CN111065089B - Internet of vehicles bidirectional authentication method and device based on crowd sensing - Google Patents
Internet of vehicles bidirectional authentication method and device based on crowd sensing Download PDFInfo
- Publication number
- CN111065089B CN111065089B CN201911071679.2A CN201911071679A CN111065089B CN 111065089 B CN111065089 B CN 111065089B CN 201911071679 A CN201911071679 A CN 201911071679A CN 111065089 B CN111065089 B CN 111065089B
- Authority
- CN
- China
- Prior art keywords
- data
- base station
- vehicle
- authentication
- assisting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/60—Context-dependent security
- H04W12/63—Location-dependent; Proximity-dependent
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/009—Security arrangements; Authentication; Protecting privacy or anonymity specially adapted for networks, e.g. wireless sensor networks, ad-hoc networks, RFID networks or cloud networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/06—Authentication
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W36/00—Hand-off or reselection arrangements
- H04W36/0005—Control or signalling for completing the hand-off
- H04W36/0083—Determination of parameters used for hand-off, e.g. generation or modification of neighbour cell lists
- H04W36/00837—Determination of triggering parameters for hand-off
- H04W36/008375—Determination of triggering parameters for hand-off based on historical data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W36/00—Hand-off or reselection arrangements
- H04W36/08—Reselecting an access point
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W36/00—Hand-off or reselection arrangements
- H04W36/24—Reselection being triggered by specific parameters
- H04W36/32—Reselection being triggered by specific parameters by location or mobility data, e.g. speed data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L2209/00—Additional information or applications relating to cryptographic mechanisms or cryptographic arrangements for secret or secure communication H04L9/00
- H04L2209/80—Wireless
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L2209/00—Additional information or applications relating to cryptographic mechanisms or cryptographic arrangements for secret or secure communication H04L9/00
- H04L2209/84—Vehicles
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The embodiment of the invention discloses a bidirectional authentication method and device for Internet of vehicles based on crowd sensing, relating to the technical field of security authentication of Internet of vehicles, wherein the method comprises the following steps: the authentication base station selects a scoring strategy and broadcasts the scoring strategy to the assisting base station; the assisting base station sends the vehicle historical state information to the authentication base station according to the scoring strategy; acquiring a first behavior prediction result by utilizing a decision tree based on the historical state information of the vehicle; constructing a criterion of credibility of the data of the assisting base station, solving a plurality of evidence conflicts according to a basic probability assignment function, and determining the quality grade of the data of the assisting base station; executing a scoring strategy and updating a Q matrix to obtain an accurate vehicle behavior prediction result; and then the vehicle and the assisting base station are authenticated. The embodiment of the invention authenticates the authenticity of the base station side through the data quality grade on one hand, and authenticates the authenticity of the vehicle information through the vehicle predicted behavior result on the other hand, thereby realizing the bidirectional authentication of the base station and the vehicle and solving the network security problem of the Internet of vehicles.
Description
Technical Field
The embodiment of the invention relates to the technical field of vehicle networking security authentication, in particular to a vehicle networking mutual authentication method and device based on crowd sensing.
Background
Under the impetus of manufacturers in the automobile industry, the technology of internet of vehicles is receiving wide attention. As an important branch of the Internet of things, the Internet of vehicles is mainly applied to information interaction among vehicles, pedestrians, road side units and base stations, and the purpose of the Internet of vehicles is to enable people to easily acquire real-time road traffic information so as to guarantee driving safety and improve user experience.
However, car networking technology faces three major challenges: safety, standard and cost. The most important problem is the security problem, if the security problem of the internet of vehicles is not guaranteed, the life security of the user will be directly threatened, for example, the attacked base station broadcasts false road condition information to cause traffic jam or the driving information of the vehicle is exposed to cause lawless persons to track, etc. Therefore, the first line of defense to solve these problems is to authenticate the identity of the vehicle.
In the conventional car networking authentication handover technology, an encryption authentication method is generally adopted to solve the security problem, but although the encryption authentication method can solve the basic requirements of security authentication, the authentication mechanism based on the encryption technology as a static authentication mechanism gradually exposes some defects, such as: the attack problem of identity embezzlement cannot be solved, and only after analysis and off-line detection can be generally carried out once the situation occurs; the calculation cost and the transmission cost are high, and the requirements of low time delay and timely switching authentication of the Internet of vehicles are difficult to meet; in addition, most of the traditional authentication mechanisms belong to active security defense mechanisms, and influence user experience; most work is to design an algorithm to authenticate a vehicle at a user end, and assume that the safety and reliability of a base station end are safe and reliable, or the top-level controller performs periodic detection after offline for a period of time, which cannot guarantee the authenticity of the base station side and the network safety after the base station side is attacked.
Therefore, it is necessary to provide a new authentication method for the above technical defects.
Disclosure of Invention
The embodiment of the invention aims to provide a vehicle networking bidirectional authentication method and device based on crowd sensing, which are used for solving the problems of authenticity of a base station side and network security caused by attack of the authenticity.
In order to achieve the above object, the embodiments of the present invention mainly provide the following technical solutions:
in a first aspect, a mutual authentication method for Internet of vehicles based on crowd sensing is provided,
the method comprises the following steps: the authentication base station selects a scoring strategy according to the prompt of the Q matrix and broadcasts the scoring strategy to the assisting base station; the assisting base station sends vehicle historical state information to the authentication base station according to the scoring strategy; predicting the vehicle behavior by utilizing a decision tree based on the historical state information of the vehicle to obtain a first behavior prediction result; constructing a criterion of the credibility of the data of the assisting base station, solving a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, and determining the quality grade of the data of the assisting base station; executing a scoring strategy and updating a Q matrix, iterating the steps until stable data is obtained, and optimizing a predicted vehicle behavior result to obtain an accurate vehicle behavior prediction result; and authenticating the vehicle by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle reaches the authentication base station, and authenticating the assistance base stations according to the data quality grades of the assistance base stations when the iteration is finished.
