Blockchain technology protects the communication framework of internet of vehicles and time delay optimization method

Abstract

The application relates to a blockchain technology protection vehicle networking communication framework and time delay optimization method, which comprises the following steps: S1, constructing a system based on a blockchain environment between edge devices; S2, calculating the time delay T of vehicle data extraction according to a Poisson point arrival process i ; S3, constructing a channel state model; S4, calculating the system time delay; S5, optimizing the system overall time delay E[T], and obtaining the most suitable lambda * by convex optimization, and adjusting the extraction data rate of the vehicle so that the system overall time delay E[T] is minimized; the application introduces a blockchain Raft consensus mechanism to ensure the integrity and safety of data transmission, dynamically adjusts the vehicle data extraction rate through a convex optimization algorithm, effectively reduces the system time delay, and meets the requirements of low delay and high reliability of vehicle networking; the application also provides a new idea for the application of the blockchain technology in the intelligent transportation field, and has important theoretical and practical values.

Description

Technical Field

[0001] This invention relates to the field of machine learning technology for vehicle networking, and in particular to a blockchain technology-based framework for protecting vehicle networking communication and a latency optimization method. Background Technology

[0002] With the continuous development of vehicle intelligence and connectivity, the amount of data that vehicles need to process has increased dramatically. This includes sensor data, map information, user interaction data, and so on, which places higher demands on the vehicle's computing power. However, due to the limitations of onboard hardware and power consumption, the computing power of a single vehicle is insufficient, and the amount of data that can be processed is limited, which cannot meet the computing services required for intelligent vehicle tasks. To solve this problem, we must rely on Internet of Vehicles (IoV) technology.

[0003] Vehicle-to-everything (V2X) technology can effectively support communication between vehicles and everything. With the development of mobile edge computing technology, when the vehicle's own computing power is insufficient to handle complex local tasks, the V2X can offload the computationally intensive parts to roadside units and base stations, thereby reducing the vehicle's local computing burden. With the development of large-scale artificial intelligence models and their gradual application in V2X, the vehicle's local computing power is unprecedentedly strained. Many cutting-edge architectures have proposed using edge computing to provide AI services to vehicles. However, when using AI to process local tasks, a large amount of private data transmission is involved. Many attackers may steal or tamper with information during data transmission, seriously affecting vehicle tasks and even causing serious security risks.

[0004] In vehicle networks, blockchain technology can establish secure data connections, solving the problem of data transmission security. When deploying blockchain technology, servers are typically set up as nodes in the blockchain network. These servers usually have sufficient computing resources and storage space to handle large amounts of data and computational tasks. Servers collect various data from vehicles, package them into blocks, and form a reliable chain-like data structure, making the data in any block difficult to modify. Therefore, blockchain technology can ensure the security of vehicle networks to a certain extent. Traditional blockchain technologies, such as the Paxos consensus algorithm, are known for being difficult to understand and accurately implement. The Raft consensus mechanism, proposed by Diego Ongaro and John Ousterhout, is a simple and easy-to-understand distributed consensus algorithm that is no less secure than the Paxos algorithm. Based on this, combining the Raft consensus mechanism with edge computing technology can solve the problem of insufficient local computing power in vehicles, while ensuring the secure transmission of information in the network.

[0005] However, the introduction of blockchain technology will bring additional time delay to vehicle-to-everything (V2X) communication. In order to meet the requirements of ultra-low latency and ultra-high reliability of vehicle communication, it is necessary to study the latency composition of the system and control it within an appropriate range through optimization methods. Summary of the Invention

[0006] In view of this, the present invention proposes a blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and a latency optimization method, which achieves efficient data transmission and processing capabilities while reducing system latency, thereby meeting the stringent requirements of V2X for low latency and high reliability.

[0007] To achieve the above objectives, this invention provides a blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and a latency optimization method, comprising the following steps:

[0008] S1. Build a blockchain-based system among edge devices, deploying N servers. The states of the N servers include leaders, followers, and candidates, with the initial state being candidates.

