Safe delivery method for crowdsourcing scene data of Internet of Vehicles

Abstract

The invention provides an Internet of Vehicles crowdsourcing scene data safety delivery method, and belongs to the technical field of Internet of Vehicles safety. The technical problems of insufficient security, low data authentication efficiency and high overhead in the vehicle sensing data transmission process under the crowdsourcing scene of the Internet of Vehicles are solved. According to the technical scheme, the method comprises the following steps that S1, a TA generates system parameters and achieves OBU and TR registration; s2, generating a crowdsourcing task by the TR, and executing task allocation scheduling; s3, the OBU uploads encrypted sensing data by using the anonymous identity, and the RSU realizes batch anonymous identity verification of the data source; s4, performing efficient data aggregation verification; and S5, tracing the real identity of the malicious vehicle. The method has the beneficial effects that in order to guarantee safe delivery of vehicle data in a crowdsourcing scene, efficient authentication of perceptual vehicle identities is realized by designing an anonymous batch authentication algorithm, integrity verification of aggregated data is realized by using a homomorphic hash function idea, and the efficiency problem of safety verification during delivery of perceptual data is solved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle networking security technology, and in particular to a method for securely delivering vehicle networking crowdsourcing scenario data. Background Art

[0002] Vehicular Networking refers to a system that interconnects vehicles, road infrastructure, and other equipment based on wireless communication technology, with the goal of improving driving experience, traffic safety, and efficiency. Crowdsourcing is a key technology for the intelligentization of vehicular networking. It can collect vehicle status information, weather conditions, and other data in real time through vehicle OBUs and sensors, and allow vehicles to instantly report traffic information such as traffic accidents, road congestion, and road construction. These data are shared with other vehicles or cloud services through the network, providing data support and intelligent decision-making for many vehicle applications, warning of potential traffic problems such as road closures or sudden accidents in advance, and helping drivers adjust routes in a timely manner, thereby improving driving experience, reducing traffic congestion, and improving traffic efficiency, effectively promoting the development of intelligent transportation.

[0003] The basic principle of crowdsourcing technology is to decompose complex tasks into small subtasks through some kind of incentive mechanism, and then publish them to a large number of volunteers through an online platform. In the Internet of Vehicles environment, crowdsourcing technology makes full use of the potential data collection capabilities of many vehicle nodes, realizes the collection, processing and analysis of large-scale data through task decomposition and vehicle collaboration, and provides information and resources for intelligent traffic decision-making. Internet of Vehicles crowdsourcing technology is divided into three stages: task distribution, data upload and incentive mechanism. Task distribution refers to the generation and distribution of tasks by data requesters; data upload refers to the data collection of participating vehicles and uploading them to the crowdsourcing server through the Internet of Vehicles to the requester; incentive mechanism is the key to maximizing crowdsourcing efficiency, which means that the crowdsourcing server or data requester evaluates the data and gives the participating vehicles corresponding rewards as incentives. It can be seen that the data sharing of Internet of Vehicles crowdsourcing technology depends on the active contributions of many participants. This openness has both significant efficiency and will inevitably bring severe network security and data privacy challenges, such as malicious user identity identification problems caused by the diversity and anonymity of participants, data quality and credibility problems caused by the extensiveness of data collection, and data privacy problems caused by data sensitivity. If these security issues are not resolved, they will not only threaten the overall security of the Internet of Vehicles system, but may also undermine users' trust and willingness to participate, and reduce the efficiency of crowdsourcing technology.

[0004] In the task distribution phase, the data requester defines, designs, and publishes the perception task, selects the perception vehicle, and assigns the task. However, the task description and related information may contain sensitive data, such as the personal identity information or commercial secrets of the data requester. The leakage of this information may lead to security and privacy issues, resulting in location attacks, background knowledge attacks, and Sybil attacks. Therefore, how to achieve the definition and distribution of tasks while avoiding the leakage of sensitive information of the data requester, and to formulate a reasonable task allocation strategy, is a key task in the task allocation phase. In the data upload phase, crowdsourcing participants execute and submit tasks, and the data requester or crowdsourcing platform reviews the received results. However, the data uploaded by the perception vehicle to complete the task may contain personal privacy or sensitive information. If this data is not properly protected during the upload process, it may lead to privacy leakage and trigger background knowledge attacks and specific task attacks. At the same time, if encryption technology is not used or the data is not processed, the uploaded data may be intercepted, tampered with, or damaged. The threat to data integrity may affect the accuracy and credibility of the task results, leading to intermediate node attacks. In addition, if the participants who upload the data are not authenticated and cannot prove their legitimacy, malicious users may upload false data, resulting in Sybil attacks. Finally, completing crowdsourcing tasks not only takes up the user's time, but also consumes the user's physical resources. Without a reasonable incentive mechanism, users cannot be encouraged to participate in crowdsourcing collaboration, but malicious transactions may occur during the transaction process, causing property losses. For example, in the incentive mechanism stage, the payment information of participants and data requesters may be leaked and vulnerable to reasoning attacks. At the same time, if the participants who receive the rewards are not legitimate participants, impersonation and fraud may occur, or participants and data requesters may manipulate the system through false tasks or evaluations to obtain improper rewards or improve their reputation, leading to collusion attacks and point association attacks.

[0005] Since the data upload stage involves communication through open channels, the user's identity privacy, data integrity and privacy protection are particularly vulnerable, and there are many related attacks. Therefore, this paper mainly studies the security issues of the data upload stage of Internet of Vehicles crowdsourcing. The security goals of this stage are:

[0006] 1. Identity verification and privacy protection, which requires reliable data collection and transmission without revealing the true identity of the participants, and the legitimacy of the pseudonym can be verified;

[0007] 2. Data integrity and privacy protection, which requires preventing data from being tampered with or sensitive information from being accessed or analyzed by unauthorized third parties during data transmission and storage.