Further, the selection of the scoring strategy specifically includes: presetting a data quality grade; making different scoring strategies according to the quality grade, the data attribute and the data utility of the data, and taking the scoring strategies as an action set of Q-learning; the authentication base station selects a corresponding scoring strategy according to a greedy algorithm and sends the scoring strategy to an assisting base station indicated by the SDN controller; and selecting Q-learning states according to the quality grade condition of the data collected at the previous moment and the executed scoring strategy, and selecting Q-learning rewards according to the grade difference of the data quality at the adjacent moment.
Further, the dividing of the data quality grade includes determining a data volume, and the determining of the data volume specifically includes: the authentication base station divides the received data set into beta +1 data subsets, and the data quantity of the data subsets is sequentially decreased at equal intervals; respectively judging the data attributes of the beta +1 data subsets, and judging the data set with the attributes of the beta +1 data subsets without conflict as data volume redundancy; judging that the data size is enough if the attributes of the first beta data subsets have no conflict but the beta +1 data subsets have conflict; judging that the data quantity is not enough when the attributes of the first beta data subsets have conflicts or the number of the data subsets is not enough; and dividing the data quality grades according to the data quantity and the arrangement and combination condition of the data attributes.
Further, the method for determining the data attribute specifically includes: taking the score of the assisting base station as first evidence; taking the abnormal data condition of the assisting base station as a second evidence; taking the comprehensive condition of the plurality of predicted vehicle behavior results as a third evidence; respectively carrying out basic probability assignment function quantization on the first evidence, the second evidence and the third evidence, and fusing to obtain a group of fused basic probability assignment functions; comparing the function value of the fusion basic probability assignment function with a preset threshold value, and judging the data attribute of the assisting base station; wherein, the data attribute comprises distustful and trustful.
Further, the scoring policy specifically includes: judging according to the data quality grade, wherein the perceived data quality grade is improved, and the score is higher; judging according to the attribute of the data, wherein the attribute of the data is an addend of trustful, and the attribute of the data is a derviation of distustful; and judging according to the utility of the data, and scoring the data with similar utility in the authentication work of the base station to the same grade.
Further, the method for selecting the authentication base station and the assisting base station specifically includes: after finishing the work of vehicle authentication, any base station executing authentication reports the authentication condition to an SDN controller, the SDN controller selects the base station which is currently reported and authenticated to be the center according to the actual road condition, and selects a plurality of base stations which are possible to pass by vehicles around the center to be used as authentication base stations for next vehicle authentication preparation; and then, selecting a plurality of base stations of unorganized sparse position points in the track which the vehicle has traveled as assisting base stations, wherein the base stations are used for sending the historical traveling data of the vehicle to the authentication base stations and assisting the authentication base stations to finish the authentication work of the vehicle.
Further, the obtaining of the first behavior prediction result specifically includes: performing data preprocessing, and performing data pair matching on the vehicle historical state information sent by the assisting base station and the data of the authentication base station; discretizing a continuous value of the data attribute of the assisting base station as the input of a decision tree, clustering the data of the authenticating base station as the prediction result of the decision tree, and establishing the decision tree; calculating the information gain of each assisting base station data attribute, selecting the attribute with the information gain higher than the average value, calculating the information gain rate of the assisting base station data attribute, and selecting the attribute with the maximum information gain rate as a root node of a decision tree; and pruning the built decision tree, taking the latest data left by the vehicle at the assisting base station as the input of the decision tree, predicting the behavior of the vehicle which is about to arrive at the authentication base station, and recording the predicted behavior as a first behavior prediction result.
Further, the method comprises: after each assisting base station sends data, a scoring strategy is executed to score the assisting base station, a Q matrix is updated, whether iteration is converged is judged, if not, a new scoring strategy is selected, iteration is continued until the assisting base station sends data with the highest data quality grade, the assisting base station sending the highest quality data is judged as a real base station, authenticity of the assisting base station is authenticated, and an accurate vehicle behavior prediction result is obtained according to the data of the real base station and is used for detecting whether vehicle behavior is abnormal.
Further, the method further comprises: when the vehicle to be tested enters the coverage area of the authentication base station, the authentication base station compares the real behavior state of the vehicle with the prediction result, if the real behavior state of the vehicle is the same as the prediction result, the vehicle to be tested is allowed to access the network, if the real behavior state of the vehicle is different from the prediction result, the vehicle to be tested is judged to be an abnormal vehicle, and the vehicle to be tested is not allowed to access the network;
in a second aspect, a mutual vehicle networking authentication device based on crowd sensing is provided, the device comprising: the learning unit is used for dividing the data grade by utilizing a Q-learning algorithm, making a dividing strategy and exciting the assisting base station to send high-quality data so as to obtain an accurate vehicle behavior prediction result; the data transmission unit is used for realizing data transmission between the authentication base station and the assistance base station; the vehicle behavior prediction unit is used for predicting the vehicle behavior by utilizing the decision tree according to the historical state information of the vehicle and acquiring a first behavior prediction result; the evidence conflict processing unit is used for constructing a criterion of credibility of the assisting base station data, solving a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, judging data attributes, and determining the quality grade of the assisting base station data so as to authenticate the authenticity of the assisting base station; and the authentication unit is used for authenticating the vehicle by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle reaches the authentication base station.