[0009] S2, according to the process of reaching the Poisson point Calculate the time delay T for vehicle data extraction i ;

[0010] S3. Construct a channel state model;

[0011] S4, computing system latency;

[0012] S401, Data extraction delay T ex T ex The expectation is Calculate the delay T when the vehicle transmits data. m The expression is:

[0013] T m =T ex +T ec

[0014] Among them, T ec This indicates the latency of the vehicle's initial local data processing.

[0015] S402. Calculate the data transmission delay T of M vehicles connected to an edge device due to data collisions, and update the data transmission delay of each vehicle. m′ And calculate the latency T of the data sent by the vehicle arriving at the edge device. ar ;

[0016] S403. Considering the latency caused by the edge device demodulating vehicle data and uploading the decoding amount, calculate the latency T;

[0017] S404, Calculate the additional time overhead E[T] incurred due to a server attack on the leader election process.ele ];

[0018] S405, Obtain the overall system delay E[T];

[0019] S5. Optimize the overall system delay E[T] by finding the extreme point through convex optimization differentiation to obtain the most suitable λ, and adjust the vehicle data extraction rate to minimize the overall system delay E[T].

[0020] Preferably, the N servers elect a leader through random voting, the leader generates an origin block to form a chain, and a timer is set to periodically increment the chain during period T. term Internal broadcast control information.

[0021] Preferably, the vehicle data extraction process A includes point {A} i}, i≥1 are non-negative real numbers, according to point {A i The arrival times are arranged in ascending order, i.e., A1≤A2≤A3≤…, and the arrival delay T is… i The expression is:

[0022]

[0023] Preferably, the construction of the channel state model includes: calculating the channel state preservation probability based on vehicle speed and channel parameters;

[0024] Channel states include ideal states and error states. The channel state preservation probability expression is:

[0025]

[0026] Among them, P i P represents the probability that an ideal channel will remain in an ideal state. p Let η represent the probability that the error channel remains in the error state, and let η represent the medium coefficient. The expression for η is:

[0027]

[0028] Where F represents the fading margin, The average probability of frame transmission failure is expressed as: ρ represents the Doppler frequency f. d The Gaussian correlation coefficient ρ between two samples of the fading channel amplitude is expressed as:

[0029] ρ=J0(2πf d / θ)

[0030] Where J0(.) represents the zeroth-order Bessel function, 1 / θ represents the frame transmission time on the channel, and f d f represents the Doppler frequency. dThe expression is:

[0031] f d =f c v / c

[0032] Among them, f c denoted by carrier frequency, v represents vehicle speed, and c represents the speed of light.

[0033] Preferably, the construction of the channel state model further includes: calculating the probability p of discarding the channel state update. d The expression is:

[0034]

[0035] Among them, v L l represents the probability of at least one frame transmission failure in an error channel. L This represents the probability of failure of at least one frame in the ideal new track.

[0036] Preferably, the vehicle transmits data with a delay T. m′ The expression is:

[0037]

[0038] Where, p c Indicates the probability of collision;

[0039] The latency T of data transmitted by the vehicle to the edge device ar The expression is:

[0040]

[0041] Where, p d This indicates the probability that a channel state update is discarded.

[0042] Preferably, the expression for the time delay T is:

[0043] T = T ar +T dc +T f +T p +T ele

[0044] Among them, T dc T represents the latency incurred by the edge device in decoding vehicle data. f T represents the delay T caused by the follower uploading the decoded variable to the leader. p T represents the time delay in which the leader communicates a message to all followers. ele This indicates the time delay in the election of leaders.

[0045] Preferably, the average time overhead E[T] for electing an additional leader due to a server attack is... ele The expression for ] is:

[0046]

[0047] Where, τ ele T represents the time it takes to elect a leader. term This represents the preset time period, E[T] represents the expected latency T, and a represents the number of servers under attack.

[0048] The expression for the overall system delay E[T] is:

[0049]

[0050] Among them, T si This represents the latency of data processing and transmission.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] 1. This invention offloads the vehicle's computing tasks to the server through edge computing, reducing the vehicle's local computing burden, while optimizing server resource allocation and improving the overall system operating efficiency.

