Internet of Vehicles Slice Resource Transaction Method Based on Blockchain

Abstract

A method for trading Internet of Vehicles (IoV) slice resources based on blockchain, in a network scenario where multiple operators collaborate to provide services for the IoV, establishes a network slice resource management architecture, and then conducts resource trading for IoV slices, the process comprising: an IoV service provider publishes requirements in a trading system through a smart contract; a resource provider encrypts data using the public key of the IoV service provider and submits encrypted information; the smart contract collects all bidding information within the bidding time, and the IoV service provider verifies the bidding information through the signature of the resource provider, and decrypts the valid bidding information if the verification is passed, otherwise a penalty bill is broadcast in the blockchain; the IoV service provider selects the final resource provider based on an auction; after being selected, the resource provider submits encrypted token information to the IoV service provider; the IoV service provider decrypts using its private key, obtains control over the resource, and pays; the blockchain completes the verification of the transaction through a credit-based PBFT consensus mechanism.

Description

Technical Field

[0001] The present invention belongs to the field of network technology, and specifically is a method for trading Internet of Vehicles slice resources based on blockchain. Background Art

[0002] With the advancement of 5G network technology and artificial intelligence, the Internet of Vehicles is gradually developing towards efficient perception, intelligent analysis and safe sharing of information between people, vehicles and roads. [1][2] In the era of autonomous driving and connected driving, vehicle services such as vehicle platooning, extended perception, autonomous driving and remote driving [3] These services have different quality of service (QoS) requirements. [4] For example, high-definition map services require sufficient storage resources, real-time positioning technology requires low latency, and audio and video entertainment services require not only communication resources but also certain cache resources. This places higher demands on network resource orchestration and QoS guarantees.

[0003] Network Slicing [5] The key technology is Network Function Virtualization (NFV). [6] and Software Defined Network (SDN) [7] . Through network slicing, a single physical network can be divided into multiple virtual networks optimized for specific services and business objectives. Network slices can be customized according to application or customer needs, and each network slice needs to be able to adapt to different business and communication scenarios to provide reasonable network control and efficient resource utilization. In particular, to achieve high-frequency periodic vehicle connections, ultra-reliable, low-latency communications are essential, and for high-precision map downloads, stability and high throughput are even more important. In addition, network slices are isolated from each other, and the service quality of each slice is not affected by other slices. Therefore, combining network slicing technology with Internet of Vehicles services [8][9] , it can optimize network resource allocation, achieve maximum cost efficiency, and meet the needs of diversified 5G Internet of Vehicles services.

[0004] The Internet of Vehicles architecture based on network slicing relies on efficient multi-dimensional resource orchestration and real-time scheduling. It is difficult for existing operators to independently provide sufficient and diverse resources to meet the diverse needs of Internet of Vehicles services. Providing Internet of Vehicles services anytime, anywhere and across the entire domain often requires the participation of multiple operators. On the one hand, the transaction environment is untrustworthy and opaque, and there are many security risks in real-time resource transactions in such an environment. For example, in traditional centralized resource transactions, transactions between operators are managed by intermediaries, and there is a single point of failure.

[10] , privacy leakage

[11] Even more problematic is that some malicious operators, for their own benefit, may advertise fraudulent resource leasing services when there are not enough available resources, which discourages honest and credible resource providers from participating. As can be seen, achieving credible resource transactions and meeting dynamically changing service needs face severe challenges.

[0005] Blockchain Technology

[12] The emergence of blockchain provides a promising management model for multi-party Internet of Vehicles resource transactions. Compared with traditional centralized architecture, the immutability, auditability and transparency of blockchain help ensure data security in distributed networks.

[13] Applying blockchain technology to network slice transactions, orchestration, and management with multi-party participation can help solve problems such as privacy leakage. However, decentralized networks can cause distrust between network nodes. In order to ensure the reliability of the network, the network system will reach a consensus through relevant protocol negotiations to achieve consistency. Summary of the invention

[0006] In the existing technology, the differentiated quality of service (QoS) requirements of 5G Internet of Vehicles require the assistance of network slicing technology and the collaboration of multiple operators to provide resources. However, the provision of Internet of Vehicles slice resources with the participation of multiple parties is often accompanied by problems such as distrust and deceptive attacks.

[0007] To address these issues, the present invention proposes a blockchain-based 5G Internet of Vehicles slicing resource trading framework, with the aim of building a secure, reliable and traceable Internet of Vehicles resource trading ecosystem.

[0008] The present invention specifically provides: a blockchain-based Internet of Vehicles slice resource trading method, in a network scenario where multiple operators collaborate to provide services for the Internet of Vehicles, a network slice resource management architecture is established, and then resource trading for Internet of Vehicles slices is conducted;

[0009] A. In the network scenario:

[0010] The IoV service provider is responsible for registration and integrating the resources provided by the operator, performing dynamic network slicing, and providing services to vehicles. Based on the vehicle service request, the IoV service provider purchases resources from the operator, slices the resources, and packages them into services for the vehicle.

[0011] As a service requester, the vehicle is equipped with an on-board unit (OBU) to interact with the network slice through the access network;

[0012] Resource providers lease and provide communication resources, cache resources and computing resources to providers of Internet of Vehicles services.

[0013] B. In the network slice resource management architecture:

[0014] The SDN controller is deployed in roadside units or ground base stations to collect user content requests and temporary records and service information;

[0015] Based on the information collected by the local controller, the blockchain predicts the user's historical service requests, defines the slice blueprint and performs end-to-end slicing;

[0016] Slicing is the joint allocation of multiple types of network resources to optimize network utility while meeting the differentiated QoS requirements of customized Internet of Vehicles services;

[0017] The Internet of Vehicles service provider automatically executes transactions with operators through distributed resource transaction smart contracts;

[0018] The network slice orchestrator pools, slices, and orchestrates the purchased resources and provides network slice services to vehicles.

[0019] The local controller assists the SDN controller in network slicing. After the slices are deployed, the corresponding local controller orchestrates the slice resources among the end users.

[0020] The SLA monitoring and violation prediction component monitors the resource utilization and QoS satisfaction of the slice status;

[0021] The local controller collects and analyzes resource monitoring data for SLA violation prediction and uploads it to the blockchain;

[0022] C. Distributed resource transactions for IoV slices are encapsulated in smart contracts and automatically executed. The resource transaction process for IoV slices includes:

[0023] 1.1) The IoV service provider publishes the demand in the transaction system; the smart contract records the demand and broadcasts the message to the entire blockchain; all operators in the blockchain network obtain the demand information;

[0024] Demand information includes demand set Θ, maximum accepted price and credit thresholds

[0025] 1.2) Resource Provider S i The encrypted bidding information submitted is B i ; Use the public key PK0 of the Internet of Vehicles service provider to encrypt the data;

[0026] 1.3) Smart contracts collect all bidding information within the bidding period, and the Internet of Vehicles service provider uses S i The signature of the bidding information is used to verify the bidding information; if the verification is successful, the Internet of Vehicles service provider uses Si Decrypt with the public key and get S i Valid bidding information: If the verification fails, the penalty transaction bill Tran i Will be broadcasted to the blockchain;

[0027] 1.4) The set of resource providers that can pass the verification is S. The Internet of Vehicles service provider selects the set of resource providers that will eventually participate in the transaction based on the auction and credit value.