The technical scheme provided by the embodiment of the invention at least has the following advantages:
according to the embodiment of the invention, historical state information of a plurality of assisting base stations is collected through a crowd sensing method, the behavior of a vehicle is predicted through a decision tree, the data credibility criterion of the assisting base stations is established, a plurality of evidence conflicts are solved according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, the data quality grade is determined, the predicted vehicle behavior result is optimized through iteration, the accurate prediction of the future driving state information of the vehicle is realized, on one hand, the vehicle is authenticated according to the predicted result, on the other hand, the base stations are authenticated according to the data quality grade, and the bidirectional authentication of the base stations and the vehicle is realized; in order to reduce misjudgment and excite the road side unit to actively participate in the bidirectional authentication task, the Q-learning algorithm in reinforcement learning is utilized for iteration, and a scoring strategy is used for continuously exciting the assisting base station to send data with higher quality level; the method can solve the problem of identity embezzlement attack of the base station and has the advantages of safety, rapidness, high efficiency and the like.
Drawings
Fig. 1 is a step diagram of a car networking bidirectional authentication method based on crowd sensing according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a method for selecting an authentication base station and an assisting base station according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of data matching of a decision tree according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a car networking bidirectional authentication device based on crowd sensing according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure.
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, interfaces, techniques, etc. in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
Before describing the embodiments of the present invention, a brief description will be made of the technical background related to the embodiments:
Crowd-Sensing, is a new data acquisition mode that combines crowdsourcing thinking and mobile device perception capabilities, and is an expression form of the internet of things. Crowd sensing refers to forming an interactive and participatory sensing network through the existing mobile equipment of people and releasing a sensing task to an individual or a group in the network to complete the sensing, so that professionals or the public are helped to collect data, analyze information and share knowledge. The concept of crowd sensing is to realize unconscious cooperation, so that a user can complete a sensing task under the unconscious condition, and the barrier of participation of professionals is broken through. The crowd sensing has the advantages of flexible and economical deployment, multi-source heterogeneous sensing data, wide and uniform coverage range, high-expansion multifunction and the like. The crowd-sourcing aware incentive mechanism achieves the maximum utility of the server platform and the participants through some incentive mode.
The goal of Q-learning reinforcement learning is to obtain a best solution strategy for the current problem through reward and punishment, reward the good strategy, punish the bad strategy, continuously reinforce the process, and finally obtain a best strategy.
The embodiment of the invention mainly makes an excitation scoring strategy based on reinforcement learning, gives each base station an initial trust value, and collects high-quality perception data in a one-step iteration process by utilizing a Q-learning algorithm to finally realize accurate authentication of the vehicle and the base station side.
Specifically, the embodiment of the invention provides a mutual authentication method for an internet of vehicles based on crowd sensing, which is applied to an identity authentication system between a vehicle and each network access point in an internet of vehicles environment, and is used for rapidly and safely authenticating the authenticity of the vehicle, each roadside unit and a base station side and simultaneously realizing the low-delay requirement of the internet of vehicles authentication system. Referring to fig. 1, the method includes:
s1, the authentication base station selects a scoring strategy according to the prompt of the Q matrix and broadcasts the scoring strategy to the assisting base station;
the present embodiment utilizes a network architecture combining currently mainstream mobile edge computing and software defined networking, and includes a bottom device, a core network middle layer composed of switches, and a top SDN controller, where the bottom device communicates with a vehicle through a wireless communication network, such as an IEEE 802.11p communication protocol. Enabling the SDN controller to master the driving track of the vehicle and the deployment condition of the base station, and guiding the vehicle authentication task by using the centralized control capability of the SDN controller.
Specifically, in the embodiment of the present invention, different base stations are required to be selected to perform respective tasks, including an authentication base station and an assisting base station, where the authentication base station is a base station that performs information authentication, and the assisting base station is a base station that assists the authentication base station to complete authentication, and a specific method for selecting the authentication base station and the assisting base station is as follows: in the embodiment, a base station reporting mechanism is adopted, that is, after any base station executing authentication finishes the work of vehicle authentication, the authentication condition is reported to an SDN controller, the SDN controller can master the track that the vehicle has run, the SDN controller selects the base station which finishes the current reported authentication as a center according to the actual road condition, and selects a plurality of base stations which are possible paths of the vehicle around the center as authentication base stations for the next vehicle authentication preparation;
in consideration of privacy safety of a user, the assisting base station is selected in a mode of randomly selecting sparse position points, namely, a plurality of base stations with unorganized sparse position points need to be selected from tracks where vehicles pass through as the assisting base stations, the assisting base stations are used for sending historical driving data of the vehicles to the authentication base stations, and the assisting base stations are used for finishing authentication work on the vehicles.
As shown in fig. 2, when a vehicle passes through base stations BS1, BS4, BS7, and BS10 in sequence, and BS10 is a base station that reports vehicle authentication conditions to an SDN controller at present, the SDN selects BS7, BS8, BS12, and BS13 as authentication base stations to perform next vehicle authentication preparation according to actual road conditions, and an assisting base station selects a few sparse location points such as BS1 and BS 10.