[0008] The crowdsourcing technology of the Internet of Vehicles can effectively improve the vehicle data perception and analysis capabilities, but the message transmission mechanism of the open channel makes the perception vehicle vulnerable to malicious attacks during the data upload process, and its identity and data security are relatively fragile. Therefore, the security of crowdsourcing communication in the Internet of Vehicles has become a hot topic of concern for scholars in recent years. Researchers have proposed some solutions based on traditional privacy protection technologies such as anonymization and differential privacy. For example, Wang et al. proposed a privacy protection scheme based on differential privacy for the evaluation of perception data in edge computing environments in the paper "Privacy-Preserving Streaming Truth Discovery in Crowdsourcing With Differential Privacy". The scheme reduces the amplitude of adding noise to the data according to the reliability of the perception participants, thereby improving the accuracy of the truth discovery process. At the same time, differential privacy is used to ensure the privacy security of the perception participants in multiple events. Lin et al. used differential privacy to implement a mobile crowd sensing scheme in the paper "Enhancing Sparse Mobile Crowd Sensing With Manifold Optimization and Differential Privacy". The scheme perturbs the perception data by locally injecting noise sampled from the truncated Gaussian mechanism to ensure the validity of the locally perturbed data and achieve data privacy protection. The use of differential privacy technology in the above-mentioned literature will inevitably affect the accuracy and availability of data, but more importantly, it does not take into account the privacy protection of the identity of the participants. Yan et al. proposed a data aggregation scheme in a crowdsourcing scenario in the paper "Verifiable, Reliable, and Privacy-Preserving Data Aggregation in Fog-Assisted Mobile Crowdsensing". The scheme has the characteristics of verifiability, reliability and privacy protection, and can achieve data privacy through secret sharing and Paillier encryption. At the same time, the scheme also designs group authentication to achieve anonymous authentication to prevent identity impersonation by external attackers. Sun et al. proposed a mobile crowdsensing privacy protection framework based on fog nodes in the paper "Toward Incentivizing Fog-Based Privacy-Preserving Mobile Crowdsensing in the Internet of Vehicles". The scheme designs a token reward exchange mechanism based on partial blind signature authentication and zero-knowledge authentication technology, which cuts off the relationship between user identity and data in the links of data upload, reputation management and incentive mechanism, and can effectively protect the privacy of participants. At the same time, the scheme uses homomorphic encryption technology to ensure data privacy in the process of realizing multi-data aggregation.Zhao et al. proposed a privacy protection scheme for mobile crowdsensing based on federated analytics in the paper CrowdFA: A Privacy-Preserving Mobile Crowdsensing Paradigm via FederatedAnalytics. The scheme relies on the efficient encryption primitive of additive secret sharing and realizes the privacy protection of aggregated data and incentive mechanism at the same time, thereby ensuring the privacy security of user identity. Scheme participants can perform local aggregation operations, promote a series of aggregation operations of private data, and protect the privacy of their original sensor data.

[0009] Since OBU computing resources are limited in the Internet of Vehicles environment, the communication link duration is short, and the network topology changes frequently, how to optimize the scheme, reduce the amount of computing and improve communication efficiency is a key indicator point. Relevant literature has made further attempts in this regard. For example, Wu proposed the idea of ​​limiting the number of pseudonyms issued in the paper "Enabling Data Trustworthiness and User Privacy in Mobile Crowdsensing", which helps to solve the problem of reduced efficiency caused by anonymous abuse in identity authentication. The scheme improves the efficiency of anonymous identity authentication through group signatures and blind signature-based protocols. Wu et al. further proposed a privacy-preserving edge task assignment framework based on MCS in the paper "From Centralized Management to Edge Collaboration: A Privacy-Preserving Task Assignment Framework for Mobile Crowdsensing". The scheme adopts attribution-based user clustering and batch group signatures, which not only improves the matching accuracy of tasks and participants, but also takes into account the efficiency and security performance of task assignment due to the use of the edge task framework. Cheng et al. attempted to study an efficient MCS anonymous authentication and data reliability evaluation scheme in the paper "Efficient Anonymous Authentication and Privacy-Preserving Reliability Evaluation for Mobile Crowdsensing in Vehicular Networks". The scheme uses bilinear mapping to achieve batch anonymous authentication. At the same time, it adopts a privacy-preserving reputation comparison algorithm to compare the reputation value with the reputation threshold, select reliable sensing vehicles, and improve the mutual authentication efficiency on the basis of ensuring the identity reliability of the sensing vehicle. Vijayakumar et al. proposed an efficient batch authentication and key exchange scheme in the paper "Ananonymous batch authentication and key exchange protocols for 6G enabled VANETs". It adopts anonymous batch authentication based on bilinear mapping, avoids communication with malicious sensing vehicles, and reduces the authentication burden of roadside units when authenticating in congested areas. The scheme also uses digital signature technology to ensure the integrity of communication messages and prevent messages from being modified during transmission.Wang et al. proposed an efficient and traceable road condition monitoring scheme based on fog in the paper "Privacy-preserving cloud-fog–based traceable road condition monitoring in VANET". The scheme uses certificateless aggregate signature encryption technology to realize multi-message aggregate verification, which greatly reduces the computational overhead of sensor vehicles. The scheme uses a pseudonym mechanism to achieve identity privacy protection and uses bilinear mapping to achieve anonymous batch authentication. However, its disadvantage is that bilinear mapping technology and multiple authentication will not only increase the computational and communication overhead, but also reduce the willingness of perception vehicles to participate, affecting the integrity of perception data results. To address this problem, the paper "Conditional privacy-preserving batch authentication scheme based on certificateless aggregate signature in vehicle networking" proposed a conditional privacy-preserving batch authentication scheme based on certificateless aggregate signature. The scheme uses certificateless public key cryptography and aggregation technology to realize batch message authentication in the vehicle networking environment, improving the authentication efficiency. Since the scheme does not use Map-to-Point hash functions and bilinear pairing operations with high computational consumption, it has potential advantages in saving costs and improving computational efficiency.

[0010] In order to improve the efficiency of data verification, most perception vehicles adopt the method of first efficiently aggregate and verify the data and then upload it. However, because the perception vehicle and the data platform cannot maintain a long-term connection in the mobile communication environment, the reliability and integrity of data upload cannot be guaranteed. Most literatures propose that hash functions or bilinear mapping can be used to verify data integrity at this stage, but hash functions cannot achieve batch verification, which will reduce the efficiency of the verification party. Although bilinear mapping can achieve linear batch verification, the computational overhead is relatively large.

[0011] In response to the above problems, the present invention proposes a method for secure data delivery in crowdsourcing scenarios of Internet of Vehicles based on edge computing. The solution uses edge nodes to complete anonymous batch authentication of perceived vehicles and supports malicious vehicle identity tracing. At the same time, efficient aggregate data integrity verification is achieved based on homomorphic hash functions. Since the solution uses related technologies such as the Paillier homomorphic encryption algorithm and homomorphic hash functions in cryptography, in order to clearly explain the principles and steps of the present invention, the relevant background technology is first described:

[0012] 1. Paillier Homomorphic Encryption Algorithm

[0013] Homomorphic encryption is a special encryption technology that allows computing operations to be performed in an encrypted state. The result is still encrypted, but the result is the same as the corresponding unencrypted operation. This property allows computing without revealing the data content, which is crucial for protecting data privacy and security. The Paillier algorithm is a type of homomorphic encryption algorithm. The algorithm satisfies additive homomorphism for the calculation of ciphertext and ciphertext, and can be used to perform calculations securely without exposing the original data. The algorithm is defined as follows:

[0014] (1) Setup: Select two large prime numbers p and q, and calculate n = p·q.

[0015] (2) KeyGen: Calculate λ = LCM (p-1, q-1), that is, generate the least common multiple of p-1 and q-1. Then select a random number As The generator of the secret parameter μ=(L(g λ modn 2 )) -1 modn, where L() is expressed as L(x) = (x-1) / n. Generate a public-private key pair (pk, sk), where the public key is pk = (n, g) and the private key is sk = (λ, μ).