[0053] 2. This invention uses blockchain technology to record all communication data between all vehicles and servers in a distributed ledger, ensuring data integrity and immutability; even if some servers are attacked, data security and integrity can still be guaranteed.

[0054] 3. This invention introduces the Raft consensus mechanism, which ensures that the system can still operate normally even if some nodes are attacked through the division of roles between leaders and followers and the periodic leader election mechanism. As long as more than half of the nodes are working normally, the system can maintain stability, effectively resist external attacks, and enhance its anti-attack capability.

[0055] 4. This invention proposes an online optimization algorithm based on convex optimization to dynamically adjust the vehicle data extraction rate and significantly reduce the overall system latency. Experimental results show that the system can achieve low latency performance requirements under different vehicle connection numbers and attack intensities through the optimization algorithm. Attached Figure Description

[0056] Figure 1 This is a diagram illustrating the communication environment in the vehicle-to-everything (V2X) network of this invention.

[0057] Figure 2 This is a diagram illustrating the identity transformation and control information broadcasting of Raft, the invention.

[0058] Figure 3This is a diagram of the vehicle and server channel model of the present invention;

[0059] Figure 4 This is a comparison chart showing the impact of the maximum number of connected vehicles on system latency.

[0060] Figure 5 This is a comparison chart showing the impact of attack intensity on system latency under different data extraction rates according to the present invention;

[0061] Figure 6 This is a comparison chart showing the maximum number of connected vehicles on the server and the impact of the number of vehicles on system latency under different data extraction rates according to the present invention. Detailed Implementation

[0062] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0063] This embodiment considers a straight-ahead road scenario, where RSUs or BSs are deployed on both sides of the road to assist vehicle computation. Vehicles need to offload some computational tasks to edge-deployed devices to alleviate local computational burden. To ensure information security and prevent tampering of vehicle-sent information, a blockchain environment is built between the edge devices (i.e., BS and RSU), employing the Raft consensus mechanism to defend against external attacks. A detailed description follows... Figure 1 As shown, each vehicle communicates with an edge device; the edge device can change its identity between three roles (follower, candidate, and leader); the leader is responsible for receiving vehicle requests and creating new log entries, which are uniquely added to the leader's log; the follower is responsible for recording and storing log information from the leader and responding to the leader's requests when necessary; the candidate can request votes from other nodes during the election process to establish its eligibility as a new leader, and is an intermediate role in the transition from follower to leader.

[0064] The vehicle first needs to extract data, and this process will be considered a Poisson process. Communication between the vehicle and the server also needs to consider error-prone channels based on the 3GPP LTE Cat.M1 protocol. In this protocol, the vehicle's speed has a significant impact on communication performance, achieved through the Doppler effect. The Doppler effect causes a change in frequency due to the object's speed, which affects the frequency of communication transmission. This frequency change affects the vehicle's communication performance. This invention will provide a detailed explanation of the communication channel analysis and calculate the probability of data loss during vehicle transmission and the optimal extraction rate. Furthermore, the number of servers is N, and the maximum number of vehicles that can be connected to the servers is M.

[0065] Figure 2 The text explains in detail how the three roles in the Raft mechanism transition: First, servers are divided into three roles: maintainers (followers), intermediate nodes (candidates), and core nodes (leaders). Maintainers need to send the information they record to the core node. An intermediate node is an intermediary role; it transforms from a maintainer node and waits to become a core node. Once a core node is selected from among the intermediate nodes, the remaining intermediate nodes become maintainers. The core node is the most crucial role and has its own term. It collects the information recorded by the other maintainers and packages it into blocks. In addition, it needs to periodically broadcast its control information to show its presence. Once the control information disappears, it means that the core node's term has ended or it has been maliciously attacked. At this point, to ensure the robustness and security of the network, a new core node needs to emerge. However, the other maintainers who originally received the control information no longer receive it and thus understand that the core node does not exist. They then become intermediate nodes to participate in the election, eventually producing a new, healthy core node.

[0066] This mechanism ensures that the core server—the core node—remains robust, thus improving the overall robustness of the framework.