Mainly, the selection of the scoring strategy specifically comprises the following steps:
the method for presetting the data quality grade specifically comprises the following steps: since the prediction result is biased due to too small data amount, and the prediction result tends to be stable as the data amount increases, the authentication base station wants to assist the base station to send data which is neither small nor redundant in consideration of calculation cost and utility. Therefore, the authentication base station divides the received data set into β +1 data subsets, and the data amount thereof decreases sequentially at equal intervals, preferably 10 data subsets are used as intervals in this embodiment; respectively judging the attributes of the beta +1 data subsets, and judging the data set with the attributes of the beta +1 data subsets without conflict as data volume redundancy; judging that the data size is enough if the attributes of the first beta data subsets have no conflict but the beta +1 data subsets have conflict; judging that the data quantity is not enough when the attributes of the first beta data subsets have conflicts or the number of the data subsets is not enough; dividing the data quality grades according to the number of data and the arrangement and combination condition of the data attributes; the data attributes include trustful and distrustful, namely trusted and untrusted. The present embodiment preferably divides the data levels into 6 levels, which are:
l1: the data quantity is less than 10;
l2: data volume redundancy;
l3: the data volume is insufficient and the attribute is disttrustful;
l4: the data volume is insufficient and the attribute is trustful;
l5: the data volume is enough and the attribute is disttrustful;
l6: the amount of data is sufficient and the attribute is trustful.
Further, different scoring strategies are formulated according to the quality grade of the data, the data attribute and the data utility, and the scoring strategies are used as an action set of Q-learning, and specifically comprise 3 scoring strategies:
strategy 1: judging according to the data quality grade, wherein the perceived data quality grade is improved, and the score is higher;
strategy 2: judging according to the attribute of the data, wherein the attribute of the data is an addend of trustful, and the attribute of the data is a derviation of distustful;
strategy 3: and judging according to the utility of the data, and scoring the data with similar utility in the authentication work of the base station to the same grade. If the attribute of insufficient data amount is trustful or untrustful, the same score of-0.5 is given, and if the attribute of insufficient data amount is trustful or the data amount is redundant, the same score of-1 is given. In practice, the specific scoring rules are shown in table 1,
table 1:
selecting Q-learning state according to quality grade condition of data collected at last moment and executed scoring strategy, counting data proportion of each grade, and recording as
Wherein l represents the number of levels,
represents the proportion of reports with the grade of l collected by the kth iteration authentication base station, q represents the number of assisting base stations participating in the authentication task, j represents the number of the base stations,
indicating the level of data transmitted by base station j for the kth iteration,
is an identification function when
The identification function takes 1. Thus, the Q-learning state is defined as s(k)＝[Nl (k-1),a(k-1)]。
Selecting Q-learning reward according to grade difference of data quality at adjacent time, wherein the formula is as follows
Wherein the content of the first and second substances,
the quantization value for the perceived data quality at time k,
the quantization value is the perceived data quality at time k-1. And the prompting of the Q matrix is to set a value according to a greedy algorithm, if the value is 0.9, the authentication base station selects an action, namely a scoring strategy, corresponding to the maximum Q value of the row in which the current state is located in the Q matrix with a probability of 90%, randomly selects an action with a probability of 10%, and then sends the corresponding scoring strategy to the assisting base station indicated by the SDN controller according to the prompting.
S2, the assisting base station sends the vehicle history state information to the authentication base station according to the scoring strategy;
data forwarding and state acquisition are provided among all base stations through wired links, and the historical state information of the vehicle comprises information such as time, longitude, latitude, speed, direction included angle and the like when the vehicle applies for network access;
s3, predicting the vehicle behavior by using a decision tree based on the vehicle historical state information to obtain a first behavior prediction result;
the method comprises the following steps: performing data preprocessing, and performing data pair matching on the vehicle historical state information sent by the assisting base station and the data of the authentication base station; specifically, the collected perception data and the historical state information stored in the authentication base station are matched into a data pair one by one, and the data pair has the following characteristics: the time stamps of the two are nearest, the time stamp of the data sent by the assisting base station is earlier than the time stamp of the data of the authenticating base station, and when the assisting base station has a plurality of pieces of data corresponding to the data of one authenticating base station, the data pairs with far time intervals are deleted.
After the data pairs are formed, discretizing continuous values of data attributes of the assisting base station as input of a decision tree, wherein the data attributes comprise time, longitude and latitude, speed, direction included angle and the like, clustering the data of the authentication data as a prediction result of the decision tree, and establishing the decision tree; the information gain of each assisting base station data attribute is calculated, the attribute with the information gain higher than the average value is selected, the information gain rate of the assisting base station data attribute is calculated, and the attribute with the maximum information gain rate is selected as the root node of the decision tree. The calculation formula of the information gain is as follows:
b is an attribute, V is a branch node corresponding to each attribute, D is a whole sample set, the sample set is clustered into K classes during data processing, and the probability of each class appearing in the sample set is pkThen, end (d) is the information entropy of the parent node attribute, and the information entropy is calculated according to the following formula:
the calculation formula of the information gain ratio is as follows:
and further pruning the built decision tree to form a decision tree which is neither over-fitted nor under-fitted, and taking the latest data left by the vehicle at the assisting base station as the input of the decision tree to predict the behavior of the vehicle which is about to arrive at the authentication base station, and recording the predicted behavior as a first behavior prediction result.
Referring to fig. 3, a data pair matching graph of an authentication base station and an assisting base station is shown, and status information of a vehicle about to arrive at the authentication base station is predicted according to status information of the vehicle currently arriving at the assisting base station, where the data in fig. 3 includes 2 parts, and a part of historical data of vehicles in the first three rows is used as data for decision tree building, and the historical data includes historical data of the assisting base station and historical data of the authentication base station. And the other part is the current data left by the last row of vehicles at the assisting base station and used as the input of the decision tree, and the output is the predicted behavior state of the vehicle to be solved when the vehicle reaches the authentication base station.