[0028] 1.5) After being selected as the winner, the relevant resource provider submits the encrypted token information to the Internet of Vehicles service provider. The token SIG i Delegate authorization to access S i The controller that owns the corresponding resource, S i Use the public key PK0 of the Internet of Vehicles service provider to SIG i Encryption to get ESIG i ;

[0029] 1.6) After obtaining the encrypted token information, the Internet of Vehicles service provider uses its private key SK0 to i Decrypt to gain control over resources and pay fees to resource providers;

[0030] Each resource provider has a wallet account to store and manage personal property. During the payment process, a random pseudonym is used as the wallet address of the resource provider's wallet account instead of the real address of the wallet account to protect privacy. The mapping relationship between the wallet account and the corresponding wallet address is recorded in a trusted authority.

[0031] Finally, a record of resource transaction events is generated;

[0032] 1.7) The blockchain completes the verification and confirmation of the transaction through the credit-based PBFT consensus mechanism; finally, the record Tran obtained in step 1.6) is added to the blockchain as a data block;

[0033] The nodes involved in transaction verification are provided by the operators involved in the transaction.

[0034] The present invention proposes a distributed resource trading method based on smart contracts to ensure the fairness and privacy security of transactions; it also proposes a credit evaluation mechanism based on VCG auctions, which serves as a reference when selecting resource traders; finally, it proposes a credit-aware Practical Byzantine Fault Tolerance (PBFT) consensus mechanism.

[0035] Security analysis and simulation results show that the method proposed in the present invention can realize secure and reliable resource transactions, improve node operation efficiency, and reduce the probability of malicious nodes participating in consensus.

[0036] Description of the drawings

[0037] Figure 1 : Schematic diagram of Internet of Vehicles scenario based on network slicing;

[0038] Figure 2 : Schematic diagram of the functional architecture of Internet of Vehicles slice based on blockchain;

[0039] Figure 3 : Schematic diagram of distributed resource transaction process for Internet of Vehicles slicing;

[0040] Figure 4 : Schematic diagram of the impact of credit threshold on slice satisfaction;

[0041] Figure 5 : Schematic diagram of the impact of the maximum accepted price on slice satisfaction;

[0042] Figure 6 :Consensus speed diagram;

[0043] Figure 7 : Schematic diagram of abnormal nodes participating in consensus. DETAILED DESCRIPTION

[0044] The present invention is further described below in conjunction with specific implementation modes. 1. Summary of the Invention

[0046] The present invention considers a scenario where multiple operators collaborate to provide services for the Internet of Vehicles. A trusted third-party organization acts as an Internet of Vehicles service provider, rents existing operator infrastructure resources, including communication, cache, computing resources, etc., and pools and orchestrates resources through virtualization methods to provide satisfactory service quality for the deployment of large-scale Internet of Vehicles applications. In this regard, the present invention proposes a 5G Internet of Vehicles slice resource trading method based on blockchain, with the goal of building a secure, reliable, and fair resource trading ecosystem.

[0047] The present invention establishes a network slice resource management architecture in a network scenario where multiple operators collaborate to provide services for the Internet of Vehicles, and then conducts resource transactions for Internet of Vehicles slices. The process includes: the Internet of Vehicles service provider publishes the demand in the transaction system through a smart contract; the resource provider uses the public key of the Internet of Vehicles service provider to encrypt the data and submit the encrypted information; the smart contract collects all bidding information within the bidding time, and the Internet of Vehicles service provider verifies the bidding information through the signature of the resource provider. If the verification is passed, the valid bidding information is decrypted, otherwise the penalty bill is broadcast in the blockchain; the Internet of Vehicles service provider selects the final resource provider based on the auction; after being selected, the resource provider submits the encrypted token information to the Internet of Vehicles service provider; the Internet of Vehicles service provider uses its private key to decrypt, obtains the control right of the resource, and pays; the blockchain completes the transaction verification through the credit-based PBFT consensus mechanism.

[0048] The main contributions of the present invention are summarized as follows:

[0049] 1) A blockchain-based IoV slice resource transaction management architecture is proposed, which utilizes the characteristics of blockchain such as immutability, traceability, and public auditability to build a secure, reliable, and traceable IoV resource transaction ecosystem.

[0050] 2) A distributed resource transaction smart contract for network slicing is designed and deployed on the blockchain to ensure the fairness and privacy of transactions. At the same time, a credit evaluation mechanism is proposed to provide support for the selection of resource providers.

[0051] 3) Design a credit-based Byzantine fault-tolerant consensus mechanism that works in conjunction with the credit assessment mechanism. This mechanism can vote for accounting nodes based on the credit assessment results of the nodes, thereby enhancing the initiative of trusted nodes and reducing the participation of abnormal nodes.

[0052] The rest is arranged as follows:

[0053] Section 2 introduces the preliminary knowledge in the blockchain-based Internet of Vehicles slice management system.

[0054] Section 3 introduces the system model of the solution of the present invention, mainly starting from the network scenario and functional architecture.

[0055] Section 4 introduces the implementation process of distributed resource transactions for network slicing in detail.

[0056] Section 5 analyzes the performance of the framework proposed in this paper from the perspective of security and verifies it through experimental results.

[0057] Finally, the whole article is summarized.

[0058] 2 Preliminary Knowledge

[0059] 2.1 Blockchain Technology

[0060] Blockchain

[14] It is a peer-to-peer distributed ledger formed by consensus. All activities on the blockchain are trustworthy, responsible, transparent, and can simplify business processes and legal constraints. Asymmetric cryptography and distributed consensus algorithms are used to achieve user security and ledger consistency. The efficient operation of distributed systems depends on an effective consensus mechanism, which is the core of blockchain technology. An efficient consensus mechanism enables the blockchain to form a consistent blockchain structure through effective negotiation. The emergence of blockchain systems has promoted the efficient development of consensus mechanisms. From Proof of Work (PoW) to Proof of Stake (PoS)

[15] , and then to Delegated Proof of Stake (DPoS)

[16] and Practical Byzantine Fault Tolerance (PBFT) consensus mechanism

[17] , the consensus mechanism is being gradually improved.