[0016] (3)Encryption: Select a random number Use pk to encrypt the message m to get the ciphertext. The encryption process is:

[0017] (4) Decryption: Use sk to ciphertext C m Decryption is performed. The decryption process is

[0018] This algorithm satisfies the additive homomorphic property, that is, for any messages m1 and m2,

[0019] 2. Homomorphic Hash Function

[0020] Homomorphic hash function is a special type of hash function, which not only has the basic properties of traditional hash functions, such as anti-collision and determinism, but also supports some forms of homomorphic operations. The properties are as follows:

[0021] (1) Collision-free. No adversary can create a tuple (m1, m2, m3, k1, k2) in probabilistic polynomial time such that m3 = k1m1 + k2m2 and

[0022] (2) Homomorphism. For any two data m1, m2 and real numbers k1, k2, we have

[0023] The present invention utilizes the properties of additive homomorphic hash functions to implement integrity verification of aggregated data, that is, for two messages x and y, there exists H(x+y)=H(x)+H(y).

[0024] 3. Difficulty issues

[0025] Elliptic Curve Discrete Logarithm Problem (ECDLP): Let G be an elliptic curve cyclic group of order p, P be a generator (i.e., base point) of G, given P, aP∈G, in probabilistic polynomial time, any adversary can compute is difficult. Summary of the invention

[0026] The present invention provides a method for secure data delivery in a crowdsourcing scenario of the Internet of Vehicles, the purpose of which is to use edge nodes to complete lightweight anonymous batch authentication of sensing vehicles and simultaneously achieve identity tracing of malicious vehicles. In addition, an aggregate data integrity verification method is designed based on homomorphic hash function technology, which reduces both the computational overhead of the user end and the communication overhead.

[0027] The present invention is implemented by the following measures: A method for secure delivery of crowdsourcing scene data in an Internet of Vehicles, comprising the following steps:

[0028] S1. TA generates system parameters and registers the physical OBU and TR;

[0029] S2, TR generates crowdsourcing tasks and performs task allocation and scheduling;

[0030] S3,OBU uses anonymous identity to upload encrypted perception data, and RSU implements batch anonymous identity authentication of data sources;

[0031] S4, data efficient aggregation verification;

[0032] S5. Tracing the true identity of malicious vehicles.

[0033] Furthermore, the S1 step includes:

[0034] S11. Input security parameter k. TA generates system parameters through the generation algorithm Gen(k). The process is as follows:

[0035] S111. Select a multiplication cyclic group G with a prime number p as its generator, select a large prime number q to calculate n=p·q, and define a hash function

[0036] S112, TA selects a random number And save {α, β} as the system master key, calculate P pub =α·P,T pub =β·P;

[0037] S113, TA public system public parameters PP = {G, P, P pub ,T pub ,H1,H2,H3,n};

[0038] S12, vehicle initialization, the vehicle will be real identity RID i Sent as a unique identifier to TA, TA randomly selects As an OBU i The private key of As a public key, TA will Sent to OBU via secure channel i ;

[0039] S13, TR is initialized, TA generates a homomorphic key pair pk for the data requester TR tr =(n,g d ), sk tr =(μ d ,λ d ), where n = p·q, and sent to TR through a secure channel;

[0040] Furthermore, the S2 step includes:

[0041] S21, TR perception task, and determine the type of sensing task T m (such as traffic flow, traffic accidents, traffic light conditions, etc.), the area range of the sensing task L m and the time threshold of the sensing task Time m TR generates a data request message {L m ,T m ,Time m ,pk tr} and sent to the crowdsourcing server MCS, where pk tr is the public key of TR;

[0042] S22, MCS generates a sensing task sequence number T according to the request message id , using it as the identifier of the sensing task and generating a task request message {T id ,T m ,Time m ,pk tr} and send it to RSU. RSU distributes its broadcast to the sensing vehicles in the area;

[0043] Furthermore, the S3 step includes:

[0044] S31, OBU self-generates pseudonym PID i =(PID i,1 ,PID i,2 ) is sent to TA through the channel, and TA generates a pseudonymous commitment for it. OBU signs after correctness verification and performs trustworthy processing on the uploaded data. The process is as follows:

[0045] S311, OBU generates an anonymous identity, TA generates a pseudonymous commitment for it, and then OBU verifies the correctness and signs it. The process is as follows:

[0046] S3111、OBU i First, pick a random number Then calculate the PID i,1 =γ i ·P, Generate pseudonymous PID i =(PID i,1 ,PID i,2 ), and finally send the pseudonym to TA through a public channel to apply for pseudonym commitment;

[0047] S3112, TA first uses the pseudonym PID i Trace whether the real identity is legal. If it is a legal user, generate a pseudonym commitment for it. For the specific tracing process, see the identity tracing and revocation stage. TA selects a random number Calculate A i =s i ·P, Z i =H2(A i ||PID i ), Generate Pseudonymous Commitment Send it to OBU i ;

[0048] S3113、OBU i Get this pseudonymous commitment to verify the correctness and judge Is it equal to T pub +Z i ·P, if they are equal, it means that this message is a pseudonymous commitment generated by TA. If they are not equal, the message is discarded and a new pseudonymous commitment is applied. The derivation proof of correctness verification is as follows:

[0049]

[0050] S3114、OBU i Calculating the Hash in It is a timestamp. A signature is generated at the same time Can verify the legitimacy of anonymous identities;

[0051] S312, select random number The data is m i , generate data ciphertext

[0052] S313. Calculate homomorphic hash verification code This is convenient for TR to verify the integrity of the data. Send to RSU;

[0053] S32, RSU verifies the legitimacy of the anonymous vehicle identity through the aggregate signature. The process is as follows:

[0054] S321, RSU receives the data packet sent by OBU and first checks the timestamp Is the time threshold of the sensing task reached? m If not, it means that it does not meet the standard of the sensing task and is invalid data, so the relevant data packet is discarded;

[0055] S322, verify the identity of the sensing vehicle to ensure the reliability of the data source. The identity authentication process of a single sensing vehicle is the verification equation Is it true? If it is true, the user is a legitimate user. Otherwise, the relevant data ciphertext is discarded. The correctness analysis is as follows:

[0056]

[0057] S323: If there are n sensing vehicles participating in the sensing task T id , then RSU performs anonymous batch authentication and verifies the equation If it is established, the reliability of the OBU identity is guaranteed and the data is received; if it is not established, the relevant data is discarded. The correctness analysis is as follows:

[0058]

[0059] S324, if the perceived vehicle passes the identity verification, the RSU will Upload to MCS;

[0060] Furthermore, the S4 step includes:

[0061] S41, MCS aggregates the perceived data ciphertext and data hash verification code, and sends the result to TR. The process is as follows:

[0062] S411. If the sensor data uploaded by n sensing vehicles is valid and the data source is reliable, MCS generates an aggregated ciphertext C and sends it to TR. The data aggregation function is expressed as

[0063] S412, MCS aggregates the data hash verification code, and the formula is expressed as Finally, the message Send to TR;