S4, constructing a criterion of credibility of the assisting base station data, solving a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, and determining the quality grade of the assisting base station data;
and (4) according to the state information predicted by the decision tree in the step (S3) and the characteristic structure of the data sent by each assisting base station, determining whether the data sent by the assisting base station is credible evidence, that is, determining the attributes of the data uploaded by the assisting base station. According to the evidences, different basic probability assignment functions m { trustful }, m { distrustful }, m { trustful } which represents the credibility of data sent by the assisting base station, and m { distrustful } which represents the untrustworthiness are given, and the concrete evidences and the corresponding basic probability assignment function quantization method are as follows:
taking the score of the assisting base station as first evidence; when the assisting base stations are scored, all the assisting base stations are endowed with the same initial score, and then in the process of Q learning each iteration, the authentication base station performs adding or subtracting on the scores of the assisting base stations on the basis of the initial scores according to the quality grades of the data sent by the assisting base stations. Assuming that the initial score of the assisting base station is z, the residual score after k iterations is z′Then the basic probability assignment function is quantized as follows:
when z is equal to z′When, this evidence is not referred to;
when z is<z′When m is1{trustful}＝1，m1{distrustful}＝0；
When z is>z′When the temperature of the water is higher than the set temperature,
m1{trustful}＝1-m1{distrustful}。
taking the abnormal data condition of the assisting base station as a second evidence; if the data with the same attribute in the data of the assisting base station correspond to a plurality of different behavior results, the data are abnormal, if the behavior results of the authentication base station are clustered into K types, the data with the same attribute in the data of the assisting base station correspond to K′And quantifying the basic probability assignment function according to the behavior results of different authentication base stations as follows:
it can be understood that: because the false base station has no real historical state information of the vehicle, the false base station can only forge the vehicle on the basis of the state information left by the vehicle at present, so that the historical data with the same attribute can correspond to a plurality of different behavior results when the vehicle arrives at the authentication base station, the attribute refers to the attribute selected when the decision tree predicts, and the more different numbers of the historical data, the more suspicious the data provided by the assisting base station.
Taking the comprehensive condition of the plurality of predicted vehicle behavior results as a third evidence; if q assisting base stations participating in the authentication task exist, that is, the authentication base station can predict q behavior results, if the behavior results corresponding to the currently determined data are the same, q behavior results exist′Then, the basic probability assignment function is quantized as follows:
it should be noted that, since a plurality of evidences are cited in the above steps, an evidence conflict may be caused, and therefore, the first evidence, the second evidence, and the third evidence are quantized by the basic probability assignment function, and are synthesized into a set of basic probability assignment functions by using a fusion formula of the D-S evidence theory, and are calculated according to the following formula:
wherein A is1，···，Am∈Θ，A1∩···∩Am＝A，к＝1/∑m1(A1)···mm(Am)。
Comparing the calculated fusion basic probability assignment function value with a preset threshold value, and determining whether the data of the assisting base station is trusted, where the preset threshold value of this embodiment is 0.5, that is, when the function value of the basic probability assignment function is higher than 0.5, the data is trusted, that is, the data attribute is trustful, and is not trusted when the function value is lower than 0.5, that is, the data attribute is distrustful, and the obtained data attribute is used as one of the bases for determining the data grade thereof, and is used for determining the data grade thereof according to the grade division policy in step S1.
S5, executing a scoring strategy and updating a Q matrix, and iterating the steps until stable data are obtained to obtain an accurate vehicle behavior prediction result;
the Q matrix is updated according to the following equation:
Q(s,a)←Q(s,a)+α[R(s,a)+γmaxQ(s′,a′)-Q(s,a)]
where α is the learning rate, γ is the discounting factor, maxQ(s)′,a′) The maximum Q value at the next moment.
In the iteration process, the vehicle behavior result output by the decision tree is used as a basis for judging the data quality grade, and when the data quality sent by each assisting base station is stable after iteration, namely the quantity of the received data with the data quality grade L6 is not changed any more, the base station which cannot send the data with the data quality grade L6 can be judged to be a false base station, so that the assisting base station is authenticated.
And then when the iteration is terminated, judging that the predicted behavior result output by the real data sent by the base station in the decision tree is the accurate vehicle behavior result. In addition, when the data quality is stable, the data is not sent to the authentication base station any more, and the effect of low time delay can be achieved.
And S6, authenticating the vehicle by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle reaches the authentication base station, and authenticating the assistance base stations according to the data quality grades of the assistance base stations when the iteration is finished.
When the vehicle to be tested enters the coverage area of the authentication base station, the authentication base station compares the real behavior state of the vehicle with the prediction result, if the real behavior state of the vehicle is the same as the prediction result, the vehicle to be tested is allowed to access the network, if the real behavior state of the vehicle is different from the prediction result, the vehicle to be tested is judged to be an abnormal vehicle, and the vehicle to be tested is not allowed to access the network;
when the iteration is terminated, the quality of the data sent by each assisting base station tends to be stable, namely when the quantity of the received data with the data quality level of L6 is not changed, the base station which cannot send the data with the data quality level of L6 can be judged to be a false base station, and therefore the assisting base station is authenticated.
According to the embodiment of the invention, historical state information of a plurality of assisting base stations is collected through a crowd sensing method, the behavior of a vehicle is predicted through a decision tree, a data credibility criterion of the assisting base stations is established, a plurality of evidence conflicts are solved according to a basic probability assignment function, the predicted vehicle behavior result is optimized, the accurate prediction of the vehicle future driving state information is realized, on one hand, the vehicle is authenticated according to the prediction result, and on the other hand, the base stations are authenticated according to the data quality grade sent by the assisting base stations; in order to reduce misjudgment and excite the road side unit to actively participate in the bidirectional authentication task, the Q-learning algorithm in reinforcement learning is utilized for iteration, and a scoring strategy is used for continuously exciting the assisting base station to send data with higher quality level; the method can solve the problem of identity embezzlement attack of the base station and has the advantages of safety, rapidness, high efficiency and the like.