[0067] To reduce latency in vehicle-to-everything (V2X) systems and ensure efficient data transmission and processing capabilities, this invention provides the following technical solution: a blockchain-based framework for protecting V2X communication and a latency optimization method.

[0068] An experimental platform was built and simulation experiments were conducted. The stability and latency optimization effects of the proposed blockchain-supported vehicle networking framework were verified by adjusting the simulation parameters. The experimental platform was built by integrating Python and C++. Python was used to implement the variational information bottleneck algorithm, while C++ was used to build a faster blockchain environment. The modified National Institute of Standards and Technology (MNIST) dataset was used, which contains 60,000 training data and 10,000 test data.

[0069] Includes the following steps:

[0070] S1. Build a blockchain-based system among edge devices, deploying N servers. The N servers have states including leader, followers, and candidates, initially as candidates. The N servers elect a leader through random voting, and the remaining candidates become followers, initially receiving random rewards. Once a leader is elected, the leader generates an origin block to form a chain, and a timer is set to periodically expire during period T. term Internal broadcast control information.

[0071] S2. Define vehicle data extraction process A, following the Poisson point arrival process. Calculate the time delay T for vehicle data extraction i ;

[0072] Poisson point arrival process It has the following two attributes:

[0073] (1) A point A([a,b)) with random numbers in the process lies in a bounded interval [a,b). The average of a Poisson random variable λ([a,b)) can be used as a Radon measure (non-negative).

[0074] (2) The points of process A fall within the interval [a1,b1), [a2,b2), ..., [a... k ,b k Form k independent random variables λ([a1,b1),[a2,b2),…,[a... k ,b k ));

[0075] For a point process A defined on non-negative real numbers, including the point {A} i}, i≥1 are non-negative real numbers, according to point {A i Arrival times are arranged in ascending order, i.e., A1≤A2≤A3≤…, with arrival delay T. i The expression is:

[0076]

[0077] In the system of this embodiment, there are the following two situations:

[0078] Homogeneous Poisson process: In {A i For a homogeneous Poisson process with speed Λ, the corresponding arrival delay is... Since they are independent, identically distributed exponential random variables with an average of 1 / Λ, the probability density function of the arrival process can be expressed as follows:

[0079] P(T i ≤t)=1-e -Λt

[0080] Non-homogeneous Poisson process: For {A} i For a non-homogeneous Poisson process with intensity Λ(t), the corresponding arrival rate estimation method is as follows:

[0081] The expression for the first arrival time T1 = A1 distribution is:

[0082]

[0083] Given the initial arrival delay T1 = A1, the distribution expression for the second conditional arrival time T2 is:

[0084]

[0085] Similarly, we can obtain the arrival times T for the i-th, 3rd, 4th, ...th arrival times. i The distribution expression;

[0086] For a nonhomogeneous Poisson process {A} i Each point is independently distributed in distribution P(A). i In the interval ≤a), P(A) i The expression for the ≤a) distribution is:

[0087]

[0088] S3. Construct a channel state model;

[0089] Communication latency is caused by data transmission between the vehicle and the server, and involves two distinct processes: uploading the potential space vector and backpropagation iteration. In the blockchain-based system proposed in this embodiment, estimating uploading and backpropagation latency by analyzing communication latency requires accurate analysis of the underlying cellular network (such as V2V and V2X LTE) channel dynamics. This embodiment considers and develops an approximate analysis of channel state dynamics from the work of Pokhrel et al., and includes basic details.

[0090] Channel states are divided into two types: ideal and poor. Frames transmitted in an ideal channel will not fail, while transmissions in a poor channel have a probability of failure. The states of both ideal and poor channels can change; the probability of an ideal channel remaining ideal is P. i The probability of the error channel maintaining the error is P. P Therefore, the probability of switching from the ideal channel to the error channel is 1-P. i The probability of the error channel transitioning to the ideal is 1-P. P It is particularly important to note that the mobility of the vehicle itself will affect the communication performance. In the section on mathematical modeling of the 0 model, this invention considers the impact of vehicle speed on communication as the impact of vehicle speed on the channel, because the Doppler effect caused by vehicle speed affects the preservation and transfer probability of the error channel and the ideal channel.