[0064] S42, TR uses the private key to decrypt the plaintext of the perception data and uses the homomorphic hash function idea to verify the integrity of the data. The process is as follows:

[0065] S421, TR receives the aggregated ciphertext C sent by MCS and uses the private key sk tr Decrypt it to obtain the plaintext data. The decryption formula is:

[0066]

[0067] S422. Use homomorphic hash functions to verify the data to ensure data integrity. Verify equation H' m =H m Is it true? If true, it means the data has not been tampered with; otherwise, it means the data has been changed and is discarded. The correctness analysis is as follows:

[0068]

[0069] Furthermore, the step S5 includes:

[0070] S51, TR will be the pseudonymous PID of the malicious data source i Send to TA, TA according to the pseudonym PID i =(PID i,1 ,PID i,2 ) and the system private key α to trace the real identity and calculate Get the real identity RID i Then revoke it. i No longer having a legitimate identity, TA completes the identity tracing and revocation of the malicious data source. The specific tracing process is as follows:

[0071]

[0072] For ease of understanding, the symbols and their meanings involved in the present invention are given as shown in the following table:

[0073]

[0074]

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] (1) The present invention proposes an anonymous batch verification algorithm for vehicle identities in a crowdsourcing scenario of the Internet of Vehicles. The present invention does not use Map-to-Point hash functions and bilinear pairing operations with high computational overhead for verification, but instead uses aggregation technology and pseudonymous commitments generated by TA to achieve authentication, greatly improving the authentication efficiency while ensuring the privacy of anonymously perceived vehicle identities.

[0077] (2) The present invention designs a data aggregation algorithm based on additive homomorphic encryption, and utilizes homomorphic hash functions to realize efficient integrity verification of aggregated data, thereby effectively reducing communication overhead while providing data confidentiality and integrity protection.

[0078] (3) The present invention has been theoretically analyzed and proven in terms of identity authentication and data security, showing that the present invention has security features such as unforgeability, anonymity, traceability, unlinkability, resistance to Sybil attacks, and resistance to replay attacks. Experimental analysis and comparison show that the present invention has many advantages in terms of computing overhead and communication overhead compared to other existing solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0080] Figure 1 The overall flow chart provided by the present invention.

[0081] Figure 2 The possible threat model diagram of each stage of the crowdsourcing life cycle in the present invention is

[0082] Figure 3 Schematic diagram of the system model in the present invention.

[0083] Figure 4 This is a flow chart of the solution for the data uploading phase provided by the present invention.

[0084] Figure 5 A flow chart of the data efficient aggregation verification phase provided by the present invention.

[0085] Figure 6 This is a comparison chart of the computational overhead of a single anonymous authentication provided by the present invention.

[0086] Figure 7 A comparison chart of the computational overhead of batch anonymous authentication provided by the present invention.

[0087] Figure 8 This is a comparison chart of the communication overhead required for each solution under a single number of participants provided by the present invention.

[0088] Fig. 9This is a comparison chart of the communication overhead required for each solution under different numbers of participants provided by the present invention. DETAILED DESCRIPTION

[0089] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0090] Example 1

[0091] This embodiment 1 provides a method for securely delivering crowdsourcing scenario data in an Internet of Vehicles. Figure 1 As shown, the following steps are included:

[0092] S1. TA generates system parameters and registers the physical OBU and TR;

[0093] S2, TR generates crowdsourcing tasks and performs task allocation and scheduling;

[0094] S3,OBU uses anonymous identity to upload encrypted perception data, and RSU implements batch anonymous identity authentication of data sources;

[0095] S4, data efficient aggregation verification;

[0096] S5. Tracing the true identity of malicious vehicles.

[0097] like Figure 3 As shown, the Internet of Vehicles architecture in the Internet of Vehicles crowdsourcing scenario data security delivery method includes: a trusted authority (TA), a crowdsourcing server (MCS), a data requester (TR), a roadside unit (RSU) and a perception vehicle (OBU).

[0098] (1) Trusted Center: TA has powerful computing and storage capabilities and is a fully trusted entity. It is responsible for generating system parameters, registering and distributing entity keys for all vehicles and RSUs, generating commitments for pseudonyms of data providers, and tracing and revoking the identities of malicious vehicles.

[0099] (2) Crowdsourcing Server: MCS has powerful computing and storage capabilities and is responsible for publishing task requirements and aggregating and submitting data ciphertext.

[0100] (3) Data requester: A person or organization that is interested in the IoV status data, such as a traffic management department. TR has a demand for certain types of data and hopes to extract valuable information from the data to restore or predict the status, such as traffic congestion estimation, accident reporting, parking behavior, or energy consumption. TR needs to verify the integrity of the data.

[0101] (4) Roadside Unit: RSU has strong computing and storage capabilities and can provide fast local response. It is mainly responsible for the allocation of perception tasks and the identity authentication of connected vehicles. It uses its edge node characteristics to reduce the computing overhead of other entities.

[0102] (5) Perception vehicle: Equipped with an onboard unit (OBU), it has certain computing and storage capabilities and can accept tasks and provide perception data in crowdsourcing scenarios. It is also responsible for self-generating pseudonyms and encrypting and uploading perception data.

[0103] Furthermore, the specific contents of S1 include the following steps:

[0104] S11. Input security parameter k. TA generates system parameters through the generation algorithm Gen(k). The process is as follows:

[0105] S111. Select a multiplication cyclic group G with a prime number p as its generator, select a large prime number q to calculate n=p·q, and define a hash function

[0106] S112, TA selects a random number And save {α, β} as the system master key, calculate P pub =α·P,T pub =β·P;

[0107] S113, TA public system public parameters PP = {G, P, P pub ,T pub ,H1,H2,H3,n};

[0108] S12, vehicle initialization, the vehicle will be real identity RID i Sent as a unique identifier to TA, TA randomly selects As an OBU i The private key of As a public key, TA will Sent to OBU via secure channel i ;

[0109] S13, TR is initialized, TA generates a homomorphic key pair pk for the data requester TR tr =(n,g d ), sk tr =(μ d ,λ d ), where n = p·q, and is sent to TR through a secure channel.

[0110] Furthermore, the specific contents of S2 include the following steps:

[0111] S21, TR perception task, and determine the type of sensing task T m (such as traffic flow, traffic accidents, traffic light conditions, etc.), the area range of the sensing task L m and the time threshold of the sensing task Time m TR generates a data request message {L m ,T m ,Time m ,pk tr} and sent to the crowdsourcing server MCS, where pk tr is the public key of TR;

[0112] S22, MCS generates a sensing task sequence number T according to the request message id , using it as the identifier of the sensing task and generating a task request message {T id ,T m ,Time m ,pk tr} and sends it to RSU. RSU distributes its broadcast to the sensing vehicles in the area.