Corresponding to the above embodiments, an embodiment of the present invention provides a bidirectional authentication device for internet of vehicles based on crowd sensing, and referring to fig. 4, the device includes:
the learning unit 01 is used for dividing data grades by using a Q-learning algorithm, making a dividing strategy and exciting the assisting base station to send high-quality data so as to obtain an accurate vehicle behavior prediction result; the method is specifically used for performing the scoring strategy to score the assisting base station after the assisting base station sends data each time, updating the Q matrix, judging whether iteration converges, if not, selecting a new scoring strategy, and continuing the iteration until the assisting base station sends data with the highest data quality grade.
A data transmission unit 02 for implementing data transmission between the authentication base station and the assisting base station;
the vehicle behavior prediction unit 03 is configured to predict a vehicle behavior by using a decision tree according to vehicle historical state information, and obtain a first behavior prediction result; the system comprises a base station, an assistant base station, an authentication base station and a communication terminal, wherein the base station is used for sending vehicle historical state information to the assistant base station; discretizing a continuous value of the data attribute of the assisting base station as the input of a decision tree, clustering the data of the authenticating base station as the prediction result of the decision tree, and establishing the decision tree; calculating the information gain of each assisting base station data attribute, selecting the attribute with the information gain higher than the average value, calculating the information gain rate of the assisting base station data attribute, and selecting the attribute with the maximum information gain rate as a root node of a decision tree; and pruning the built decision tree, taking the latest data left by the vehicle at the assisting base station as the input of the decision tree, predicting the behavior of the vehicle which is about to arrive at the authentication base station, and recording the predicted behavior as a first behavior prediction result.
The evidence conflict processing unit 04 is configured to construct an assisting base station data credibility criterion, solve a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, determine data attributes, and determine quality levels of assisting base station data to authenticate authenticity of the assisting base station; in particular for using the score of the assisting base station as a first evidence; taking the abnormal data condition of the assisting base station as a second evidence; taking the comprehensive condition of the plurality of predicted vehicle behavior results as a third evidence; respectively carrying out basic probability assignment function quantization on the first evidence, the second evidence and the third evidence, and fusing to obtain a group of fused basic probability assignment functions; and comparing the function value of the fusion basic probability assignment function with a preset threshold value, and judging whether the data of the assisting base station is credible or not.
And the authentication unit 05 is used for authenticating the vehicle and the assisting base station by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle arrives at the authentication base station.
According to the embodiment of the invention, historical state information of a plurality of assisting base stations is collected through a crowd sensing method, the behavior of a vehicle is predicted through a decision tree, a data credibility criterion of the assisting base stations is established, a plurality of evidence conflicts are solved according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, the data quality grade is determined, the predicted vehicle behavior result is optimized through iteration, the accurate prediction of the future driving state information of the vehicle is realized, on one hand, the vehicle is authenticated according to the predicted result, and on the other hand, the base stations are authenticated according to the data quality grade sent by the assisting base stations; in order to reduce misjudgment and excite the road side unit to actively participate in the bidirectional authentication task, the Q-learning algorithm in reinforcement learning is utilized for iteration, and a scoring strategy is used for continuously exciting the assisting base station to send data with higher quality level; the method can solve the problem of identity embezzlement attack of the base station and has the advantages of safety, rapidness, high efficiency and the like.
In the implementation process of the embodiments disclosed in the present invention, a computer-readable storage medium is involved, in which computer program instructions are stored, and when the computer program instructions are run on a computer, the computer is caused to execute the method provided by the embodiments of the present invention.
The storage medium may be a memory, for example, which may be volatile memory or nonvolatile memory, or which may include both volatile and nonvolatile memory.
The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash Memory.
The volatile Memory may be a Random Access Memory (RAM) which serves as an external cache. By way of example and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), SLDRAM (SLDRAM), and Direct Rambus RAM (DRRAM).
The storage media described in connection with the embodiments of the invention are intended to comprise, without being limited to, these and any other suitable types of memory.
Those skilled in the art will appreciate that the functionality described in the present invention may be implemented in a combination of hardware and software in one or more of the examples described above. When software is applied, the corresponding functionality may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.
Claims (10)
1. A mutual authentication method for Internet of vehicles based on crowd sensing is characterized by comprising the following steps:
the authentication base station selects a scoring strategy according to the prompt of the Q matrix and broadcasts the scoring strategy to the assisting base station;
the assisting base station sends vehicle historical state information to the authentication base station according to the scoring strategy;
predicting the vehicle behavior by utilizing a decision tree based on the historical state information of the vehicle to obtain a first behavior prediction result;
constructing a criterion of the credibility of the data of the assisting base station, solving a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, and determining the quality grade of the data of the assisting base station;
executing a scoring strategy and updating a Q matrix, performing iterative operation until stable data are obtained, and optimizing a first behavior prediction result to obtain an accurate vehicle behavior prediction result;
and authenticating the vehicle by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle reaches the authentication base station, and authenticating the assistance base stations according to the data quality grades of the assistance base stations when the iteration is finished.
2. The Internet of vehicles mutual authentication method based on crowd sensing as claimed in claim 1, wherein the selection of the scoring policy specifically comprises:
presetting a data quality grade;
making different scoring strategies according to the quality grade, the data attribute and the data utility of the data, and taking the scoring strategies as an action set of Q-learning;
the authentication base station selects a corresponding scoring strategy according to a greedy algorithm and sends the scoring strategy to an assisting base station indicated by the SDN controller;
and selecting Q-learning states according to the quality grade condition of the data collected at the previous moment and the executed scoring strategy, and selecting Q-learning rewards according to the grade difference of the data quality at the adjacent moment.