[0061] Smart Contract

[18] It is an event-driven, stateful code contract and algorithmic contract, which has received extensive attention and research with the in-depth development of blockchain technology. Smart contracts are computer programs that can automatically execute contract terms and are enforced by code. Once started, they will run automatically without any intervention from their initiator. Smart contracts have the characteristics of decentralization, autonomy, observability, verifiability, and information sharing. Smart contracts based on blockchain technology

[19] It can not only give full play to the cost-efficiency advantages of smart contracts, but also prevent malicious behavior from interfering with the normal execution of contracts.

[0062] 2.2 Network Slicing and Service Level Agreement

[0063] Network slicing is a new type of network architecture. Software Defined Network (SDN) and Network Function Virtualization (NFV) are important technologies supporting 5G network slicing.

[20] . SDN aims to provide a higher level of control over network resources through centralized control and management functions. SDN separates the control layer of the core network slice from the user data layer. The control layer is scheduled by a logically centralized programmable controller, and the user data layer is composed of SDN switches. NFV virtualizes network functions and resources to decouple the virtual network elements of the core network slice from the physical layer hardware and configure virtual network elements. Network slicing technology implements virtualized management and maintenance based on network SDN / NFV. Through the SDN / NFV platform, network resources are virtually matched and mapped to generate different logical functions, and then the orchestrator provides corresponding services to meet the network capability requirements of various application scenarios.

[0064] Service Level Agreement (SLA)

[21] As a service guarantee agreement between slice service providers and slice users, it plays a key role in the expansion and commercialization of 5G industry applications. First, operators can have a deeper understanding of the industry's application needs for 5G in the process of participating in the formulation of SLA standards, and develop network slicing services that meet the needs of most industries. Secondly, SLA standards help reduce operators' operating costs. Operators formulate flexible differentiated pricing services based on different SLA standards, and provide a communication platform for negotiating service prices and guarantee levels, which helps to achieve an effective balance between meeting the diversified customized network needs of industry customers and reducing 5G operation and maintenance costs. Finally, SLA standards can improve the service quality of network slicing, enabling each operator to optimize services and enhance competitiveness.

[0065] 3 System Model

[0066] 3.1 Network Scenario

[0067] The present invention considers a network scenario where multiple operators collaborate to provide services for the Internet of Vehicles, such as Figure 1 As shown, it mainly includes the following roles:

[0068] Internet of Vehicles service providers: provide a variety of services for the Internet of Vehicles in this framework, are responsible for registration, integrate resources provided by operators, perform dynamic network slicing, and provide services to vehicles. They are generally trusted third-party organizations. Internet of Vehicles service providers purchase resources from operators based on vehicle service requests, slice the resources, and package them into services for vehicles.

[0069] Vehicle: An entity traveling on the road, generally equipped with an on-board unit that interacts with the network slice through the access network, acting as a service requester in this framework.

[0070] Resource providers: including access networks, core networks, cellular wireless networks, ground data and processing centers, etc. Ground network service operators lease and provide various resource services to providers of Internet of Vehicles services, including communication resources, cache resources, computing resources, etc.

[0071] 3.2 Functional Architecture

[0072] The present invention designs a decentralized network slice resource management architecture based on blockchain, such as Figure 2 shown.

[0073] SDN controllers can be deployed in roadside units or ground base stations. In addition to collecting user content requests, they can also temporarily record some transaction and service information. The blockchain is responsible for predicting the user's historical service requests based on the information collected from the local controller, defining the slice blueprint and end-to-end slicing. Slices should jointly allocate multiple types of network resources (including: communication, computing and caching) to optimize network utility, and at the same time meet the differentiated QoS requirements of customized Internet of Vehicles services. Providers of Internet of Vehicles services automatically execute transactions with operators through distributed resource transaction smart contracts. The network slice orchestrator pools, slices, and orchestrates the purchased resources (including communication, caching, and computing resources), and provides network slice services to vehicles. The local controller is responsible for assisting the SDN controller in network slicing. After the slice deployment is completed, the corresponding local controller is responsible for orchestrating slice resources between end users.

[0074] The SLA monitoring and violation prediction components are used to monitor the resource utilization and QoS satisfaction of the slice status. The local controller collects and analyzes the resource monitoring data for SLA violation prediction and uploads it to the blockchain, which can promptly detect service quality degradation, resource utilization changes, and system configuration errors.

[0075] 4 Distributed resource transactions for Internet of Vehicles slicing

[0076] Distributed resource transactions for IoV slicing are encapsulated in smart contracts and automatically executed. This section will introduce the transaction process, credit evaluation, auction-based resource provider selection, and credit-based PBFT consensus mechanism of resource transactions in detail. The main symbols and variables used are listed in Table 1.

[0077] Table 1 Meaning of main symbols

[0078] Table 1:Summary of major notations

[0079]

[0080]

[0081] 4.1 Transaction Process

[0082] like Figure 3 As shown in the figure, the resource transaction process for Internet of Vehicles slicing includes:

[0083] 1) The IoV service provider publishes the demand in the trading system, including the demand set Θ and the maximum accepted price and credit thresholds The smart contract records the demand and broadcasts the message to the entire blockchain. All operators in the blockchain network can obtain the demand information.

[0084] 2) Resource Provider S i The encrypted bidding information submitted is B i .

[0085]

[0086]

[0087] Indicates that the public key PK0 of the Internet of Vehicles service provider is used to access the data (b i , c i ) is encrypted. The present invention uses an asymmetric encryption algorithm to encrypt the bid information of resource providers. This not only protects the real bidding information, but also ensures that the information will not be leaked to other resource providers. The set of all bidding information of multiple resource providers is represented as

[0088] 3) Smart contracts collect all bidding information within the bidding period Internet of Vehicles service providers use S i Signature To verify the bidding information. If the verification is successful, the Internet of Vehicles service provider uses S i Decrypt with the public key and get S i Valid bidding information:

[0089]

[0090] It is a function that uses the private key SK0 of the Internet of Vehicles service provider for decryption.

[0091] If the verification fails, the penalty transaction bill for the operator will be Tran i It will be broadcasted to the blockchain.

[0092] Tran i =(pun i, signs K0 (pun i )) (4)

[0093] pun i =(cID i , timestamp, PK0, PK i ) (5)

[0094] pun i It includes the pseudonym, private key, timestamp of the resource provider, and the public key of the Internet of Vehicles service provider.

[0095] 4) Auction-based resource provider selection: The set of resource providers that can pass the verification is S, and the Internet of Vehicles service provider selects the set of resource providers that will eventually participate in the transaction based on the auction. The specific algorithm design will be introduced in detail in Section 4.3.

[0096] 5) After being selected as the winner, the relevant resource provider submits the encrypted token information to the Internet of Vehicles service provider.

[0097]

[0098] ESIG i =EncPK0(SIG i ) (7)

[0099] Token SIG i Representatives can authorize access to S i The controller that owns the corresponding resource, S i Use the public key PK0 of the Internet of Vehicles service provider to SIG i encryption.