[0113] like Figure 4 As shown, the specific contents of S3 include the following steps:

[0114] S31, OBU self-generates pseudonym PID i =(PID i,1 ,PID i,2 ) is sent to TA through the channel, and TA generates a pseudonymous commitment for it. OBU signs after correctness verification and performs trustworthy processing on the uploaded data. The process is as follows:

[0115] S311, OBU generates an anonymous identity, TA generates a pseudonymous commitment for it, and then OBU verifies the correctness and signs it. The process is as follows:

[0116] S3111、OBU i First, pick a random number Then calculate the PID i,1 =γ i ·P, Generate pseudonymous PID i =(PID i,1 ,PID i,2 ), and finally send the pseudonym to TA through a public channel to apply for pseudonym commitment;

[0117] S3112, TA first uses the pseudonym PID i Trace whether the real identity is legal. If it is a legal user, generate a pseudonym commitment for it. For the specific tracing process, see the identity tracing and revocation stage. TA selects a random number Calculate Ai =s i ·P, Z i =H2(A i ||PID i ), Generate Pseudonymous Commitment Send it to OBU i ;

[0118] S3113、OBU i Get this pseudonymous commitment to verify the correctness and judge Is it equal to T pub +Z i ·P, if they are equal, it means that this message is a pseudonymous commitment generated by TA. If they are not equal, the message is discarded and a new pseudonymous commitment is applied. The derivation proof of correctness verification is as follows:

[0119]

[0120] S3114、OBU i Calculating the Hash in It is a timestamp. A signature is generated at the same time Can verify the legitimacy of anonymous identities;

[0121] S312, select random number The data is m i , generate data ciphertext

[0122] S313. Calculate homomorphic hash verification code This is convenient for TR to verify the integrity of the data. Send to RSU;

[0123] S32, RSU verifies the legitimacy of the anonymous vehicle identity through the aggregate signature. The process is as follows:

[0124] S321, RSU receives the data packet sent by OBU and first checks the timestamp Is the time threshold of the sensing task reached? m If not, it means that it does not meet the standard of the sensing task and is invalid data, so the relevant data packet is discarded;

[0125] S322, verify the identity of the sensing vehicle to ensure the reliability of the data source. The identity authentication process of a single sensing vehicle is the verification equation Is it true? If it is true, the user is a legitimate user. Otherwise, the relevant data ciphertext is discarded. The correctness analysis is as follows:

[0126]

[0127] S323: If there are n sensing vehicles participating in the sensing task T id , then RSU performs anonymous batch authentication and verifies the equation If it is established, the reliability of the OBU identity is guaranteed and the data is received; if it is not established, the relevant data is discarded. The correctness analysis is as follows:

[0128]

[0129] S324, if the perceived vehicle passes the identity verification, the RSU will Upload to MCS.

[0130] like Figure 5 As shown, the specific contents of S4 include the following steps:

[0131] S41, MCS aggregates the perceived data ciphertext and data hash verification code, and sends the result to TR. The process is as follows:

[0132] S411. If the sensor data uploaded by n sensing vehicles is valid and the data source is reliable, MCS generates an aggregated ciphertext C and sends it to TR. The data aggregation function is expressed as

[0133] S412, MCS aggregates the data hash verification code, and the formula is expressed as Finally, the message Send to TR;

[0134] S42, TR uses the private key to decrypt the plaintext of the perception data and uses the homomorphic hash function idea to verify the integrity of the data. The process is as follows:

[0135] S421, TR receives the aggregated ciphertext C sent by MCS and uses the private key sk tr Decrypt it to obtain the plaintext data. The decryption formula is:

[0136]

[0137] S422. Use homomorphic hash functions to verify the data to ensure data integrity. Verify equation H' m =H m Is it true? If true, it means the data has not been tampered with; otherwise, it means the data has been changed and is discarded. The correctness analysis is as follows:

[0138]

[0139] Furthermore, the specific contents of S5 include the following steps:

[0140] S51, TR will be the pseudonymous PID of the malicious data source i Send to TA, TA according to the pseudonym PID i =(PID i,1 ,PID i,2 ) and the system private key α to trace the real identity and calculate Get the real identity RID i Then revoke it. i No longer having a legitimate identity, TA completes the identity tracing and revocation of the malicious data source. The specific tracing process is as follows:

[0141]

[0142] In order to verify the feasibility of the present invention, the security of this embodiment 1 is analyzed:

[0143] 1. Security Analysis

[0144] Lemma 1: Under the random oracle model, assuming that ECDLP is hard, there exists an unforgeable, i.e., secure, solution under the adaptive chosen-message attack and chosen-identity attack of adversary A.

[0145] Proof: If there is an adversary A who can forge a legitimate message with a non-negligible advantage after a polynomial time bounded query Then there is a challenger c who can use subroutine A to solve the ECDLP problem with a non-negligible advantage. Challenger c will simulate Oracle and play the following game with adversary A:

[0146] Initialization phase: Challenger c initializes a system and randomly selects α. As TA's private key, calculate Ppub = α·P, T pub =β·P, generating system parameters {p,G,P,P pub ,T pub ,H1,H2,H3,n}, send the system parameters to adversary A.

[0147] H asks: Challenger c creates an initially empty list L H , whose element type is {T id ,PID i ,Tim i ,h i}. When adversary A uses PID i When requesting a query, challenger c queries list L H If the corresponding element exists in the list, h will be returned directly. i Otherwise, challenger c selects a random value In response, {T id ,PIDi ,Tim i ,h i}Add to list L H middle.

[0148] User creates a query: Challenger c creates an initially empty list L PK , whose element type is {PID i ,Cert i ,sk i ,pk i}. When adversary A uses PID i When requesting a query, challenger c queries list L PK If the corresponding element exists in the list, pk will be returned directly. i Otherwise, challenger c randomly selects Setting A i =α i ·P, Z i =H2(α i ||PID i ), Cert i =(β+Z i )modp,sk i =⊥, Q = x·P. Verifiable equation Cert i P = T pub +Z i P is established, then pk i In response, {PID i ,Cert i ,sk i ,pk i}Save to list L PK middle.

[0149] Commitment query: Adversary A requests information about PID i Commitment query, challenger c query list L PK Does {PID already exist in i ,Cert i ,sk i ,pk i}. If it already exists, the challenger c will Cert i as a response; otherwise, the challenger c responds ⊥.

[0150] Key acquisition query: Adversary A requests information about PID i The challenger c queries the list L PK Does {PID already exist in i ,Cert i ,sk i ,pki}. If it already exists, the challenger c will sk i as a response; otherwise, the challenger c responds ⊥.

[0151] Signature query: Adversary A requests information about PID i The challenger c randomly selects Calculate T pub =Sig i ·PZ i Ph i ·pk i . Verifiable equation Sig i P = T pub +Z i P+h i ·pk i Established. Challenger c will Sig i It is returned to adversary A as a response.