[0091] θ represents the maximum rate at which the link layer frames can be transmitted to the vehicle (θ = bit rate / frame bit rate), v represents the vehicle speed, and f c The carrier frequency and Doppler frequency f are represented by f. d The expression is:

[0092] fd =f c v / c

[0093] Considering F as the fading margin, and given the physical modulation and coding scheme, if the received signal-to-noise ratio is below a threshold:

[0094] SNR threshold =E[SNR] / F

[0095] If the channel is in poor condition, it is considered to be in ideal condition; otherwise, it is considered to be in ideal condition. The average probability of frame transmission failure due to errors in the channel. for:

[0096]

[0097] The expression for the dielectric coefficient η is:

[0098]

[0099] Where ρ represents the Doppler frequency f d The Gaussian correlation coefficient ρ between two samples of the fading channel amplitude is expressed as:

[0100] ρ=J0(2πf d / θ)

[0101] Where J0(.) represents the zeroth-order Bessel function, 1 / θ represents the frame transmission time on the channel, and f d Indicates the Doppler frequency.

[0102] Channel states include ideal states and error states. The channel state preservation probability expression is:

[0103]

[0104] Among them, P i P represents the probability that an ideal channel will remain in an ideal state. p η represents the probability that the error channel remains in the error state, and η represents the medium coefficient.

[0105] Calculate the probability p that an update in the channel is discarded. d The expression is:

[0106]

[0107] Among them, v L l represents the probability of at least one frame transmission failure in an error channel. L This represents the probability of failure of at least one frame in the ideal new track.

[0108] S4, computing system latency;

[0109] S401, the vehicle needs to extract data, which means there is a data extraction delay T. ex According to the homogeneous Poisson property, T ex The expected expression is:

[0110]

[0111] Considering the latency T of the vehicle's local preliminary data processing ec Then, the data transmission delay T is... m The expression is:

[0112] T m =T ex +T ec

[0113] S402. Calculate the data transmission delay T of M vehicles connected to an edge device due to data collisions, and update the data transmission delay of each vehicle. m′ And calculate the latency T of the data sent by the vehicle arriving at the edge device. ar ;

[0114] A server is connected to M vehicles. Data transmission between these M vehicles has a probability of collision. The LTE CATM1 specification stipulates an interval of 2 to 3 time slots to avoid collisions. This embodiment uses 3 time slots:

[0115] τ c =3τ t

[0116] Where, τ t Represents the time slot interval, and the collision probability p. c The expression is:

[0117]

[0118] in, Let be the time it takes for any two vehicles to send data. This is the Poisson process delay plus a constant, and it is still a Poisson process. Therefore:

[0119]

[0120] Assuming the vehicle retrieves data again after a collision, the delay T for the updated vehicle to send the data is... m′ The expression is:

[0121]

[0122] Also considering the probability of transmission failure in the channel, the delay T of data arriving at the server ar The expression is:

[0123]

[0124] Where, p d This indicates the probability that a channel state update is discarded.

[0125] S403. Considering the latency caused by the edge device demodulating vehicle data and uploading the decoding data, calculate the latency T. The expression for latency T is:

[0126]

[0127] Among them, T dc T represents the latency incurred by the edge device in decoding vehicle data. f T represents the delay T caused by the follower uploading the decoded variable to the leader. p T represents the time delay in which the leader communicates a message to all followers. ele This indicates the time delay in the election of leaders;

[0128] S404, Calculate the additional time overhead E[T] incurred due to a server attack on the leader election process. ele ];

[0129] Assuming no attacks occur and each term is not interrupted, the average election delay per epoch is expressed as follows:

[0130]

[0131] Where, τ ele The time allotted for electing a leader, without loss of generality, is determined by keeping the control information interval, carrier frequency, and other conditions constant.