[0100] 6) After obtaining the encrypted token information, the Internet of Vehicles service provider uses its private key SK0 to i Decryption to gain control over resources and pay fees to resource providers.

[22] Similarly, each resource provider has a wallet account to store and manage personal property. During the payment process, in order to protect privacy, the present invention uses a random pseudonym as the wallet address of the resource provider wallet account to replace the real address of the wallet account to protect privacy. The mapping relationship between the wallet account and the corresponding wallet address is recorded in a trusted authority. Finally, a record of the resource transaction event is generated:

[0101]

[0102] Data i =(timestamp, p i, cID i , PKi, PK0) (9)

[0103] 7) Credit-based PBFT consensus mechanism: The blockchain uses a credit-based PBFT consensus mechanism. The record Tran of resource transaction events is added to the blockchain as a data block. The specific introduction of the consensus process is in Section 4.4. At the same time, in order to make the consensus mechanism lightweight, the nodes participating in transaction verification are all provided by the operators participating in the transaction.

[0104] 4.2 Credit Assessment

[0105] The local controller will regularly collect the detection data of network slices from the SLA detection module, including resource utilization, slice satisfaction, etc. These data are written into the blockchain to form records such as behavioral characteristics. At the same time, the blockchain records the historical behavior of resource providers participating in transactions. The blockchain aggregates these historical behaviors and detection data into a credit value (full score is 100) to provide a reference for dynamic sorting selection in the subsequent consensus process.

[0106] The credit assessment parameter indicators considered in the present invention include:

[0107] 1) Resource reliability index I i1 :Communication reliability is another obstacle that needs to be overcome to achieve satisfaction with connected vehicle services, especially for safety and online vehicle traffic management applications. The calculation method is as follows:

[0108] I i1 =r i 100 (10)

[0109] r i Indicates the communication reliability of the current operator i.

[0110] 2) Average access success rate index I i2 :T i The average access success rate of users accessing slice resources, calculated as follows:

[0111] I i2 =T i 100 (11)

[0112] 3) Average transmission delay index I ia :Data transmission between nodes will cause transmission delay due to various factors. When the average transmission delay L i When it is less than the threshold σ, the present invention considers it tolerable and the node is trustworthy. As the transmission delay exceeds the threshold, the possibility of abnormal behavior of the node increases, which is reflected in the rapid decline of the transmission delay index. The calculation method is as follows:

[0113]

[0114] 4) Credit Index I for Participating in Consensus i4 :The verification process of each transaction is completed by the operator nodes involved in the transaction, and the credit index participating in the consensus is the average of the credit values ​​of all nodes. The calculation method is as follows:

[0115]

[0116] Among them, n i Indicates the number of nodes in the current operator, c ij is the credit value of the jth node in operator i.

[0117] Blockchain standardizes the above evidence data indicators. i1 , I i2 , I i3 , I i4 The corresponding evaluation weight coefficients are r1, r2, r3, r4, (r1+r2+r3+r4=1). Depending on the type of service, the corresponding network slice weight coefficients are also different. The blockchain can obtain the credit value of a resource provider by combining the historical data recorded in the block. Algorithm 1 describes the credit evaluation method.

[0118] The algorithm takes as input the cID of the resource provider i To query its historical behavior data. Introduce the weight coefficient w i , historical transaction data at different times have different weight coefficients. The decay factor m is a constant greater than 0, so older data has a smaller weight coefficient. As time goes by, the impact of historical data on the comprehensive credit value of resource providers will become smaller and smaller. The workload of the algorithm depends on the number of executions of the loop, so the computational complexity of Algorithm 1 is O(N), where N represents the number of recent historical transactions and service provision selected in the blockchain.

[0119]

[0120] 4.3 Auction-based resource provider selection

[0121] The Internet of Vehicles service provider purchases three types of resources (communication, cache, and computing resources) from multiple resource providers according to network slicing requirements.

[23] The dynamic slicing algorithm in is used to realize the provision of Internet of Vehicles services. The Internet of Vehicles service provider publishes the demand Θ = {θ1, θ2, θ3}, where θ1, θ2, θ3 represent the required number of communication, cache, and computing resources respectively. The Internet of Vehicles service provider hopes that more operators can participate in the resource transaction. The set of providers participating in the transaction is represented as S. Each resource provider can provide multiple types of resources. The Internet of Vehicles service provider expects resource providers to and task deadlines When publishing the demand, the Internet of Vehicles service provider will give a credit threshold. and the maximum accepted price The reward received by the resource provider who ultimately participates in the transaction shall not be higher than Credit value must not be less than The operator's credit value and bid are the main factors that Internet of Vehicles service providers need to consider when conducting transactions.

[0122] The local controller will regularly collect the detection data of network slices from the SLA detection module, including resource utilization, QoS completion, etc. The blockchain will record S i The historical transaction behaviors and these detection data are aggregated into credit values ​​through Algorithm 1. The Internet of Vehicles service provider can enter the cID of the resource provider i and the number of historical transactions N to query the credit value of the operator γ i .

[0123] The revenue of Internet of Vehicles service providers from purchasing resources to packaging network slices into services and distributing them to users is It can be expressed as:

[0124]

[0125] Where ψ(W) represents the expected reward for packaging network slices into services and distributing them to users, and W is the set of resource providers that ultimately participate in resource transactions.

[0126] ψ(W)=∑ i∈W v i (15)

[0127] v i =λ·c i ·∑ j∈1，2，3 μ ij (16)

[0128] v in formula (15) i represents the expected benefits of resources provided by any resource provider, v iIt is composed of λ, which is a coefficient that can integrate various resources into monetary rewards. μ1, μ2, and μ3 represent the number of communication resources, computing resources, and cache resources provided by the resource provider, respectively. i The value of is proportional to the operator's credit value. i and c i ′, if c i ≥c i ′, both have v i (c i )≥v i ′(c i ′).

[0129] The goal of Internet of Vehicles service providers is to maximize profits. Operators with different types of resources are the core part of network slicing deployment. They are often selfish but behave rationally.

[24] A Vickrey-Clarke-Groves (VCG) auction mechanism is proposed to determine the bid price b i This auction mechanism achieves efficiency, authenticity and rationality.