[0152] Output: Finally, adversary A outputs the element set This verifies the equation Sig i P=Cert i P+h i ·pk i According to the fork theorem, adversary A replays the above process to forge another set of messages So we can get the equation According to the equation, we can get

[0153]

[0154] but

[0155] Challenger C output To solve a given ECDLP instance, the challenger can use the adversary's ability to solve the ECDLP problem with a non-negligible advantage, which is contrary to the difficulty problem. Therefore, the present invention can meet the unforgeability under adaptive selected message attack and selected identity attack. The proof is complete.

[0156] Theorem 1 This embodiment can achieve anonymity.

[0157] Prove that in the mobile car networking crowdsourcing scenario, all vehicles are perceived to communicate with a fake identity PID i =(PID i,1 ,PID i,2 ) as the identity. The real identity RID i Only the sensing vehicle and TA know it, and the false identity of the sensing vehicle is PID i,1 =β i ·P, where γi P is based on the difficulty of ECDLP. Therefore, this embodiment has the anonymity of communication, and the proof is complete.

[0158] Theorem 2 This embodiment can achieve traceability of malicious vehicles.

[0159] It proves that TA is an authoritative organization that can supervise the crowdsourced data transmission process. If a malicious data source is found, TA can quickly trace its identity. TA has the system master key α and calculates γ i P pub =γ i α·P=α·PID i,1 , thereby restoring the true identity Therefore, this embodiment can achieve efficient traceability.

[0160] Theorem 3 This embodiment can achieve the unlinkability of pseudonyms.

[0161] Prove that in this embodiment, for different perception tasks, the perception vehicle can have multiple pseudonyms at the same time, and one pseudonym can only participate in one perception task. Perception vehicle is randomly selected γ i , to generate a pseudonymous PID i =(PID i,1 ,PID i,2 ), PID i,1 =γ i ·P, Considering γ i The randomness of the pseudonym is different each time, so the γ i are different, so it is impossible for the adversary to link any two pseudo-identities FPID i and QPID i Therefore, this embodiment can realize the unlinkability of pseudonyms.

[0162] Theorem 4 This embodiment can resist Sybil attacks.

[0163] The Sybil attack refers to an adversary creating a large number of vehicle pseudonyms to send forged messages, thereby affecting the normal operation of the network. In this embodiment, after the vehicle successfully completes the authentication registration, the pseudonyms it generates are all obtained from the TA. This makes it easy for RSU to verify its legitimacy. When uploading data to RSU, unauthenticated vehicle pseudonyms will not be recognized by RSU. Therefore, when the adversary creates a large number of pseudonyms and sends a large number of forged messages, RSU will not pay attention to the credibility of the message itself, but directly check whether the corresponding pseudonym identity is legal. If the pseudonym is illegal, it will directly reject the sent message. Therefore, this embodiment can resist witch attacks.

[0164] Theorem 5 This embodiment can achieve data confidentiality.

[0165] Proving data confidentiality means protecting the data to ensure that unauthorized access cannot view or obtain the data content. This embodiment uses Paillier homomorphic encryption technology to protect the confidentiality of the data. The sensing vehicle uses the public key pk of the data requester to tr =(n,g d ) encrypts the perception data, and only those who have the corresponding private key sk tr =(μ d ,λ d ) can be successfully decrypted by the data requester TR. Considering that RSU is a semi-trusted facility, OBU independently selects the random number For plaintext m i Symmetric encryption is performed and then uploaded to the RSU, so that the RSU cannot obtain the plaintext and read or tamper with the data. Based on the elliptic curve discrete logarithm problem, the attacker cannot obtain the relevant information of the plaintext. Therefore, this embodiment has data confidentiality, and the proof is complete.

[0166] Theorem 6 This embodiment can realize data integrity verification.

[0167] Proving data integrity means that the data has not been tampered with or damaged in each link of storage, transmission, and processing, and the data remains original, accurate, and complete. i Process and obtain the homomorphic hash verification code While data aggregation, data hash verification code is also aggregated, that is, When TR needs to verify data integrity, it performs cumulative point multiplication on the plaintext, i.e. (m1+m2+…+m n )·P, compared with equation H m =(m1+m2+…m n )·P is established. If the equation is established, the integrity of the data is verified. If the equation is not established, the corresponding plaintext is discarded. Therefore, this embodiment can realize data integrity verification.

[0168] Theorem 7 This embodiment can resist replay attacks.

[0169] Prove that the perception vehicle v i Messages sent Timestamp with sender Any neighboring vehicle or RSU can be checked To verify the freshness of the message, it prevents repeated broadcast of messages in the RSU domain Therefore, this embodiment can resist replay attacks.

[0170] Example 2

[0171] In order to demonstrate the performance superiority of the present invention, Example 2 compares the functionality and communication overhead of the present invention with the security solutions in other Internet of Vehicles crowdsourcing scenarios, and also compares the computational overhead of the present invention with the anonymous batch verification solution. The comparison schemes include the scheme proposed by Yan et al. in "Verifiable, Reliable, and Privacy-Preserving Data Aggregation in Fog-Assisted Mobile Crowdsensing" 1, the scheme proposed by Cheng et al. in "Efficient Anonymous Authentication and Privacy-Preserving Reliability Evaluation for Mobile Crowdsensing in Vehic-ular Networks" 2, the scheme proposed by Sucasas et al. in "A Signature Scheme with Unlinkable-yet-Accountable Pseudonymity for Privacy-Preserving Crowdsensing" 3, the scheme proposed by Yan et al. in "P2SIM: Privacy-Preserving and Source-Reliable Incentive Mechanism for Mobile Crowdsensing" 4, the scheme proposed by Vijayakumar et al. in "An anonymous batch authentication and key exchange protocols for 6G enabled VANETs" 5, and the scheme proposed by Maurya et al. in "Efficient Anonymous Batch Authentication Scheme This paper analyzes Solution 6 proposed in "With Conditional Privacy in the Internet of Vehicles (IoV) Applications" and Solution 7 proposed by Li et al. in "RPPM: A Reputation-Based and Privacy-Preserving Platoon Management Scheme in Vehicular Networks".

[0172] 1. Functional comparison

[0173] Functional comparison includes: whether it supports self-generated pseudonyms, whether it supports batch anonymous authentication, whether it supports integrity verification of aggregated data, and whether malicious vehicles can be revoked in a timely manner:

[0174]

[0175] First, compared with the other four schemes, this embodiment supports self-generated pseudonyms, does not rely on allocation by trusted institutions, and can have multiple pseudonyms to participate in multiple perception tasks at the same time; batch verification is adopted in both identity authentication and data verification, which greatly reduces the computing overhead on the user side; at the same time, the identity of malicious vehicles can be revoked in real time, and the illegal behavior of malicious vehicles can be restricted in time to ensure safety.

[0176] Scheme 1 and Scheme 2 rely on TA when generating pseudonyms, which has single point failure, centralization risk and scalability problems, and does not support integrity verification of aggregated data. Scheme 3 uses Paillier homomorphic encryption technology to process data and supports data aggregation, but does not consider integrity verification of aggregated data. When verifying the integrity of data, it still verifies it sequentially, which reduces the efficiency of verification. Scheme 4 does not support self-generated pseudonyms and cannot guarantee the integrity of aggregated data. In addition, Scheme 4 cannot revoke the source of malicious data in a timely manner.