3. The networking for vehicle mutual authentication method based on crowd sensing as recited in claim 2, wherein the dividing of the data quality level includes determining a data volume, and the determining of the data volume specifically includes:
the authentication base station divides the received data set into beta +1 data subsets, and the data quantity of the data subsets is sequentially decreased at equal intervals;
respectively judging the data attributes of the beta +1 data subsets, and judging the data set with the attributes of the beta +1 data subsets without conflict as data volume redundancy;
judging that the data size is enough if the attributes of the first beta data subsets have no conflict but the beta +1 data subsets have conflict;
judging that the data quantity is not enough when the attributes of the first beta data subsets have conflicts or the number of the data subsets is not enough;
and dividing the data quality grades according to the data quantity and the arrangement and combination condition of the data attributes.
4. The Internet of vehicles mutual authentication method based on crowd sensing as claimed in claim 2, wherein the data attribute determination method specifically comprises:
taking the score of the assisting base station as first evidence;
taking the abnormal data condition of the assisting base station as a second evidence;
taking the comprehensive condition of the plurality of predicted vehicle behavior results as a third evidence;
respectively carrying out basic probability assignment function quantization on the first evidence, the second evidence and the third evidence, and fusing to obtain a group of fused basic probability assignment functions;
comparing the function value of the fusion basic probability assignment function with a preset threshold value, and judging the data attribute of the assisting base station; wherein, the data attribute comprises distustful and trustful.
5. The Internet of vehicles mutual authentication method based on crowd sensing as claimed in claim 1 or 2, wherein the scoring policy specifically comprises:
judging according to the data quality grade, wherein the perceived data quality grade is improved, and the score is higher;
judging according to the attribute of the data, wherein the attribute of the data is an addend of trustful, and the attribute of the data is a derviation of distustful;
and judging according to the utility of the data, and scoring the data with similar utility in the authentication work of the base station to the same grade.
6. The Internet of vehicles mutual authentication method based on crowd sensing as claimed in claim 1, wherein the selection method of the authentication base station and the assisting base station specifically comprises:
after finishing the work of vehicle authentication, any base station executing authentication reports the authentication condition to an SDN controller, the SDN controller selects the base station which is currently reported and authenticated to be the center according to the actual road condition, and selects a plurality of base stations which are possible to pass by vehicles around the center to be used as authentication base stations for next vehicle authentication preparation;
and then, selecting a plurality of base stations of unorganized sparse position points in the track which the vehicle has traveled as assisting base stations, wherein the base stations are used for sending the historical traveling data of the vehicle to the authentication base stations and assisting the authentication base stations to finish the authentication work of the vehicle.
7. The internet of vehicles mutual authentication method based on crowd sensing as claimed in claim 1, wherein the obtaining of the first behavior prediction result specifically comprises:
performing data preprocessing, and performing data pair matching on the vehicle historical state information sent by the assisting base station and the data of the authentication base station;
discretizing a continuous value of the data attribute of the assisting base station as the input of a decision tree, clustering the data of the authenticating base station as the prediction result of the decision tree, and establishing the decision tree; calculating the information gain of each assisting base station data attribute, selecting the attribute with the information gain higher than the average value, calculating the information gain rate of the assisting base station data attribute, and selecting the attribute with the maximum information gain rate as a root node of a decision tree;
and pruning the built decision tree, taking the latest data left by the vehicle at the assisting base station as the input of the decision tree, predicting the behavior of the vehicle which is about to arrive at the authentication base station, and recording the predicted behavior as a first behavior prediction result.
8. The networking-over-vehicle mutual authentication method based on crowd sensing as claimed in claim 1, wherein the method comprises:
after each assisting base station sends data, a scoring strategy is executed to score the assisting base station, a Q matrix is updated, whether iteration is converged is judged, if not, a new scoring strategy is selected, iteration is continued until the assisting base station sends data with the highest data quality grade, the assisting base station sending the highest quality data is judged as a real base station, authenticity of the assisting base station is authenticated, and an accurate vehicle behavior prediction result is obtained according to the data of the real base station and is used for detecting whether vehicle behavior is abnormal.
9. The networking for vehicle mutual authentication method based on crowd sensing as claimed in claim 1, wherein said method further comprises:
when the vehicle to be tested enters the coverage area of the authentication base station, the authentication base station compares the real behavior state of the vehicle with an accurate vehicle behavior prediction result, if the real behavior state of the vehicle is the same as the accurate vehicle behavior prediction result, the vehicle to be tested is allowed to access the network, if the real behavior state of the vehicle is different from the accurate vehicle behavior prediction result, the vehicle to be tested is judged to be an abnormal vehicle, and the vehicle to be tested is not allowed.