[0130] Resource transactions in Internet of Vehicles slices need to solve the problem of resource provider selection. Given a demand set Θ, a set of resource providers participating in the bidding S, and a maximum accepted price and credit thresholds The goal of this problem is to find a subset W, Maximize the benefits of the Internet of Vehicles service provider. This problem can be expressed as the following linear programming:

[0131]

[0132]

[0133]

[0134] Algorithm 2 describes the resource provider selection process. First, the smart contract queries the operator’s bidding information based on the decrypted information and excludes the bidders with unit prices higher than The operator (lines 3-5). Then, its credit value is obtained through the credit evaluation algorithm, which will not meet the credit threshold. The resource providers are put into the supplementary set S - In (lines 6-9), we complement the set S by subtracting it from S - Get the current resource provider set W + (Line 11). Execute the main loop, and the smart contract determines whether the number of resources provided by the current resource provider set W meets the total demand. If not, find the current W +The operator with the highest ratio of credit value to unit resource price is supplemented (line 9) and the current resource quantity is recalculated. + All resources in cannot meet the needs of the Internet of Vehicles service provider, and find the supplementary set S - The operators in the list are supplemented (lines 17-22) until the total demand of Internet of Vehicles service providers is met.

[0135] 4.4 Credit-based PBFT consensus mechanism

[0136] The present invention uses PBFT based on credit

[25] Improvements are made. PBFT mainly consists of a consistency protocol, a view change protocol, and a checkpoint protocol. The present invention defines a credit evaluation to describe the state of a node, and proposes a voting mechanism based on credit value and node state to reward correct nodes and punish incorrect nodes. This can enhance the initiative of trusted nodes and reduce the participation of abnormal nodes. At the same time, in order to make the consensus mechanism lightweight, the nodes involved in transaction verification are all provided by the operators participating in the transaction.

[0137] Set three thresholds from small to large, δ1, δ2, and δ3, and divide the credit value into four credit levels. The specific values ​​of the three thresholds can be divided according to the actual security requirements of the Internet of Vehicles slice. See Table 2:

[0138] Table 2 Credit rating and authority

[0139] Table 2: Credit Level and Authority

[0140]

[0141] The credit-based PBFT consensus process is as follows:

[0142] 1) Select the master node to participate in the consensus based on the voting results. All qualified nodes participate in the vote to select the master node of their organization. The voting includes the options of "support", "oppose" and "abstain". In each election, a node can support (oppose) a node or abstain. The credit value of the qualified node needs to be higher than the threshold δ1. Let N ij N represents node j in operator i. ij The voting results are expressed as

[0143]

[0144] Among them, S ij Represents node N ij Credit rating; c ij YesN ij The credit value; W is the set of operators participating in the transaction; n kis the number of nodes participating in the election of operator k; V ij N ij The voting options include "support", "abstain" and "oppose", with corresponding values ​​of 1, 0 and -1 respectively.

[0145] 2) Master nodes are selected to generate blocks according to a credit-based consensus protocol. "Excellent" nodes have priority in elections, and "good" nodes can be elected as master nodes after all "excellent" nodes have been elected or after no "excellent" nodes have participated in the election. "Passing" nodes cannot be master nodes, but can be deputy nodes. "Invalid" nodes cannot participate in consensus and voting at all. Permission classification effectively prevents "passing" and "invalid" nodes from becoming master nodes, which not only reduces the frequency of view change protocols and communication overhead between nodes, but also reduces the time consumed in generating blocks.

[0146] If a node successfully generates a block, the node is assigned a credit reward. The credit value is updated to When a node fails to generate a block or blocks the generation of a block due to malicious attacks, the node will be punished and the credit value will be updated to The credit rating was changed to "Pass".

[0147] 3) Check whether the master node has timed out. The block generation of the master node is monitored by the timeout monitoring scheme. If the master node is within the given time threshold If the block cannot be produced within 10 seconds, the master node is changed through the view change protocol, otherwise the new block is written into the blockchain. The relationship between nodes in the consensus protocol is defined by the view "v", and the working process of the view change protocol is divided into the following three stages:

[0148] a) “View Change” phase: A replica node enters view “v+1” and broadcasts a “view change” message to all nodes when it determines that the primary node is inactive.

[0149] b) "View Change-Feedback" phase: After a node receives 2f+1 "View Change" messages (including from itself), the new master node enters the "New-View" phase after receiving the "View Change" and "View Change-Feedback" messages.

[0150] c) “New-View” phase: The new master node selects a checkpoint as the starting state for the “new-view” request, and then executes the consistency protocol based on the local blockchain data.

[0151] 4) Checkpoint protocol. The blockchain is verified in chronological order. According to the characteristics of the blockchain, the content and order written on the blockchain are integrated throughout the network and cannot be tampered with or disrupted. When a block is written to the blockchain, the previous verification message has been executed and recorded on the blockchain. Therefore, the previous message record in the local memory is redundant and can be safely deleted.

[0152] 5. Safety Analysis and Performance Evaluation

[0153] 5.1 Security Analysis

[0154] Different from traditional communication security and privacy protection, the 5G Internet of Vehicles slicing resource transaction framework proposed in this invention uses blockchain and smart contract technology to ensure security and privacy protection in the bidding and transaction process. Blockchain ensures the traceability of data, and the automatic execution of smart contracts ensures the safe conduct of transactions. The use of pseudonyms in the bidding and transaction process brings privacy protection to Internet of Vehicles service providers and resource providers. The blockchain-related security performance considered in this invention includes:

[0155] 1) Resist external node attacks: Some malicious operators who are not involved in the framework may steal other people's privacy preference information during the transaction and accounting process of each transaction, thereby destroying the availability of the entire transaction. Therefore, signature authentication, asymmetric encryption algorithms, and encrypted tokens can prevent external nodes from obtaining this private information in the proposed framework.

[0156] 2) Data unforgeability: No attacker can destroy the data in the blockchain. This is because the attacker cannot forge the digital signature of any vehicle or operator, nor can he control most of the network’s resources.

[26] If a small number of vehicles or ground base stations in the blockchain are attacked, hijacked or controlled, the original data will not be accessible because they are all encrypted with keys.

[0157] 3) Secure self-execution: The distributed resource transaction smart contract running on the blockchain is an autonomous, self-executing, and self-maintaining computer code. Smart contracts do not require mutual trust and are completely automatic. Therefore, once the smart contract takes effect,

[27] , no human factors are required and there are no human factors that can be controlled.

[0158] 5.2 Analysis of simulation results

[0159] The present invention establishes a simulation environment consisting of Internet of Vehicles service providers and operators to verify the feasibility of resource transactions. It also simulates the blockchain system to measure the block efficiency of the credit-based PBFT consensus mechanism and the case of malicious nodes participating in the consensus mechanism.

[0160] In order to evaluate the impact of credit threshold and maximum acceptance price on resource trading, a detailed auction algorithm is implemented using MATLAB. In the simulation environment, the ns-3 simulator is used to set up the wireless communication environment between the Internet of Vehicles service provider and the operator node. The Internet of Vehicles service provider uses IEEE 802.11p and TCP protocols to communicate with the operator node.