[0177] 2. Computational Cost Comparison

[0178] In order to ensure the fairness and credibility of the comparison, this embodiment uses third-party data, such as Zheng et al., who simulated basic cryptographic operations in "Hybrid Message Authentication Scheme for Internet of Vehicles Based on ZeroKnowledge Proof" to obtain the basic calculation running time described in the following table. The calculation time mainly includes the time of a bilinear mapping operation T bp , first power calculation time T s , one ECC-related modular multiplication operation time T epm , one modular multiplication operation time T mul And the hash function operation time T h .

[0179]

[0180] In the perception vehicle selection phase, the RSU or cloud server uses anonymous authentication to verify the validity of the perception vehicle. Therefore, we compare the computational overhead of this embodiment with other similar schemes 1 and 5-7 in the anonymous authentication phase. The specific computational overhead of each scheme is shown in the following table.

[0181]

[0182]

[0183] In this embodiment, for a perception task, a single anonymous authentication process consists of 3 ECC modular multiplication operations, and a batch anonymous authentication process includes (n+2) ECC modular multiplication operations. Therefore, the computational overhead of single anonymous authentication and batch anonymous authentication is 3T respectively. emp and (n+3)T epm .

[0184] In Scheme 1, a single anonymous authentication process consists of 1 pairing operation and 1 ECC modular multiplication operation. Meanwhile, a batch anonymous authentication process includes 1 pairing operation and n ECC modular multiplication operations. However, in the scheme, e(P2, P1) is assumed to be a public parameter and does not participate in the calculation, but is not indicated in the public parameter list, and the multiplication operation when calculating the product is not considered. Therefore, it is necessary to consider that in this embodiment, a single anonymous authentication process consists of 2 pairing operations and 1 ECC modular multiplication operation, and a batch anonymous authentication process consists of 2 pairing operations, n ECC modular multiplication operations, and 2n modular multiplication operations, and the calculation overhead is T respectively. bp +T epm and T bp +nT epm +2nT mul .

[0185] In scheme 5, Z = e(g1, g2) is known by default and does not require calculation, but it is not shown in the public parameters, and the bilinear mapping operation is not considered, but it needs to be considered in this embodiment. Therefore, the single anonymous authentication process consists of 2 bilinear mapping operations and 1 modular multiplication operation, and the batch anonymous authentication process includes 2 bilinear mapping operations and (2n-1) modular multiplication operations. Therefore, the computational overhead of single anonymous authentication and batch anonymous authentication is 2T respectively. bp +T mul and 2T bp +(2n-1)T mul .

[0186] In Scheme 6, the single anonymous authentication process consists of 3 bilinear mapping operations, 5 ECC modular multiplication operations, and 2 hash operations, and the batch anonymous authentication process includes 3 bilinear mapping operations, (3n+2) ECC modular multiplication operations, 2n hash operations, and n modular multiplication operations. Therefore, the computational overhead of single anonymous authentication and batch anonymous authentication is 3T respectively. bp +5T emp +2T h and 3T bp +(3n+2)T epm +2nT h+nT mul .

[0187] In Scheme 7, the single anonymous authentication process includes 2 ECC modular multiplication operations, and the batch anonymous authentication process consists of (n+2) ECC modular multiplication operations and 2n modular multiplication operations. Therefore, the computational overhead of single anonymous authentication and batch anonymous authentication is 2T respectively. emp and (n+2)T epm +2nT mul .

[0188] Since this embodiment is based on ECC like most other solutions, only a small amount of ECC modular multiplication is used in the anonymous batch verification process, avoiding the use of additional bilinear mapping operations, thereby reducing the computational overhead of anonymous authentication. When N=1, the comparison result is as follows Figure 5 As shown, compared with the first three solutions, this embodiment has obvious advantages, but compared with solution 7, the single anonymous authentication calculation overhead of this embodiment is slightly higher. Figure 6 It can be seen that when N changes, the computational overhead of this embodiment, Scheme 1, and Schemes 5-7 in the batch anonymous authentication stage increases with the increase in the number of participants. However, compared with other schemes, the advantage of this embodiment in terms of computational overhead gradually increases.

[0189] 3. Communication overhead comparison

[0190] The communication overhead comparison of this embodiment is based on the cryptographic function library MIRACL, in which the ordinary cyclic group G in the library T The element size on |G T | is 32Bytes, the element size |G| on the elliptic curve group G is 40Bytes, the element size |G1| on the group G after bilinear transformation is 64Bytes, the integer domain The element size |Z| in is 20Bytes, the element size |H| of the hash function (SHA-256) output is 20Bytes, and it is assumed that the identity and timestamp size |T| is 4Bytes.

[0191]

[0192] In this embodiment, the data requester sends a perception task {L m ,T m ,Time m ,pk tr}, the bit length of the sensing task is 4|Z|+|G|. After the sensing participant accepts the task, he generates a pseudonym, obtains the verification credential, and generates the sensing ciphertext, which is sent to the RSU, that is, The bit length is 2|T|+|Z|+|G T|+3|G|. RSU will verify the received sensing data Sent to MCS, its bit length is |Z|+|T|+|G|+|G T |. MCS aggregates all the received perception data and Sent to TR, bit length is |Z|+|G T |+|G|. Therefore, the total communication overhead of this embodiment is 7×20+6×40+3×32+3×4=488 Bytes.

[0193] In the entire crowdsourcing phase of Scheme 1, the data requester needs to send 3 elements in the integer field, 3 elements on the elliptic curve group G, and 4 general cyclic groups G. T The length of the elements on is 3|Z|+3|G|+4|G T |. The crowdsourcing server needs to send 4 elements in the integer domain, 2 elements on the elliptic curve group G, 1 element output by the hash function, 1 element on the group G after bilinear transformation, and 1 identity element, with a bit length of 4|Z|+2|G|+|H|+|G1|+|T|. The perceiving participant needs to send 1 element in the integer domain, 1 identity element, 2 elements output by the hash function, 1 element on the elliptic curve group G, and 2 ordinary cyclic groups G T The bit length of the elements on is |Z|+|T|+2|H|+|G|+2|G T |. Therefore, the total communication overhead of scheme

[11] is 11×20+6×(40+32)+64+2×4=724Bytes.

[0194] In scheme 4, the data requester needs to send 8 elements in the integer field and 1 general cyclic group G T The bit length of the elements on is 8|Z|+|G T |. The crowdsourcing server needs to send 1 element in the integer domain, 3 elements output by the hash function, and 1 general cyclic group G T The bit length of the elements on is |Z|+3|H|+|G T |. The perceiving participant needs to send 1 element in the integer domain, 3 elements output by the hash function, and 5 general cyclic groups G T The element on the bit length is |Z|+3|H|+5|G T |. Therefore, the total communication overhead of scheme

[11] is 10×20+7×32+6×20=544Bytes.