[0132] τ ele =τ b log N

[0133] Where N represents the number of servers, τ b Indicates the election time constant;

[0134] For an attack of strength of 1 This means that one of the N servers has been attacked and its computing power has been paralyzed. In this embodiment's architecture, as long as the paralyzed server is not the leader, the impact on the system is relatively small. If the leader is paralyzed, its control information stops, and the election will begin immediately. Assuming that each server has an equal probability of being attacked, the probability of the leader being attacked is [value missing]. This results in additional election time overhead, and the election delay expression is:

[0135]

[0136] Where, τ eleT represents the time it takes to elect a leader. term This represents the preset time period, E[T] represents the expected latency T, and a represents the number of servers under attack.

[0137] S405. Obtain the overall system delay E[T], expressed as:

[0138]

[0139] T si =T dc +T f +T p

[0140] Among them, T si This represents the probability density of the arrival process;

[0141] S5. Optimize the overall system delay E[T]. The expression for the overall system delay E[T] is a convex optimization function with respect to λ. Take the derivative of the overall system delay E[T] with respect to λ to find the extreme point. Adjust the vehicle data extraction rate to minimize the overall system delay E[T].

[0142] shilling:

[0143]

[0144] B=M(M-1)τ c / 2

[0145] but:

[0146]

[0147] Differentiation yields:

[0148]

[0149] We can obtain:

[0150]

[0151] Then there is one and only one:

[0152] (BT ec λ 2 +Bλ-1)=0

[0153] E ′ Only when [T] is 0 can E[T] reach an extreme value;

[0154] According to Vieta's formulas:

[0155]

[0156] Since λ>0, then λ * It has only one possible value:

[0157]

[0158] It is easy to obtain that in (0,λ) * ), E′[T]<0, E[T] decreases, in (λ * Since E'[T] > 0, E[T] is increasing; therefore, λ = λ * When E[T] is at its minimum value;

[0159] When the vehicle detects that the data extraction rate is not equal to λ * At that time, adjust your data extraction rate to λ. * .

[0160] The specific algorithm pseudocode is shown in Algorithm 1:

[0161]

[0162]

[0163] like Figure 4 As shown, the impact of the maximum number of connected vehicles on system latency demonstrates the relationship between system latency and the rate λ. The optimal point of λ varies with different values ​​of m: all three curves show that the system latency can have an optimal point. When m=2, the curve of λ first drops rapidly and then rises rapidly, with the optimal point at 0.22. When m=3, the curve of λ similarly drops rapidly and then rises rapidly, with the optimal point at 0.18. When m=4, the curve of λ similarly drops rapidly and then rises rapidly, with the optimal point at 0.16. This is due to the properties of the formula, and the minimum value for m=4 is smaller than the minimum value for m=2 or m=3, indicating that there is also a suitable value for the number of vehicles connected to the same server.

[0164] like Figure 5As shown, this study investigates the impact of attack intensity on system latency under different data extraction rates. The parameter m is set to 3, and the attack intensity is represented by simulating server paralysis, where intensity 1 represents paralyzing one server and intensity 2 represents paralyzing two servers. The figure clearly shows that system latency increases with increasing attack intensity. Before the attack intensity is below 5, the increase in system latency is not significant, indicating that the system can cope relatively effectively with low-intensity attacks. However, after the attack intensity exceeds 5, the system latency begins to rise sharply, exhibiting a clear exponential function. This phenomenon indicates that once the system suffers a higher-intensity attack, its latency growth will accelerate rapidly, potentially leading to a sharp decline in system performance. By comparing the curves under different parameters, it can be found that the system performs significantly better at a data extraction rate of 0.2 than at data extraction rates of 0.3 or 0.4. This shows that a lower data extraction rate can better maintain system latency performance when facing attacks, making the system more robust and stable. Therefore, when designing a system, it is necessary to carefully select the data extraction rate to ensure that the system can maintain good performance when facing various attacks.

