerated urbanization process, China's transportation industry has flourished, especially with the rise of new energy vehicle industry, the number of private cars has increased rapidly, which has brought great challenges to traffic safety. In order to meet these challenges, researchers have begun to focus on using advanced communication technologies to improve traffic management. Vehicle Ad Hoc Network (VANET) as an extended form of Mobile Ad Hoc Network (MANET) has attracted much attention. Due to the highly dynamic topology of VANET, its security and privacy issues have attracted much attention. Many researchers have proposed a trust management system based on VANET to deal with the security threats brought by potential malicious users in the Internet of Vehicles.

[0003] Traditional trust management systems mainly include centralized, semi-centralized and distributed types. Centralized systems have problems such as low reliability and single point failure, and semi-centralized systems can only partially solve these problems. Blockchain technology is considered to be a feasible technical solution in VANET due to its decentralization, consistency, tamper-proof and transparency. The Internet of Vehicles (IoV) model based on blockchain realizes data sharing and transaction verification through pre-authorized RSU and lightweight vehicle nodes, improving the system's identity authentication and data security. However, the blockchain in VANET currently generally has the problems of high update frequency and many update nodes, which brings storage and computing pressure to the nodes of the Internet of Vehicles. Therefore, the consensus and block rate of traditional blockchain cannot meet the VANET system.

[0004] With the rapid development of new economic models such as online ride-hailing and shared cars, the relationship between vehicles and drivers has become complicated. The traditional IoV trust management model that only considers the trust of vehicles has the problem of trust bias. In vehicle social networks (VSNs), the behavior of drivers will affect their trust and importance in the social system, and this trust relationship is essentially personal social trust between people. In shared transportation models, vehicles and drivers are often in a non-1-to-1 state. A vehicle with honest historical behavior does not necessarily mean that the current driver is trustworthy, and a driver with a high historical reputation value does not necessarily mean that the vehicle he is currently driving is trustworthy.

[0005] At present, most trust management research ignores the consideration of the trust relationship between drivers and vehicles. The existing trust assessment mechanism cannot prevent malicious vehicle attacks caused by trust bias. At the same time, the blockchain architecture in the current Internet of Vehicles trust management system is not suitable for the Internet of Vehicles scenario, resulting in problems such as high system storage pressure and slow node consensus efficiency. Summary of the invention

[0006] In order to solve the above existing technical problems, the present invention adopts a vehicle-to-vehicle separation trust management method based on a double-layer blockchain. The vehicle-to-vehicle trust management system includes: vehicles, local auxiliary roadside units, trusted leader roadside units and a double-layer blockchain; the double-layer blockchain includes: a vehicle reputation value blockchain VBC and a driver reputation value blockchain DBC;

[0007] The trust management methods for separating people from vehicles include:

[0008] S1. Authentication of the driver and vehicle is performed, and the successfully authenticated vehicle is added to the Internet of Vehicles trust management system;

[0009] S2, the data requesting vehicle in the Internet of Vehicles trust management system requests information, and the data providing vehicle in the Internet of Vehicles trust management system provides information to the data requesting vehicle;

[0010] S3. After the data requesting vehicle receives the information of the data providing vehicle, the data providing vehicle is trusted and evaluated to obtain a new reputation value of the data providing vehicle.

[0011] S4. Provide the new reputation value of the vehicle based on the data and update the vehicle reputation value blockchain VBC based on the AS-PBFT algorithm;

[0012] S5. Establish a driver's recent trust table based on the updated vehicle credit value blockchain VBC;

[0013] S6. Evaluate the driver's active status, calculate the driver's new reputation value based on the driver's active status and the driver's recent trust table, and update the driver's reputation value blockchain DBC based on the driver's new reputation value.

[0014] The parameters of the Internet of Vehicles trust management system include: an elliptic curve with a base point P and a one-way hash function H(·); the driver and vehicle authentication process includes:

[0015] S11、Driver i Personal information and vehicle keys Input to vehicle c j Log in; the personal information includes identity information and biometrics

[0016] S12, vehicle c j Select a random number N1 to calculate N1P, and send N1P and current timestamp T1 to the local auxiliary roadside unit L-RSU k and to the trusted leadership roadside unit Send authentication request and signature information; where k is the index of the local auxiliary roadside unit and k0 is the index of the trusted leader roadside unit;

[0017] S13, Query the data of the vehicle credit value blockchain VBC and obtain the registration information And register information and the current timestamp T2 as confirmation information sent to L-RSU k ;in, It is a dynamic identity password for people and vehicles. is the hash value key of the driver's biometrics, To assist with public information;

[0018] S14, L-RSU k Select a random number N2 to calculate N2P, and send N2P and confirmation information Passed to vehicle c j ;

[0019] S15, Vehicle c j The random key is recovered using the recovery algorithm Rep(·) in the fuzzy extraction technology Calculate the hash key of the driver's biometrics Calculate the hash value key K based on the random number generated r =H(N2·N1P||N2P||N1P), calculate vehicle c j Authentication key Will Send to L-RSU k ;L-RSU k Calculate the hash value key K based on the random number generated r =H(N2·N1P||N2P||N1P) and the authentication key

[0020] S16, L-RSU k judge Is it equal to If they are equal, then reply to vehicle c j Verify the successful message and transfer the vehicle j Join the Internet of Vehicles Trust Management System; otherwise, the vehicle c j Join the Internet of Vehicles Trust Management System.