[28] In Experiment 1, the credit thresholds were set to 60, 70, and 80 to evaluate the slice satisfaction. Other parameters remained unchanged, and the number of operators was changed from 10 to 30 with a step size of 5. Figure 5 In Experiment 2, the maximum accepted price is set to 80, 90, and 100 respectively to evaluate the slice satisfaction. Other parameters remain unchanged, and the number of operators is changed from 10 to 30 with a step size of 5. Figure 6 As shown in the figure, as the credit threshold or the maximum accepted price increases, the slice satisfaction increases. At the same time, the more operators participate in resource trading, the higher the slice satisfaction. Under moderate constraints, the IoV service provider can improve the IoV slice service with higher slice satisfaction while spending a lower price.

[0161] In order to analyze the performance of the credit-based PBFT consensus mechanism, the system environment is configured according to the requirements of Hyperledger Fabric V1.1 and a blockchain infrastructure network is established. The Hyperledger Fabric platform supports both TCP and UDP protocols. A simulation platform consisting of 15 computers is established in the blockchain network, running the Linux operating system, 8GB of memory, an I7-6700 CPU, and a GTX960 graphics card. There are five to twenty numbered nodes in an operator. Transactions are continuously initiated by Node11 in the simulation to test the time it takes for transactions to reach consensus.

[0162] In Experiment 3, the present invention compares PBET with the credit-based PBFT consensus mechanism proposed by the present invention. The present invention sets the number of nodes participating in the consensus to 10 to 50 (with a step size of 2) for testing. Figure 7 It can be seen that as the number of gateway nodes increases, the time required for the PBFT-based solution to reach consensus increases rapidly. The system performance of the PBFT consensus mechanism is greatly affected by the number of nodes, and will be greatly reduced when the number of nodes exceeds the threshold. However, the performance of the solution proposed in this invention remains stable as the number of nodes increases.

[0163] In PBFT, all nodes participate in consensus, and the probability of abnormal nodes becoming master nodes is very high. The consensus mechanism proposed in the present invention uses a voting mechanism to significantly reduce this probability, and selects 3 proxy nodes in each round to participate in the final consensus. The status of "excellent", "good", "passed" or "invalid" nodes is set to 1.1, 1.0, 0.8 and 0.0 respectively. In Experiment 4, the voting results of the nodes Node1, Node2, and Node3 that entered the top three in the first round of consensus were 78, 70, and 68 respectively. When generating blocks, Node1 was found to have malicious block generation behavior, so its status was changed to "passed". In this simulation, voting elections were carried out through 100 repeated experiments, and the election results of the "passed" nodes after the second round of consensus were analyzed. From the simulation results, in 100 elections, the "passed" node only entered the top 3 once, with a probability of less than 2%, which is much lower than PBFT. Therefore, the PBFT consensus mechanism based on credit value improvement of the present invention effectively reduces the probability of abnormal nodes participating in consensus and improves the security of the system.

[0164] 6 Conclusion

[0165] The deep combination of 5G network slicing technology and blockchain technology provides a new direction for services for the Internet of Vehicles. The present invention designs a blockchain-based Internet of Vehicles slice resource trading method. An independent trusted third-party organization for Internet of Vehicles services rents existing operator infrastructure resources, including communication, cache, computing resources, etc. Resources are pooled and orchestrated through virtualization methods to provide various differentiated services for vehicles. The present invention deploys a distributed resource transaction smart contract on the blockchain, collects behaviors during the transaction process, achieves fairness in the transaction process, and ensures the traceability and non-tamperability of the data. In addition, a credit evaluation mechanism is defined to score the operator's historical behavior and records using the slice monitoring data recorded in the blockchain. Finally, the present invention proposes a voting mechanism based on credit and node status, which effectively reduces the participation of abnormal nodes. Compared with the PBFT-based solution, the consensus time of the improved consensus mechanism of the present invention does not increase significantly with the increase in the number of nodes.

[0166] References:

[0167] [1]Hamid U Z A,Zamzuri H,Limbu D K.Internet of vehicle(IoV)applications in expediting the implementation of smart highway of autonomousvehicle:Asurvey[M] / / Performability in Internet of Things.Springer,Cham,2019:137-157.

[0168] [2]Benalia E,Bitam S,Mellouk A.Data dissemination for Internet ofvehicle based on 5G communications:Asurvey[J]. Transactions on EmergingTelecommunications Technologies,2020,31(5):e3881.

[0169] [3]Ning,Z.,Hu,X.,Chen,Z.,Zhou,M.,Hu,B.,Cheng,J.,&Obaidat,M.S.Acooperative quality-aware service access system for social Internet ofvehicles.IEEE Internet of Things Journal,2017,5(4):2506-2517.

[0170] [4]Zhang,W.,Zhang,Z.,&Chao,H.C.Cooperative fog computing for dealingwith big data in the internet of vehicles: Architecture and hierarchicalresource management.IEEE Communications Magazine,2017,55(12):60-67.

[0171] [5]Elayoubi S E,Jemaa S B,Altman Z,et al.5G RAN slicing forverticals:Enablers and challenges[J].IEEE Communications Magazine,2019,57(1):28-34.

[0172] [6]Mijumbi R,Serrat J,Gorricho J L,et al.Management and orchestrationchallenges in network functions virtualization[J].IEEE CommunicationsMagazine,2016,54(1):98-105.

[0173] [7]Yi B,Wang X,Li K,et al.A comprehensive survey of network functionvirtualization[J].Computer Networks,2018,133: 212-262.

[0174] [8]Yang Rong-yue,Zhang Peng-zhou,Song Qing.Research and prospect ofintelligent internet of vehicles based on 5G technology[J].TelecommunicationsScience,2020,36(5):106.

[0175] [9]Lyu F,Yang P,Wu H,et al.Service-oriented dynamic resource slicingand optimization for space-air-ground integrated vehicular networks[J].IEEETransactions on Intelligent Transportation Systems,2021.

[0176]

[10] Bawany N Z,Shamsi J A,Salah K.DDoS attack detection andmitigation using SDN:methods,practices,and solutions[J].Arabian Journal forScience and Engineering,2017,42(2):425-441.

[0177]

[11] Levine B N,Shields C,Margolin N B.A survey of solutions to thesybil attack[J].University of Massachusetts Amherst, Amherst,MA,2006,7:224.

[0178]

[12] Mollah M B,Zhao J,Niyato D,et al.Blockchain for the internet ofvehicles towards intelligent transportation systems:A survey[J].IEEE Internetof Things Journal,2020,8(6):4157-4185.

[0179]

[13] LI Shu,ZHAO Pei-pei,YU Jin-gang,WANG Hai-ting.Research and Designof Data Trading Platform Based on Blockchain[J].Journal of Chinese ComputerSystems 2021,42(5):1109-1114.