[0165] like Figure 6 As shown in the figure, the impact of the maximum number of connected vehicles and the number of vehicles on system latency under different data extraction rates presents the relationship between system latency and parameters m and n; from Figure 6 As can be observed, the system latency gradually increases with the increase of parameter m. This is because the increase of m leads to more computation and processing requirements, thus increasing the system response time. Similarly, when the parameter n increases, the system latency also increases, because more n means more data transmission and processing operations, which in turn affects the system's latency performance. However, as m increases, the rate of increase in system latency gradually accelerates. This is because as m increases, the system processing complexity increases exponentially, leading to a faster increase in latency. Conversely, as n increases, the rate of increase in system latency gradually slows down. This is because the increase of n brings linearly increasing data transmission and processing requirements, which have a relatively small impact compared to the exponentially increasing m.

[0166] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and a latency optimization method, characterized in that, Includes the following steps: S1. Build a blockchain-based system among edge devices, deploying N servers. The states of the N servers include leaders, followers, and candidates, with the initial state being candidates. S2, according to the process of reaching the Poisson point Calculate the interval delay of vehicle data arrival. ; S3. Construct a channel state model; S4, computing system latency; S401, Data extraction delay , The expectation is Calculate the latency when the vehicle sends data. The expression is: ； in, This indicates the latency of the vehicle's initial local data processing. S402. Calculate the data transmission latency of M vehicles connected to an edge device due to data collisions and update the data transmission latency of each vehicle. And calculate the latency of data sent by the vehicle arriving at the edge device. ; The vehicle data transmission delay The expression is: ； in, Indicates the probability of collision; Latency of vehicle data reaching edge devices The expression is: ； in, This indicates the probability that a channel state update is discarded. S403. Considering the latency caused by edge devices demodulating vehicle data and uploading decoded data, calculate the latency. ; The delay The expression is: ； in, This indicates the latency incurred by the edge device in decoding vehicle data. This represents the delay caused by the follower uploading the decoded variable to the leader. This refers to the time delay that occurs when a leader communicates a message to all followers. This indicates the time delay in the election of leaders; S404, Calculate the additional time overhead of leader election caused by an attack on the server. The expression is: ； in, Indicates the time for electing a leader. Indicates a preset time period. Indicates delay Expectations Indicates the number of servers that were attacked; The overall system delay The expression is: ； ； in, This indicates the transmission interval, set to avoid data transmission conflicts. Indicates the delay in data processing and transmission. S405, Obtain the overall system delay ; S5. Overall system delay Optimization is performed by finding the extreme point through convex optimization differentiation to obtain the most suitable result. When the vehicle detects that the data extraction rate is not equal to At that time, adjust your data extraction rate to This results in overall system latency Minimum.

2. The blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and the latency optimization method according to claim 1, characterized in that, The N servers elect a leader through random voting. The leader generates an origin block to form a chain, and a timer is set to periodically... Internal broadcast control information.

3. The blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and the latency optimization method according to claim 1, characterized in that, The process of extracting vehicle data, A, includes points. For non-negative real numbers, according to the point Arrival order of arrival time points, i.e. The arrival delay The expression is: 。 4. The blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and the latency optimization method according to claim 1, characterized in that, The construction of the channel state model includes: calculating the channel state preservation probability based on vehicle speed and channel parameters; Channel states include ideal states and error states. The channel state preservation probability expression is: ； in, This represents the Markum Q function. This represents the probability that an ideal channel will remain in an ideal state. This represents the probability that the error channel remains in an error state. Represents the dielectric constant. The expression is: ； in, Indicates fading margin, The average probability of frame transmission failure is expressed as: , Indicates the Doppler frequency as The Gaussian correlation coefficient of the two samples of the fading channel amplitude. The expression is: ； in, Represents the zeroth-order Bessel function. Indicates the frame transmission time on the channel. Indicates the Doppler frequency. The expression is: ； in, Indicates the carrier frequency. Indicates vehicle speed. It represents the speed of light.

5. The blockchain technology-based framework for protecting vehicle-to-everything (V2X) communication and the latency optimization method according to claim 4, characterized in that, The construction of the channel state model also includes: calculating the probability of discarding a channel state update. The expression is: ； in, This represents the probability that at least one frame transmission fails in an error channel. This represents the probability of failure of at least one frame in the ideal new track.

Images