[0021] Trust assessment of data providing vehicles includes:

[0022] S31, L-RSU k Calculate the information quality evaluation value of the vehicle provided by the data;

[0023] S32, L-RSU k Obtain the direct shared interaction history between the data requesting vehicle and the data providing vehicle from the DBC, and calculate the direct interaction trust value based on the direct shared interaction history;

[0024] S33, L-RSU k Obtain the neighbor vehicle information of the data requesting vehicle and the data providing vehicle from the DBC, and calculate the indirect interaction trust value based on the neighbor vehicle information;

[0025] S34, L-RSU k The new reputation value of the vehicle is provided based on the information quality evaluation value, direct interaction trust value and indirect interaction trust value calculation data.

[0026] Information Quality Evaluation Value InfoQ (j′,j,t) The calculation formula is:

[0027]

[0028] Among them, rel represents the message correlation index, Dist represents the data request vehicle c j′ Provide vehicle with data c j The distance coefficient TL represents the temporal freshness of information request and information reception.

[0029] Indirect interaction trust value The calculation formula is:

[0030]

[0031] Where M is the number of neighbor vehicles m, and the neighbor vehicle is the data request vehicle c. j′ Provide vehicle with data c j During communication, all vehicles that can communicate directly with the two vehicles, represents the recommended rating of the neighbor vehicle m at time t for the data provided by the vehicle, InfoS (m,t) represents the evaluation value of neighbor vehicle m at time t, TR (m,t) represents the weight of the reputation value of the driver of neighbor vehicle m at time t.

[0032] The vehicle reputation blockchain VBC is updated based on the data to provide the new reputation value of the vehicle, including:

[0033] S41. Select a consensus committee in the vehicle reputation blockchain VBC;

[0034] S42. The consensus committee conducts AS-PBFT consensus on the new reputation value of the data-providing vehicle. If a consensus is reached, the new reputation value of the data-providing vehicle is updated to VBC.

[0035] The nodes of the vehicle reputation blockchain VBC are vehicles, and each node includes: the reputation value of the vehicle; the consensus committee selected includes:

[0036] S411, setting threshold Cre thr , select the reputation value greater than the threshold Cre in VBC thr Nodes of, get the first node group;

[0037] S412, calculating the activity level sleep of the nodes in the first node group, selecting the nodes with activity level sleep=0 in the first node group, and obtaining the second node group;

[0038] S413, setting the size σ of the consensus committee. If the number of nodes in the second node group exceeds σ, then selecting nodes with reputation values ​​ranked in the top σ to form the consensus committee;

[0039] S414. Randomly generate a leader node in the consensus committee, and use the remaining nodes of the consensus committee as auxiliary nodes.

[0040] The AS-PBFT consensus includes:

[0041] S421, L-RSU k Notify all nodes in the consensus committee to participate in the consensus;

[0042] S422, L-RSU k The request message is transmitted to the leader node, and the leader node confirms the request message;

[0043] S423. After confirming the request message, the leader node sends a pre-preparation message to the auxiliary node. The auxiliary node verifies the pre-preparation message. If the verification is successful, the auxiliary node enters the preparation phase; otherwise, the auxiliary node consensus fails.

[0044] S424: The auxiliary node entering the preparation stage sends preparation information to all other nodes in the consensus committee. All other nodes verify the preparation information. If the verification is successful, the auxiliary node enters the confirmation stage; otherwise, the auxiliary node consensus fails.

[0045] S425, the auxiliary node entering the confirmation phase transmits a confirmation message to the leading node;

[0046] S426: If the leader node receives more than 3f+1 confirmation messages, consensus is reached and all other nodes in the VBC and L-RSUk Reply the confirmation result reached by consensus; otherwise, the consensus fails; where f is the number of malicious nodes in the consensus committee.

[0047] Building a driver's recent trust table includes:

[0048] S51, L-RSU k Send a request to DBC to get the driver d i Most recently driven vehicle j 、Vehicle j Credibility (j,t) 、Driver i Driving a vehicle j Driving distance and the driver i Driving distance of all vehicles RDD all , and according to the driving distance and RDD all Calculate driver d i With vehicle c j The matching degree

[0049] S52, L-RSU k According to the driver i Most recently driven vehicle j 、Vehicle j Credibility (j,t) and the driver i With vehicle c j The matching degree Construct the driver's recent trust table; where t is the vehicle c j The timestamp of reputation calculation.