[0180]

[14] Zheng Z,Xie S,Dai H,et al.An overview of blockchain technology:Architecture,consensus,and future trends[C] / / 2017 IEEE international congresson big data(BigData congress).Ieee,2017:557-564.

[0181]

[15] Cao B,Zhang Z,Feng D,et al.Performance analysis and comparisonofPoW,PoS and DAG based blockchains[J].Digital Communications and Networks,2020,6(4):480-485.

[0182]

[16] Xu G,Liu Y,Khan P W.Improvement of the DPoS consensus mechanismin Blockchain based on vague sets[J].IEEE Transactions on IndustrialInformatics,2019,16(6):4252-4259.

[0183]

[17] Hao X,Yu L,Zhiqiang L,et al.Dynamic practical byzantine faulttolerance[C] / / 2018IEEE conference on communications and networksecurity(CNS).IEEE,2018:1-8.

[0184]

[18] Ouyang Li-wei,Wang Shuai,Yuan Yong,et al.Smart Contracts:Architecture and Research Progress[J].Journal of Automation,2019,45(3):445-457.

[0185]

[19] He Hai-wu,Yan-an,Chen Ze-hua.Survey of smart contract technologyand application based on blockchain[J]. Computer Research and Development,2018,55(11):2452.

[0186]

[20] LU Xu,YI Bo,WANG Xing-wei,HUANG Min.5G Network Resource SliceManagement Mechanism in Software-defined Networking / Network FunctionVirtualization[J].Journal of Chinese Computer Systems 2021,42(5):1082-1087.

[0187]

[21] Jin L,Machiraju V,Sahai A.Analysis on service level agreement ofweb services[J].HP June,2002,19:1-13.

[0188]

[22] Li Z,Kang J,Yu R,et al.Consortiumblockchainfor secure energytrading in industrial internet of things[J].IEEE transactions on industrialinformatics,2017,14(8):3690-3700.

[0189]

[23] Lyu F,Yang P,Wu H,et al.Service-oriented dynamic resource slicingand optimizationfor space-air-ground integrated vehicular networks[J].IEEETransactions on Intelligent Transportation Systems,2021.

[0190]

[24] Wang X,Chen X,Wu W.Towards truthful auction mechanisms for taskassignment in mobile device clouds[C] / / IEEE INFOCOM 2017-IEEE Conference onComputer Communications.IEEE,2017:1-9.

[0191]

[25] Dinh TTA,Wang J,Chen G,et al.Blockbench:A framework for analyzing private blockchains[C] / / Proceedings of the 2017ACM international conference on management of data.2017:1085-1100.

[0192]

[26] Zyskind G,Nathan O.Decentralizing privacy: Using blockchain to protect personal data[C] / / 2015IEEE Security and Privacy Workshops.IEEE,2015:180-184.

[0193]

[27] Yue L, Junqin H, Shengzhi Q, et al. Big data model of securitysharing based on blockchain [C] / / 2017 3rd International Conference on Big DataComputing and Communications (BIGCOM). IEEE, 2017: 117-121.

[0194]

[28] Na W, Dao NN, Cho S. Mitigating WiFi interference to improve throughput for in-vehicle infotainment networks[J]. IEEE WirelessCommunications, 2016, 23(1):22-28.

[0195] Attached Chinese references:

[0196] [8] Yang Rongyue, Zhang Pengzhou, Song Qing. Research and prospects of intelligent vehicle networking based on 5G technology[J]. Telecommunications Science, 2020, 36(5):106.

[0197]

[13] Li Shu, Zhao Peipei, Yu Jingang, Wang Haiting. Research and design of data trading platform based on blockchain[J]. Small and Micro Computer Systems, 2021, 42(5): 1109-1114

[0198]

[18] Ouyang Liwei, Wang Shuai, Yuan Yong, et al. Smart contracts: architecture and progress[J]. Acta Automatica Sinica, 2019, 45(3): 445-457.

[0199]

[19] He Haiwu, Yan An, Chen Zehua. A review of smart contract technology and applications based on blockchain[J]. Computer Research and Development, 2018, 55(11): 2452.

[0200]

[20] Lu Xu, Yi Bo, Wang Xingwei, et al. 5G network resource slice management mechanism under SDN / NFV[J]. Small and Micro Computer Systems, 2021, 42(5): 1082-1087.

Claims

1. A method for trading Internet of Vehicles slice resources based on blockchain, characterized in that a network slice resource management architecture is established in a network scenario where multiple operators collaborate to provide services for the Internet of Vehicles, and then resource transactions for Internet of Vehicles slices are conducted; A. In the network scenario: The IoV service provider is responsible for registration and integrating the resources provided by the operator, performing dynamic network slicing, and providing services to vehicles. Based on the vehicle service request, the IoV service provider purchases resources from the operator, slices the resources, and packages them into services for the vehicle. As a service requester, the vehicle is equipped with an on-board unit (OBU) to interact with the network slice through the access network; Resource providers lease and provide communication resources, cache resources and computing resources to providers of Internet of Vehicles services; B. In the network slice resource management architecture: The SDN controller is deployed in roadside units or ground base stations to collect user content requests and temporary records and service information; Based on the information collected by the local controller, the blockchain predicts the user's historical service requests, defines the slice blueprint and performs end-to-end slicing; Slicing is the joint allocation of multiple types of network resources to optimize network utility while meeting the differentiated QoS requirements of customized Internet of Vehicles services; The Internet of Vehicles service provider automatically executes transactions with operators through distributed resource transaction smart contracts; The network slice orchestrator pools, slices, and orchestrates the purchased resources and provides network slice services to vehicles. The local controller assists the SDN controller in network slicing. After the slices are deployed, the corresponding local controller orchestrates the slice resources among the end users. The SLA monitoring and violation prediction component monitors the resource utilization and QoS satisfaction of the slice status; The local controller collects and analyzes resource monitoring data for SLA violation prediction and uploads it to the blockchain; C. Distributed resource transactions for IoV slices are encapsulated in smart contracts and automatically executed. The resource transaction process for IoV slices includes: 1.1) The IoV service provider publishes the demand in the transaction system; the smart contract records the demand and broadcasts the message to the entire blockchain; all operators in the blockchain network obtain the demand information; Demand information includes demand set Θ, maximum accepted price and credit thresholds 1.2) Resource Provider S i The encrypted bidding information submitted is B i , where the data is encrypted using the public key PK0 of the Internet of Vehicles service provider; 1.3) Smart contracts collect all bidding information within the bidding period, and the Internet of Vehicles service provider uses S i The signature of the bidder is used to verify the bidding information; If the verification is successful, the Internet of Vehicles service provider uses S i Decrypt with the public key and get S i Valid bidding information: If the verification fails, the penalty transaction bill Tran i Will be broadcasted to the blockchain; 1.4) The set of resource providers that can pass the verification is S. The Internet of Vehicles service provider selects the set of resource providers that will eventually participate in the transaction based on the auction and credit value. 1.5) After being selected as the winner, the relevant resource provider S i Submit encrypted token information to the Internet of Vehicles service provider, token SIG i Delegate authorization to access S i The controller that owns the corresponding resource, S i Use the public key PK0 of the Internet of Vehicles service provider to SIG i Encryption to get ESIG i ; 1.6) After obtaining the encrypted token information, the Internet of Vehicles service provider uses its private key SK0 to i Decrypt to gain control over resources and pay fees to resource providers; Each resource provider has a wallet account to store and manage personal assets; During the payment process, a random pseudonym is used as the wallet address of the resource provider wallet account to replace the real address of the wallet account to protect privacy; the mapping relationship between the wallet account and the corresponding wallet address is recorded in a trusted authority; Finally, a record of resource transaction events is generated; 1.7) The blockchain completes the verification and confirmation of the transaction through the credit-based PBFT consensus mechanism; finally, the record Tran obtained in step 1.6) is added to the blockchain as a data block; The nodes involved in transaction verification are provided by the operators involved in the transaction.