[0195] Figure 7 The communication overhead of this embodiment compared with other related solutions when N=1 is depicted in the form of a histogram. Specifically, compared with Solution 1 and Solution 4, this embodiment reduces the additional communication overhead by 47% and 11%, respectively. Figure 8 As shown, the communication overhead increases with the increase of the number of participants N, but compared with other schemes, the communication overhead of this embodiment increases significantly slower.

[0196] The above analysis shows that the use of edge computing and batch anonymous authentication in the crowdsourcing process has certain advantages in reducing communication overhead.

[0197] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for secure delivery of crowdsourcing scenario data in Internet of Vehicles, characterized in that: The steps include: S1. TA generates system parameters and registers the physical OBU and TR; S2, TR generates crowdsourcing tasks and performs task allocation and scheduling; S3,OBU uses anonymous identity to upload encrypted perception data, and RSU implements batch anonymous identity authentication of data sources; S4, data efficient aggregation verification; S5. Tracing the true identity of malicious vehicles.

2. The method for secure delivery of crowdsourcing scene data in Internet of Vehicles according to claim 1 is characterized in that: The S1 step includes the following steps: S11, TA selects a random number Save it as the system master key and calculate P pub =α·P,T pub =β·P as the system public key, and disclose the system public parameters PP={G,P,P pub ,T pub ,H1,H2,H3,n}; S12, vehicle initialization, the vehicle will be real identity RID i Sent as a unique identifier to TA, TA randomly selects As an OBU i The private key of As a public key, TA will Sent to OBU via secure channel i ; S13, TR is initialized, TA generates a homomorphic key pair pk for the data requester TR tr =(n,g d ), sk tr =(μ d ,λ d ) and send it to TR through a secure channel; In step S11, the vehicle initialization process includes the following steps: S111. Input security parameter k. TA generates system parameters through the generation algorithm Gen(k). Then select the multiplication cyclic group G with order p as prime number, P is its generator, select a large prime number q to calculate n=p·q, and define the hash function S112, TA selects a random number And save {α, β} as the system master key, calculate P pub =α·P,T pub =β·P; S113, TA public system public parameters PP = {G, P, P pub ,T pub ,H1,H2,H3,n}.

3. The method for secure delivery of crowdsourcing scene data in Internet of Vehicles according to claim 1 is characterized in that: The S2 step includes the following steps: S21, TR perception task, and determine the type of sensing task T m , the area range of the sensing task L m and the time threshold of the sensing task m , TR generates a data request message {L m ,T m ,Time m ,pk tr } and sent to the crowdsourcing server MCS, where pk tr is the public key of TR; S22, MCS generates a sensing task sequence number T according to the request message id , using it as the identifier of the sensing task and generating a task request message {T id ,T m ,Time m ,pk tr }Send to the sensing task area L m The RSU in the area distributes its broadcast to the sensing vehicles in the area.

4. The method for secure delivery of crowdsourcing scenario data in Internet of Vehicles according to claim 1 is characterized in that: The S3 step includes the following steps: S31, OBU self-generates pseudonym PID i =(PID i,1 ,PID i,2 ) is sent to TA through the channel, TA generates a pseudonym commitment for it, OBU signs it after correctness verification, and performs trustworthy processing on the uploaded data; S32, RSU verifies the legitimacy of the anonymous vehicle identity through the aggregate signature; In step S31, the OBU anonymous transmission process includes the following steps: S311, OBU generates anonymous identity PID i =(PID i,1 ,PID i,2 ), TA generates a pseudonymous commitment for it, which is then signed by OBU after correctness verification; S312, select random number The data is m i , generate data ciphertext S313. Calculate homomorphic hash verification code To facilitate TR to verify the integrity of the data, then Send to RSU; In step S32, the RSU batch authentication process includes the following steps: S321, RSU receives the data packet sent by OBU and first checks the timestamp Is the time threshold of the sensing task reached? m If not, it means that it does not meet the standard of the sensing task and is invalid data, so the relevant data packet is discarded; S322. Verify the identity of the sensing vehicle to ensure the reliability of the data source. The identity authentication process of a single sensing vehicle is the verification equation Is it true? If it is true, the user is a legitimate user. Otherwise, the relevant data ciphertext is discarded. S323: If there are n sensing vehicles participating in the sensing task T id , then RSU performs anonymous batch authentication and verifies the equation If it is established, the reliability of the OBU identity is guaranteed and the data is received; if it is not established, the relevant data is discarded; S324, if the perceived vehicle passes the identity verification, the RSU will Upload to MCS; In step S311, the process of generating an anonymous identity includes the following steps: S3111、OBU i First, pick a random number Then calculate the PID i,1 =γ i ·P, Generate pseudonymous PID i =(PID i,1 ,PID i,2 ), and finally send the pseudonym to TA through a public channel to apply for pseudonym commitment; S3112, TA first uses the pseudonym PID i Trace whether the real identity is legal. If it is a legal user, generate a pseudonym commitment for it. The tracing process is shown in the identity tracing and revocation stage. TA selects a random number Calculate A i =s i ·P, Z i =H2(A i ||PID i ), Generate Pseudonymous Commitment Send it to OBU i ; S3113、OBU i Get this pseudonymous commitment to verify the correctness and judge Is it equal to T pub +Z i P, if equal, indicates that this message is a pseudonymous commitment generated by TA; S3114、OBU i Calculating the Hash in For timestamp, generate signature at the same time Can verify the legitimacy of anonymous identities.

5. The method for secure delivery of crowdsourcing scene data in Internet of Vehicles according to claim 1, characterized in that: The S4 step includes the following steps: S41, MCS aggregates the perceived data ciphertext and data hash verification code, and sends the result to TR; S42, TR uses private key sk tr Decrypt to obtain the plaintext of the perceived data, and use the idea of ​​homomorphic hash function to verify the integrity of the data; In step S41, the MCS data aggregation process includes the following steps: S411. If the sensor data uploaded by n sensing vehicles is valid and the data source is reliable, MCS generates an aggregated ciphertext C and sends it to TR. The data aggregation function is expressed as S412, MCS aggregates the data hash verification code, and the formula is expressed as Finally, the message Send to TR; In step S42, the TR decryption verification process includes the following steps: S421, TR receives the aggregated ciphertext C sent by MCS and uses the private key sk tr Decrypt it to obtain the plaintext data. The decryption formula is: S422, use homomorphic hash function to verify the data, ensure the integrity of the data, calculate Verify equation H' m =H m Is it true? If so, it means the data has not been tampered with; otherwise, it means the data has been changed and is discarded.

6. The method for secure delivery of crowdsourcing scene data in Internet of Vehicles according to claim 1, characterized in that: The S5 step includes the following steps: S51, TR will be the pseudonymous PID of the malicious data source i Send to TA, TA according to the pseudonym PID i =(PID i,1 ,PID i,2 ) and the system private key α to trace the real identity and calculate Get the real identity RID i Then revoke it.

Images