[0050] Calculate the driver's new reputation include:

[0051]

[0052] Where N is the number of rows in the driver's recent trust table, η is a parameter about the active state, and j n Indicates the vehicle c corresponding to the nth row of the driver's recent trust table j , t n is the vehicle c corresponding to the nth row of the driver's recent trust table j The timestamp of reputation value calculation, n is the row index of the driver’s recent trust table, Rep max is the maximum driver credibility value preset by the system, and λ is a parameter used to adjust the change of the exponential function.

[0053] Beneficial effects:

[0054] 1. The present invention establishes a trust management architecture for separating people from vehicles in the IoV scenario, and adopts a two-layer blockchain to store the trust information of people and vehicles respectively, thereby improving the accuracy of the system's trust evaluation mechanism; 2. The present invention designs a new method for human-vehicle authentication through biometrics and key verification, thereby realizing the trust conversion from driver to vehicle; 3. The present invention scores the reputation of vehicle behavior through direct interaction, indirect interaction and message quality evaluation to ensure the overall trust and security of the system; 4. The present invention proposes an AS-PBFT consensus algorithm that adaptively adjusts the group size, and selects nodes by activity and reputation value to improve the consensus rate and reduce storage pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A model diagram of a vehicle networking reputation management system based on a double-layer blockchain provided in an embodiment of the present invention;

[0056] Figure 2 A flowchart of a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles provided by an embodiment of the present invention;

[0057] Figure 3 A flowchart of a driver and vehicle authentication process provided by an embodiment of the present invention;

[0058] Figure 4 A consensus committee selection flow chart provided for an embodiment of the present invention;

[0059] Figure 5 A schematic diagram of the overall process of the consensus algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] like Figure 1 As shown, the Internet of Vehicles trust management system of the present invention is composed of a trusted leader road side unit (Trusted Leader Road Side Unit, T-RSU), a local secondary road side unit (Local Secondary Road Side Unit, L-RSU), a vehicle and a double-layer blockchain;

[0062] Among them, T-RSU is a trusted entity, which is difficult to be hacked. It has strong computing power and serves as the management center of vehicle registration information in the system. It is also responsible for maintaining the upper blockchain, generating and verifying blocks in turn, and ensuring the security and integrity of data in the system; L-RSU is a semi-trusted entity, responsible for collecting information affecting reputation evaluation, and calculating the current reputation of local vehicle nodes according to a predetermined algorithm; at the same time, L-RSU exists in large numbers in the local vehicle network to provide communication services for local vehicles, including data transmission and information broadcasting, and summarizes and reports important data to T-RSU; vehicles in the system act as full nodes of the blockchain, participate in the maintenance and consensus process of the blockchain, ensure the distributed storage and security of blockchain data, and inherit their historical reputation when the driver logs in, and move in the area; the behavior of the vehicle will affect the driver's reputation, and changes in the reputation are recorded on the blockchain to achieve transparent reputation management.

[0063] Blockchain acts as the core technology of trust management and data sharing in the system, and has the characteristics of decentralization, transparency and traceability. All transactions and changes in reputation values ​​are recorded on the blockchain to prevent data tampering and fraud. The blockchain of the present invention is composed of a driver trust value blockchain (DBC) and a vehicle trust value blockchain (VBC) to store long-term and short-term trust information.

[0064] Specifically, DBC, as the upper blockchain, records all the driver information of global historical registrations, is maintained by T-RSU with better global performance, and will be updated at the update time node set by the system; VBC, as the lower blockchain, records the information exchange records between local vehicles in the current area, traffic accidents that occurred on the day, and real-time traffic information, so that vehicles can obtain timely information in a timely manner; vehicle nodes cannot directly access VBC information. Vehicle nodes will initiate requests to L-RSU when necessary. L-RSU will read relevant information and forward it to vehicle nodes only after confirming that conditions permit; each region is maintained by a separate VBC for reputation maintenance. When a vehicle interacts with reputation in a certain area and enters another area, the traffic information of the previous area does not affect its activities in the local area; L-RSU and local vehicles in a certain area jointly maintain the blockchain. VBC is small in scale and updated frequently, and AS-PBFT is used for consensus.

[0065] The system has two data sharing modes: general service requests and traffic condition notifications. General service requests are mostly information query services between vehicles, such as navigation requests, parking lot or gas station location search, entertainment information services, etc. When a vehicle needs service, it can request service from nearby vehicles. Traffic condition notifications are more likely to occur when traffic events (such as accidents, congestion) occur, and the witnessing vehicle will broadcast the information to VBC and other vehicles.

[0066] Ordinary service requests are used as a low-priority data sharing method. The specific steps are as follows:

[0067] 1) When data request vehicle c j′ When a service is needed, the vehicle j′ To the nearby L-RSU k Send a service request. After receiving the service request and confirming its legitimacy, L-RSU k Broadcast the service request message to all vehicle nodes in the VBC; where k is the index of the L-RSU and j′ is the index of the data requesting vehicle;

[0068] 2) Data providing vehicle c j To L-RSU k Send shared data, L-RSU k Vehicle inspection required j Current driver reputation Is it greater than a certain threshold Rep thr This is to prevent malicious vehicles from sending a large number of malicious messages to hinder the operation of the system. If they meet the requirements, their data will be uploaded to VBC.

[0069] 3) Data request vehicle c j′ Other vehicles interested in the data can directly download relevant data from VBC, and the downloaded data can be used to meet the service request needs of the vehicle; where j is the index of the data requesting vehicle.

[0070] 4) The vehicle that downloads the data will evaluate the quality of the information to ensure the accuracy and reliability of the data. This evaluation information will be recorded on the blockchain for reference by other vehicles.

[0071] Traffic condition notifications are usually urgent messages, so they have a high priority. The relevant steps are:

[0072] 1) When traffic conditions notify vehicle c j When a traffic event (such as an accident or congestion) is detected, it broadcasts the information to the L-RSU k Specifically, the data providing vehicle will broadcast traffic event information to the VBC node and nearby vehicles. This information may include the accident location, road closures, traffic congestion level, etc.

[0073] 2) Information receiver vehicle c j′ After obtaining traffic information, appropriate measures can be taken to avoid traffic accidents or optimize driving routes, for example, changing the driving route to avoid congested areas or accident scenes; the received information helps drivers make timely decisions to ensure driving safety and efficiency.

[0074] 3) After a certain period of time, the information receiving vehicle c j′ You can choose to provide information to the vehicle c j Conduct reputation evaluation. These evaluation information will be recorded on the blockchain to help maintain the reputation management mechanism of the entire Internet of Vehicles trust management system.

[0075] Based on the above trust management system architecture, the present invention provides a trust management method for vehicle-to-vehicle separation based on a double-layer blockchain. The process is as follows: Figure 2 As shown, including:

[0076] S1. Authentication of the driver and vehicle is performed, and the successfully authenticated vehicle is added to the Internet of Vehicles trust management system;

[0077] S2, the data requesting vehicle in the Internet of Vehicles trust management system requests information, and the data providing vehicle in the Internet of Vehicles trust management system provides information to the data requesting vehicle;

[0078] S3. After the data requesting vehicle receives the information of the data providing vehicle, the data providing vehicle is trusted and evaluated to obtain a new reputation value of the data providing vehicle.

[0079] S4. Provide a new reputation value for the vehicle based on the data and update the vehicle reputation value blockchain VBC based on the AS-PBFT algorithm; the AS-PBFT is an adaptive Byzantine fault tolerance algorithm.

[0080] S5. Establish a driver's recent trust table based on the updated vehicle credit value blockchain VBC;

[0081] S6. Evaluate the driver's active status, calculate the driver's new reputation value based on the driver's active status and the driver's recent trust table, and update the driver's reputation value blockchain DBC based on the driver's new reputation value.

[0082] Furthermore, due to the particularity and importance of the driver's identity, the present invention proposes a dual authentication scheme based on a key and biometrics based on a fuzzy extractor; a cyclic group G of order q, an elliptic curve with a base point P, and a secure and collision-resistant one-way hash function H(·) are defined; when the driver logs in to a previously registered vehicle, the system will perform identity authentication on it to ensure that the identity information is accurate, and then verify the driver's biometrics through fuzzy extraction technology and random keys, and finally confirm the driver's identity and the legitimacy of the vehicle.

[0083] Specifically, Figure 3 As shown, the driver and vehicle authentication process includes:

[0084] S11、Driver i Personal information and vehicle keys Input to vehicle c j Log in; the personal information includes identity information and biometrics

[0085] S12, vehicle c j Select a random number N1 to calculate N1P, and send N1P and current timestamp T1 to L-RSU k , and to Send identity information Certification Request and signature information Signature; where k is the index of the local auxiliary roadside unit, and k0 is the index of the trusted leader roadside unit;

[0086] S13, Query the data of the vehicle credit value blockchain VBC and obtain the registration information And register information and the current timestamp T2 as confirmation information sent to L-RSU k ;in, It is a dynamic identity password for people and vehicles. is the hash value key of the driver's biometrics, To assist with public information;

[0087] The registration information is the information provided when the person or vehicle is first registered. The relevant verification information stored, the relevant steps of registration include:

[0088] Driver i Personal information and vehicle keys Input to vehicle c j Register; the personal information includes identity information and biometrics By vehicle cj The random key generation function Gen() of the fuzzy extractor on the above process the input biometric features to generate a random key and auxiliary public information Calculate the encryption password Select a random number N0 to calculate the dynamic identity password of the person and the vehicle Delete the key After that, the registration message Send to Verify and if the verification is successful, the registration is successful.

[0089] S14, L-RSU k Select a random number N2 to calculate N2P, and send N2P and confirmation information Passed to vehicle c j ;

[0090] S15, Vehicle c j Use the recovery algorithm Rep() in fuzzy extraction technology to recover the random key Calculate the hash key of the driver's biometrics Calculate the hash value key K based on the random number generated r =H(N2·N1P||N2P||N1P), calculate vehicle c j Authentication key Will Send to L-RSU k ;L-RSU k Calculate the hash value key K generated by a random number r =H(N2·N1P||N2P||N1P) and the authentication key

[0091] S16, L-RSU k judge Is it equal to If they are equal, then reply to vehicle c j Verify the successful message and transfer the vehicle j Join the Internet of Vehicles Trust Management System; otherwise, the vehicle c j Join the Internet of Vehicles Trust Management System.

[0092] By data request vehicle c j′ Request data, data provided by vehicle c j After the vehicle interaction process that provides data is completed, L-RSU k It is necessary to calculate the vehicle c after this behavior j The vehicle reputation value;

[0093] Specifically, the trust assessment of the data providing vehicle includes:

[0094] S31. In order to consider the credibility of the message itself and filter out untrue or misleading vehicle messages, in L-RSU k The information quality of the vehicle is evaluated in InfoQ, and the information quality evaluation value of the vehicle is obtained. (j′,j,t) , the calculation formula is:

[0095]

[0096] TL=η(t-t0) -ε

[0097] Among them, rel represents the message relevance index, which is L-RSU k A subjective parameter in evaluating vehicle reputation, Dist represents the distance coefficient, TL represents the time freshness, t0 represents the information request time, t represents the time of information reception, d v represents the diameter of the VBC area, d represents the data requesting vehicle c j′ Provide vehicle with data c j , ε and η represent the parameters introduced to adjust the speed of change of time freshness.

[0098] S32, L-RSU k The direct shared interaction history between the data requesting vehicle and the data providing vehicle is obtained from the DBC, and the direct interaction trust value is calculated according to the direct shared interaction history using the reputation function based on the beta distribution. The calculation formula is:

[0099]

[0100] in, Indicates vehicle c j′ For vehicle c j The direct interaction reputation value of vehicle c, pos and neg represent j′ With vehicle c j The number of positive and negative interactions. After each information exchange between the data requesting vehicle and the data providing vehicle, the data requesting vehicle will evaluate the communication interaction once, evaluating it as positive or negative.

[0101] S33. In the Internet of Vehicles environment, direct interaction evaluation based on a single vehicle cannot accurately evaluate vehicle interaction. It is necessary to introduce indirect interaction evaluation as a supplement to the overall trust evaluation of vehicle behavior to obtain the indirect interaction trust value.

[0102] The calculation process of indirect interaction trust value includes:

[0103] S331, L-RSUk Obtaining the neighbor vehicle information of the data requesting vehicle and the data providing vehicle in the DBC and VBC, wherein the neighbor vehicle refers to all vehicles that can directly communicate with the data requesting vehicle and the data providing vehicle when the two vehicles are communicating; the neighbor vehicle information includes the reputation value of the neighbor vehicle, the recommended rating for the data providing vehicle, the driver's residence, the license plate location, the vehicle condition rating, the driving distance, and the distance between the neighbor vehicle and the data providing vehicle;

[0104] S332, constructing a neighbor vehicle evaluation table according to the residence, license plate location, vehicle condition rating, driving distance, and vehicle distance from the data providing vehicle of the driver of the neighbor vehicle, and calculating the neighbor vehicle evaluation value according to the neighbor vehicle evaluation table;

[0105] As shown in Table 1, the neighbor vehicle evaluation table includes multiple records, each of which includes: an indicator, a category interval, and a score for the indicator (o a , o b ), where a and b are the indexes of the indicators;

[0106] Table 1 Neighborhood vehicle evaluation table

[0107]

[0108] S333, calculating the indirect interaction trust value according to the reputation value, recommendation rating and evaluation value of the neighbor vehicle;

[0109] Specifically, the indirect trust value calculation formula is:

[0110]

[0111] Where M is the number of neighbor vehicles m, is the recommendation rating of neighbor vehicle m at time t, indicating the evaluation of the neighbor vehicle on the data providing vehicle, 1 and 0 represent positive and negative evaluations respectively; InfoS (m,t) represents the evaluation value of neighbor vehicle m at time t, TR (m,t) represents the historical trust weight, is the reputation value of the driver of neighbor vehicle m at time t, and l is the index of the neighbor vehicle.

[0112] S34, when the information quality evaluation, direct interaction evaluation and indirect interaction evaluation are completed, L-RSU k It is necessary to combine and score the multiple factors that affect the vehicle's reputation to obtain a new reputation value. The calculation formula for the new reputation value is:

[0113]

[0114] Among them, α1, α2, and α3 are weights, and α1+α2+α3=1.

[0115] The vehicle reputation blockchain VBC is updated based on the data to provide the new reputation value of the vehicle, including:

[0116] S41. Select a consensus committee in the vehicle reputation blockchain VBC;

[0117] S42. When VBC needs to generate a block, the consensus committee conducts AS-PBFT consensus on the new reputation value of the data-providing vehicle. If a consensus is reached, the new reputation value of the data-providing vehicle is updated to VBC.

[0118] The nodes of the vehicle reputation blockchain VBC are vehicles. Each node includes: the reputation value of the vehicle, the specific process of selecting the consensus committee is as follows: Figure 4 As shown, the specific process is as follows:

[0119] S411, setting threshold Cre thr , select the VBC with a reputation value greater than the threshold Cre thr Nodes of, get the first node group;

[0120] S412, constructing a candidate node group, calculating the activity level sleep of the nodes in the first node group, selecting nodes with activity level sleep=0 in the first node group to obtain a second node group, and adding the remaining nodes in the first node group to the candidate node group;

[0121] Calculating the activity level of the selected node sleep includes:

[0122] Activity level sleep = {0, 1, 2}. The smaller the level number, the higher the activity. Sleep can be assessed based on factors such as the vehicle driver's active time and login interval, including:

[0123] The system time of the vehicle driver logging in and out of the Internet of Vehicles trust management system is obtained in DBC, and the active interval is calculated based on the system time; two active thresholds are set. If the active interval is greater than the first active threshold, it is level zero, sleep = 0; if the active interval is less than the first active threshold and greater than the second active threshold, it is level one, sleep = 1; if the active interval is less than the second active threshold, it is level two, sleep = 2.

[0124] S413, setting the size σ of the consensus committee. If the number of nodes in the second node group exceeds σ, sorting the nodes in the second node group according to the reputation value, selecting the nodes with the reputation value ranking R in the front σ to form the consensus committee, and adding the remaining nodes to the candidate node group;

[0125] S413. Randomly generate a leader node in the consensus committee, and use the remaining nodes of the consensus committee as auxiliary nodes.

[0126] The specific process of AS-PBFT consensus is as follows: Figure 5 As shown, the details are as follows:

[0127] The size of the consensus committee σ must meet the malicious node fault tolerance criterion of PBFT (σ>3f+1); where f is the number of malicious nodes in the consensus committee, that is, the number of nodes that are allowed to fail;

[0128] S421, L-RSU k Notify all nodes in the consensus committee to participate in the consensus, and wait for the consensus to start;

[0129] S422, L-RSU k The request message is transmitted to the leader node, and the leader node confirms the request message;

[0130] S423. After confirming the request message, the leading node sends a pre-preparation message to the auxiliary node, and the auxiliary node verifies the pre-preparation message. If the verification is successful, the auxiliary node enters the preparation stage; otherwise, the auxiliary node consensus fails.

[0131] S424: The auxiliary node entering the preparation stage sends preparation information to all other nodes in the consensus committee. All other nodes verify the preparation information. If the verification is successful, the auxiliary node enters the confirmation stage; otherwise, the auxiliary node consensus fails.

[0132] S425, the auxiliary node entering the confirmation phase transmits a confirmation message to the leading node;

[0133] S426: When the leader node receives more than 3f+1 confirmation messages, consensus is reached and all other nodes in the VBC and L-RSU k Reply to confirm the result; otherwise, consensus fails.

[0134] Building a driver's recent trust table includes:

[0135] S51, L-RSU k Send a request to DBC to get the driver d i Most recently driven vehicle j 、Driver i Driving the most recently driven vehiclec j Credibility (j,t) 、Driver i Driving the most recently driven vehiclec j Driving distance and the driver i Driving distance of all vehicles RDD all , and according to the driving distance and RDD allCalculate driver d i With vehicle c j The matching degree of people and cars The matching degree between driver and vehicle can measure the i Recent and driving vehicles c j The degree of binding relationship;

[0136] S52, L-RSU k According to the driver i Most recently driven vehicle j 、Vehicle j Credibility (j,t) and the driver i With vehicle c j The matching degree Construct the driver's recent trust table; where t is the vehicle c j The timestamp of the reputation value calculation; The driver's recent trust table is shown in Table 2:

[0137] Table 2 Drivers' recent trust table

[0138]

[0139] Assessment of the driver's active state includes:

[0140] S61, L-RSU k Send a request to DBC to get the driver d i System time for logging in and out of the Internet of Vehicles Trust Management System T log and T exit , and according to the system time T log and T exit Calculate the active interval TD di =T log -T exit ;

[0141] S62: Setting the activity value threshold TD thr , according to the active interval TD di and activity threshold TD thr Determine the driver's activity level. If the active interval TD di Greater than the active value threshold TD thr , then it is an active member; otherwise, it is an inactive member.

[0142] Calculate the driver's new reputation value Rep based on the driver's recent trust table and the driver's active status di , the formula is as follows:

[0143]

[0144] Where N is the number of rows in the driver's recent trust table, η is a parameter about the active state, and j n Indicates the vehicle c corresponding to the nth row of the driver's recent trust table j , t n is the vehicle c corresponding to the nth row of the driver's recent trust table j The timestamp of reputation value calculation, n is the row index of the driver’s recent trust table, Rep max is the maximum driver credibility value preset by the system, and λ is a parameter used to adjust the change of the exponential function.

[0145] The above embodiments further illustrate the purpose, technical solutions and advantages of the present invention in detail. It should be understood that the above embodiments are only preferred implementation modes of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made to the present invention within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A trust management method for vehicle-to-vehicle separation in an Internet of Vehicles based on a double-layer blockchain, characterized in that: The Internet of Vehicles trust management system includes: vehicles, local auxiliary roadside units, trusted leader roadside units and double-layer blockchains; the double-layer blockchains include: vehicle reputation value blockchain VBC and driver reputation value blockchain DBC; the vehicle-to-human separation trust management method includes: S1. Authentication of the driver and vehicle is performed, and the successfully authenticated vehicle is added to the Internet of Vehicles trust management system; S2, the data requesting vehicle in the Internet of Vehicles trust management system requests information, and the data providing vehicle in the Internet of Vehicles trust management system provides information to the data requesting vehicle; S3. After the data requesting vehicle receives the information of the data providing vehicle, the data providing vehicle is trusted and evaluated to obtain a new reputation value of the data providing vehicle. S4. Provide the new reputation value of the vehicle based on the data and update the vehicle reputation value blockchain VBC based on the AS-PBFT algorithm; S5. Establish a driver's recent trust table based on the updated vehicle credit value blockchain VBC; S6. Evaluate the driver's active status, calculate the driver's new reputation value based on the driver's active status and the driver's recent trust table, and update the driver's reputation value blockchain DBC based on the driver's new reputation value.

2. According to claim 1, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: The parameters of the Internet of Vehicles trust management system include: an elliptic curve with a base point P and a one-way hash function H(·); the driver and vehicle authentication process includes: S11、Driver i Personal information and vehicle keys Input to vehicle c j Log in; the personal information includes identity information and biometrics S12, vehicle c j Select a random number N1 to calculate N1P, and send N1P and current timestamp T1 to the local auxiliary roadside unit L-RSU k and to the trusted leadership roadside unit Send authentication request and signature information; where k is the index of the local auxiliary roadside unit and k0 is the index of the trusted leader roadside unit; S13, Query the data of the vehicle credit value blockchain VBC and obtain the registration information And register information and the current timestamp T2 as confirmation information sent to L-RSU k ;in, It is a dynamic identity password for people and vehicles. is the hash value key of the driver's biometrics, To assist with public information; S14, L-RSU k Select a random number N2 to calculate N2P, and send N2P and confirmation information Passed to vehicle c j ; S15, Vehicle c j The random key is recovered using the recovery algorithm Rep(·) in the fuzzy extraction technology Calculate the hash key of the driver's biometrics Calculate the hash value key K based on the random number generated r =H(N2·N1P||N2P||N1P), calculate vehicle c j Authentication key Will Send to L-RSU k ;L-RSU k Calculate the hash value key K based on the random number generated r =H(N2·N1P||N2P||N1P) and the authentication key S16, L-RSU k judge Is it equal to If they are equal, then reply to vehicle c j Verify the successful message and transfer the vehicle j Join the Internet of Vehicles Trust Management System; otherwise, the vehicle c j Join the Internet of Vehicles Trust Management System.

3. According to claim 1, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: Trust assessment of data providing vehicles includes: S31, L-RSU k Calculate the information quality evaluation value of the vehicle provided by the data; S32, L-RSU k Obtain the direct shared interaction history between the data requesting vehicle and the data providing vehicle from the DBC, and calculate the direct interaction trust value based on the direct shared interaction history; S33, L-RSU k Obtain the neighbor vehicle information of the data requesting vehicle and the data providing vehicle from the DBC, and calculate the indirect interaction trust value based on the neighbor vehicle information; S34, L-RSU k The new reputation value of the vehicle is provided based on the information quality evaluation value, direct interaction trust value and indirect interaction trust value calculation data.

4. According to claim 3, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: Information Quality Evaluation Value InfoQ (j′,j,t) The calculation formula is: Among them, rel represents the message correlation index, Dist represents the data request vehicle c j′ Provide vehicle with data c j The distance coefficient TL represents the temporal freshness of information request and information reception.

5. According to claim 3, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: Indirect interaction trust value The calculation formula is: Where M is the number of neighbor vehicles m, and the neighbor vehicle is the data request vehicle c. j′ Provide vehicle with data c j During communication, all vehicles that can communicate directly with the two vehicles, represents the recommended rating of the neighbor vehicle m for the data at time t, InfoS (m,t) represents the evaluation value of neighbor vehicle m at time t, TR (m,t) represents the weight of the reputation value of the driver of neighbor vehicle m at time t.

6. According to claim 1, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: The vehicle reputation blockchain VBC is updated based on the data to provide the new reputation value of the vehicle, including: S41. Select a consensus committee in the vehicle reputation blockchain VBC; S42. The consensus committee conducts AS-PBFT consensus on the new reputation value of the data-providing vehicle. If a consensus is reached, the new reputation value of the data-providing vehicle is updated to VBC.

7. According to claim 6, a vehicle-to-vehicle separation trust management method based on a double-layer blockchain is characterized in that: The nodes of the vehicle reputation blockchain VBC are vehicles, and each node includes: the reputation value of the vehicle; the consensus committee selected includes: S411, setting threshold Cre thr , select the reputation value greater than the threshold Cre in VBC thr Nodes of, get the first node group; S412, calculating the activity level sleep of the nodes in the first node group, selecting the nodes with activity level sleep=0 in the first node group, and obtaining the second node group; S413, setting the size σ of the consensus committee. If the number of nodes in the second node group exceeds σ, then selecting nodes with reputation values ​​ranked in the top σ to form the consensus committee; S414. Randomly generate a leader node in the consensus committee, and use the remaining nodes of the consensus committee as auxiliary nodes.

8. According to claim 6, a vehicle-to-vehicle separation trust management method based on a double-layer blockchain is characterized in that: The AS-PBFT consensus includes: S421, L-RSU k Notify all nodes in the consensus committee to participate in the consensus; S422, L-RSU k The request message is transmitted to the leader node, and the leader node confirms the request message; S423. After confirming the request message, the leader node sends a pre-preparation message to the auxiliary node. The auxiliary node verifies the pre-preparation message. If the verification is successful, the auxiliary node enters the preparation phase; otherwise, the auxiliary node consensus fails. S424: The auxiliary node entering the preparation stage sends preparation information to all other nodes in the consensus committee. All other nodes verify the preparation information. If the verification is successful, the auxiliary node enters the confirmation stage; otherwise, the auxiliary node consensus fails. S425, the auxiliary node entering the confirmation phase transmits a confirmation message to the leading node; S426: If the leader node receives more than 3f+1 confirmation messages, consensus is reached and all other nodes in the VBC and L-RSU k Reply the confirmation result reached by consensus; otherwise, the consensus fails; where f is the number of malicious nodes in the consensus committee.

9. According to claim 1, a method for trust management of vehicle-to-vehicle separation based on a double-layer blockchain in an Internet of Vehicles, characterized in that: Building a driver's recent trust table includes: S51, L-RSU k Send a request to DBC to get the driver d i Most recently driven vehicle j 、Vehicle j Credibility (j,t) 、Driver i Driving a vehicle j Driving distance and the driver i Driving distance of all vehicles RDD all , and according to the driving distance and RDD all Calculate driver d i With vehicle c j The matching degree S52, L-RSU k According to the driver i Most recently driven vehicle j 、Vehicle j Credibility (j,t) and the driver i With vehicle c j The matching degree Construct the driver's recent trust table; where t is the vehicle c j The timestamp of reputation calculation.

10. A vehicle network human-vehicle separation trust management method based on a double-layer blockchain according to claim 9, characterized in that: Calculate the driver's new reputation value Rep di include: Where N is the number of rows in the driver's recent trust table, η is a parameter about the active state, and j n Indicates the vehicle c corresponding to the nth row of the driver's recent trust table j , t n is the vehicle c corresponding to the nth row of the driver's recent trust table j The timestamp of reputation value calculation, n is the row index of the driver’s recent trust table, Rep max is the maximum driver credibility value preset by the system, and λ is a parameter used to adjust the change of the exponential function.

Images