2. The method for trading Internet of Vehicles slice resources based on blockchain according to claim 1 is characterized in that In the step 1.2), the bidding information of the resource provider is encrypted using an asymmetric encryption algorithm.

3. The method for trading Internet of Vehicles slice resources based on blockchain according to claim 1 is characterized in that Credit Value i The evaluation is: The local controller regularly collects the detection data of the network slice by the SLA detection module; writes the detection data into the blockchain to form records such as behavioral characteristics; at the same time, the blockchain records the historical behavior of resource providers participating in transactions; The blockchain aggregates these historical behaviors and detection data into a credit value, with a full credit value of 100; the detection data includes resource utilization and slice satisfaction; Credit assessment parameters include: ① Resource reliability index I i1 :I i1 =r i 100, r i Indicates the communication reliability of the current operator i; ② Average access success rate index I i2 :I i2 =T i 100, T i The average access success rate of users accessing slice resources; ③Average transmission delay index I i3 : When the average transmission delay L i When it is less than the threshold σ, it is considered to be tolerable and the node is trustworthy; when the transmission delay exceeds the threshold, the possibility of abnormal behavior of the node is also increasing, which is reflected in the rapid decline of the transmission delay index; ④ Credit Index I of Participating Consensus i4 : The verification process of each transaction is completed by the operator nodes participating in the transaction, and the credit index participating in the consensus is the average credit value of all nodes; where n i Indicates the number of nodes in the current operator, c ij is the credit value of the jth node in resource provider / operator i; Blockchain standardizes the above-mentioned evidence data indicators of credit assessment parameter indicators; i1 , I i2 , I i3 , I i4 The corresponding evaluation weight coefficients are r1, r2, r3, r4, (r1+r2+r3+r4=1) respectively; the corresponding network slice weight coefficients are different according to different types of services; the blockchain combines the historical data recorded in the block to obtain the credit value γ of the resource provider i i .

4. The method for trading Internet of Vehicles slice resources based on blockchain according to claim 3 is characterized in that In step 1.4), the resource provider selection method based on auction and credit value is: The Internet of Vehicles service provider purchases three types of resources, namely communication, cache, and computing resources, from multiple resource providers according to network slicing requirements; a dynamic slicing algorithm is used to provide Internet of Vehicles services; each resource provider participating in the transaction provides multiple types of resources; IoV service providers expect resource providers to and task deadlines Participate in bidding; Internet of Vehicles service providers give a credit threshold when publishing requirements and the maximum accepted price The reward received by the resource provider who ultimately participates in the transaction shall not be higher than And the credit value is not less than Resource provider / operator credit value γ i and bidding are the main factors that Internet of Vehicles service providers need to consider when conducting transactions.

5. The method for trading Internet of Vehicles slice resources based on blockchain according to claim 3 is characterized in that In step 1.7), the establishment of the credit-based PBFT consensus mechanism is as follows: PBFT includes a consensus protocol, a view change protocol, and a checkpoint protocol. It defines a credit evaluation to describe the state of a node and proposes a voting mechanism based on credit value and node state to reward correct nodes and punish incorrect nodes. Set three thresholds δ1, δ2, and δ3 from small to large, and divide the credit value into four credit levels. The specific values ​​of the three thresholds are divided according to the actual security requirements of the Internet of Vehicles slice, and the credit levels and permissions are divided into: The credit-based PBFT consensus process is as follows: First, the master node is selected to participate in the consensus based on the voting results. All qualified nodes participate in the voting to select the master node of their organization. The voting options include "support", "oppose" and "abstain"; In each election, nodes support / oppose a node, or abstain; the credit value of qualified nodes needs to be higher than the threshold δ1; let N ij represents node j in resource provider / operator i, N ij The voting results are expressed as Among them, S ij Represents node N ij Credit rating; c ij YesN ij The credit value; W is the set of operators participating in the transaction; n k is the number of nodes participating in the election of operator k; V ij N ij The voting options include "support", "abstain" and "oppose", with corresponding values ​​of 1, 0 and -1 respectively; Then, the master node is selected to generate blocks according to the credit-based consensus protocol. "Excellent" nodes have priority during elections. "Good" nodes are elected as master nodes after all "excellent" nodes are elected or no "excellent" nodes participate in the election. "Passing" nodes cannot be master nodes, but serve as secondary nodes. "Invalid" nodes cannot participate in consensus and voting. If a node successfully generates a block, the node is assigned a credit reward. The credit value is updated to When a node fails to generate a block or blocks the generation of a block due to malicious attacks, the node will be punished and the credit value will be updated to The credit rating was changed to "Pass"; Next, check whether the master node has timed out. The block generation of the master node is monitored by a timeout monitoring scheme; if the master node If the block cannot be produced within 10 seconds, the master node is changed through the view change protocol, otherwise the new block is written into the blockchain; the relationship between nodes in the consensus protocol is defined by the view "v", and the working process of the view change protocol is divided into the following three stages: a) "View Change" phase: A replica node enters view "v+1" and broadcasts a "view change" message to all nodes when it determines that the primary node is inactive; b) "View Change-Feedback" phase: After a node receives 2f+1 "View Change" messages, the new master node enters the "New-View" phase after receiving the "View Change" and "View Change-Feedback" messages; c) "New-View" phase: The new master node selects a checkpoint as the starting state of the "New-View" request, and then executes the consistency protocol based on the local blockchain data; Finally, the checkpointing protocol The blockchain is verified in chronological order; according to the characteristics of the blockchain, the content and order written on the blockchain are integrated throughout the network; when a block is written to the blockchain, the previous verification message has been executed and recorded on the blockchain; therefore, the previous message record in the local memory is redundant and can be deleted.

Images