10. A car networking mutual authentication device based on crowd sensing, characterized in that the device includes:
the learning unit is used for dividing the data grade by utilizing a Q-learning algorithm, making a dividing strategy and exciting the assisting base station to send high-quality data so as to obtain an accurate vehicle behavior prediction result;
the data transmission unit is used for realizing data transmission between the authentication base station and the assistance base station;
the vehicle behavior prediction unit is used for predicting the vehicle behavior by utilizing the decision tree according to the historical state information of the vehicle and acquiring a first behavior prediction result;
the evidence conflict processing unit is used for constructing a criterion of credibility of the assisting base station data, solving a plurality of evidence conflicts according to a basic probability assignment function obtained by a D-S evidence theory fusion rule, judging data attributes, and determining the quality grade of the assisting base station data so as to authenticate the authenticity of the assisting base station;
and the authentication unit is used for authenticating the vehicle by using the accurate vehicle behavior prediction result and the real-time behavior data when the vehicle reaches the authentication base station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911071679.2A CN111065089B (en) | 2019-11-05 | 2019-11-05 | Internet of vehicles bidirectional authentication method and device based on crowd sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911071679.2A CN111065089B (en) | 2019-11-05 | 2019-11-05 | Internet of vehicles bidirectional authentication method and device based on crowd sensing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111065089A CN111065089A (en) | 2020-04-24 |
| CN111065089B true CN111065089B (en) | 2020-11-27 |
Family
ID=70298463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911071679.2A Active CN111065089B (en) | 2019-11-05 | 2019-11-05 | Internet of vehicles bidirectional authentication method and device based on crowd sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111065089B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112363822B (en) * | 2020-09-22 | 2024-02-13 | 长沙理工大学 | Collaborative task allocation method for crowdsourcing platform |
| CN113283516B (en) * | 2021-06-01 | 2023-02-28 | 西北工业大学 | Multi-sensor data fusion method based on reinforcement learning and D-S evidence theory |
| CN113297597B (en) * | 2021-06-09 | 2022-09-06 | 河南科技大学 | Social networking communication group establishing method based on position privacy protection |
| CN113452552B (en) * | 2021-06-16 | 2022-10-21 | 西安电子科技大学 | Information entropy perception-based super-multi-target controller placement method |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902794B (en) * | 2010-08-02 | 2012-10-17 | 清华大学深圳研究生院 | Double-layer fast switching method based on motion prediction in V2I communication |
| US8681023B2 (en) * | 2011-03-02 | 2014-03-25 | Ford Global Technologies, Llc | Method and apparatus for vehicle tracking |
| GB201401009D0 (en) * | 2014-01-21 | 2014-03-05 | Qinetiq Ltd | Vehicle identification |
| CN108595922B (en) * | 2015-11-03 | 2020-10-16 | 深圳腾信百纳科技有限公司 | System and method for identity security access authentication |
| CN107086996B (en) * | 2017-04-14 | 2020-06-16 | 重庆邮电大学 | Internet of vehicles security authentication device and method |
| US10618498B2 (en) * | 2017-07-27 | 2020-04-14 | Uatc Llc | Systems and methods for providing user access to an autonomous vehicle |
| CN108900634A (en) * | 2018-08-02 | 2018-11-27 | 安徽江淮汽车集团股份有限公司 | A kind of remote vehicle control method and system based on V2X |
| CN109862114B (en) * | 2019-03-12 | 2021-08-10 | 南京邮电大学 | Safe vehicle crowd-sourcing sensing method based on fog calculation |
| CN109946661A (en) * | 2019-04-26 | 2019-06-28 | 陕西师范大学 | A kind of trailer-mounted radar data processing algorithm verifying system |
| CN110121171B (en) * | 2019-05-10 | 2022-09-27 | 青岛大学 | Trust prediction method based on exponential smoothing method and gray model |
| CN110139214A (en) * | 2019-06-26 | 2019-08-16 | 湖南大学 | Vehicle position privacy protection method based on virtual location in a kind of VANET |
| CN110364026A (en) * | 2019-08-09 | 2019-10-22 | 山东理工大学 | A kind of vehicle follow-up strategy safe verification method and system based on state reachable set |
-
2019
- 2019-11-05 CN CN201911071679.2A patent/CN111065089B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111065089A (en) | 2020-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111065089B (en) | Internet of vehicles bidirectional authentication method and device based on crowd sensing | |
| Hbaieb et al. | A survey of trust management in the Internet of Vehicles | |
| García-Magariño et al. | Security in vehicles with IoT by prioritization rules, vehicle certificates, and trust management | |
| CN110602117B (en) | Vehicle networking node consistency consensus method based on block chain | |
| Saravanan et al. | Routing using reinforcement learning in vehicular ad hoc networks | |
| Chowdhury et al. | Trustworthiness of self-driving vehicles for intelligent transportation systems in industry applications | |
| Hussain et al. | Comparative studied based on attack resilient and efficient protocol with intrusion detection system based on deep neural network for vehicular system security | |
| Oluoch | A distributed reputation scheme for situation awareness in vehicular ad hoc networks (VANETs) | |
| US11250362B2 (en) | Machine learning based decentralized business planning system | |
| CN113301047A (en) | Vehicle networking node consistency consensus method based on malicious node attack detection | |
| CN113380024B (en) | Reputation updating method and trust calculation method based on Internet of vehicles | |
| Bhargava et al. | DUEL: Dempster uncertainty-based enhanced-trust level scheme for VANET | |
| Evangeline et al. | A two-phase fuzzy based access network selection scheme for vehicular ad hoc networks | |
| CN112672321B (en) | Relay vehicle selection method and device | |
| CN109982327A (en) | A kind of ad hoc network communication method, device, equipment and readable storage medium storing program for executing | |
| Sarangi | Malicious Attacks Detection Using Trust Node Centric Weight Management Algorithm in Vehicular Platoon | |
| Ji et al. | A blockchain-based vehicle platoon leader updating scheme | |
| CN114915940A (en) | Vehicle-road communication link matching method and system based on edge cloud computing | |
| CN115802358A (en) | Multi-step DDoS prediction poisoning attack based on reinforcement learning and defense method thereof | |
| Wang et al. | PDTM: A Provenance-Driven dynamic Trust Management model for IoVs | |
| CN114048482A (en) | LSTM risk prediction access control method based on block chain | |
| Zhao et al. | Blockchain-enabled targeted information dissemination framework in vehicular networks | |
| US20240102817A1 (en) | Road accident detection and avoidance using blockchain | |
| JP7262690B1 (en) | Traffic jam prediction device and traffic jam prediction method | |
| CN116707913A (en) | Intra-domain identity trust management method based on differential privacy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |