CN110992690A - False data detection method based on space-time outliers in Internet of vehicles - Google Patents

False data detection method based on space-time outliers in Internet of vehicles Download PDF

Info

Publication number
CN110992690A
CN110992690A CN201911200558.3A CN201911200558A CN110992690A CN 110992690 A CN110992690 A CN 110992690A CN 201911200558 A CN201911200558 A CN 201911200558A CN 110992690 A CN110992690 A CN 110992690A
Authority
CN
China
Prior art keywords
data
time
space
neighborhood
vehicles
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.)
Granted
Application number
CN201911200558.3A
Other languages
Chinese (zh)
Other versions
CN110992690B (en
Inventor
刘伎昭
王佩雪
张俊宝
杨昌霖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhongyuan University of Technology
Original Assignee
Zhongyuan University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhongyuan University of Technology filed Critical Zhongyuan University of Technology
Priority to CN201911200558.3A priority Critical patent/CN110992690B/en
Publication of CN110992690A publication Critical patent/CN110992690A/en
Application granted granted Critical
Publication of CN110992690B publication Critical patent/CN110992690B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/12—Detection or prevention of fraud
    • H04W12/121—Wireless intrusion detection systems [WIDS]; Wireless intrusion prevention systems [WIPS]
    • H04W12/122—Counter-measures against attacks; Protection against rogue devices
    • 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)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides a false data detection method based on space-time outliers in a vehicle networking, which comprises the following steps: the method comprises the steps of collecting data reported by vehicles by using RSUs deployed on roadsides in the Internet of vehicles, grouping the data according to time and space attributes, quantifying the similarity degree of the data in each group, and detecting an abnormal data group with significant data divergence based on a Bayesian method. The method comprises the steps of defining a highly relevant area of traffic data in a road space as a space-time neighborhood, establishing the space-time neighborhood based on historical data, and detecting false data which obviously deviates from the whole body by using an outlier detection algorithm. The false data is preliminarily judged by utilizing the time-space correlation characteristics of traffic data in the real traffic environment, and the deviation degree of each data in the time-space neighborhood and the whole is quantized by combining a time-space outlier detection algorithm to find out the false data, so that the road traffic condition can be correctly and effectively evaluated, the traffic management behavior capability is improved, and thus, abundant and diverse upper-layer applications such as traffic safety, intelligent navigation and the like are realized.

Description

False data detection method based on space-time outliers in Internet of vehicles
Technical Field
The invention relates to the technical field of Internet of vehicles safety, in particular to a false data detection method based on space-time outliers in the Internet of vehicles.
Background
The internet of vehicles enables the vehicles to sense the motion states (position, speed, direction, instantaneous acceleration and the like) of the vehicles and surrounding vehicles and road environment information (weather, road conditions and the like) during the driving process by installing various sensors (GPS, radar, acceleration sensor and the like) on the vehicles. The wireless communication between vehicles, between vehicle-RoadSide infrastructures or between vehicle-background management centers is realized by using a special Short-distance communication technology (DSRC) and a 5G technology, and the information is shared to surrounding vehicles or is provided to the background traffic management center after being subjected to data aggregation through a pre-deployed RoadSide infrastructure (RSU). Based on the information, the road traffic condition can be effectively evaluated, the vehicle travel time can be calculated, or abnormal events such as traffic jam and traffic accidents can be sensed, so that abundant and diverse upper-layer applications such as traffic safety and intelligent navigation can be realized.
The network architecture of the Internet of vehicles has the characteristics of distribution and self-organization, vehicles have equal positions, and any vehicle can send traffic data. Vehicle nodes may provide inaccurate or even erroneous data due to malicious attacks, user selfishness, sensor failures, etc. Such as: in order to reach the destination faster, the selfish vehicle driver sends the fake high vehicle density and low driving speed, the false image of traffic jam is made, and the surrounding vehicles are induced to drive out of the current road, so that the aim of clearing the road is fulfilled. Vehicles may be misled by wrong data to make wrong reactions, such as sudden braking, lane change, etc., which not only interfere with normal traffic order, but may even cause traffic accidents in extreme cases. The background management center can also make wrong judgment on the traffic condition, and make wrong traffic management behaviors, so that traffic disorder is caused.
Disclosure of Invention
Aiming at the defects in the background technology, the invention provides a false data detection method based on space-time outliers in the Internet of vehicles, and solves the technical problem of traffic confusion caused by injecting false traffic data into an internal enemy in the existing Internet of vehicles.
The technical scheme of the invention is realized as follows:
a false data detection method based on space-time outliers in the Internet of vehicles comprises the following steps:
s1, data collection and grouping: the RSU deployed on the roadside in the Internet of vehicles is responsible for collecting data reported by vehicle nodes in a coverage area, and grouping the data according to the time and space attributes of the data to obtain a data group;
s2, data abnormity detection: calculating a data consistency factor C of the data group by using the RSU, judging whether the value of the data consistency factor C is obviously changed on a time axis based on a Bayesian decision and hypothesis test method, if so, marking the data group as an abnormal data group, and executing the step S3, otherwise, returning to the step S1;
s3, establishing a space-time neighborhood: constructing a space-time field of the abnormal data set by using the associated entropy;
s4, space-time outlier detection: and detecting outliers in the space-time neighborhood of the abnormal data set by using an outlier detection algorithm, and determining the false data injected by the attacker.
The method for detecting data abnormality in step S2 includes:
s21, the calculation method of the data consistency factor C of the data group is as follows:
Figure BDA0002295777050000021
wherein G iss(t)＝{d1,d2,…dNThe data group of the traffic data collected by a group of N different vehicles in the road section s and the time slot T, where T is 1,2, … T, i, j is 1,2, …, N;
s22, establishing a hypothesis model according to a hypothesis testing principle:
Figure BDA0002295777050000022
wherein, thetaT＝{C1,…CTIs the consistency factor vector, H, of the current data set at time slots 1,2, … T0Is represented by CtDoes not significantly change in value of (A), H1Is represented by CtAt time slot t0A significant change occurs;
and S23, determining the probability that the value of the data consistency factor C is changed significantly in time t by using posterior probability according to the Bayes principle:
Figure BDA0002295777050000023
wherein q is ∈ {1, 2}, P (H)q|ΘT) As a likelihood function, P (H)q) Is a prior probability;
s24, judging the consistency factor C of the data at t-t according to the likelihood formula0A significant change occurred:
Figure BDA0002295777050000024
s25, data group Gs(t0) And (4) regarding the data as an abnormal data set, constructing a space-time neighborhood, and determining the false data injected by an attacker.
The space-time neighborhood comprises a direct time neighborhood, a direct space neighborhood and an indirect space-time neighborhood; the direct temporal neighborhood: data group G comprising same position and adjacent time slot collections(t-1)、Gs(t + 1); the direct spatial neighborhood: including data G of adjacent segments in the same time slots-1(t)、Gs+1(t); the indirect space-time neighborhood: including data sets that are both indirectly adjacent in time and space.
The method for constructing the space-time domain of the abnormal data set by using the associated entropy in the step S3 includes:
s31, making X be a random variable, representing traffic data collected by vehicles, and discretizing the X value into Q small subintervals;
s32, data group Gs(t)＝(d1,d2,…dN) Data reported for N vehicles d1,d2,…dNWherein d isi∈[0,Q-1](ii) a Among the N' velocity values, the number of velocity values falling in the j-th subinterval is sjWherein j is more than or equal to 0 and less than or equal to Q-1, and the entropy of the random variable X is as follows:
Figure BDA0002295777050000031
s33, order XG1As a data group G1(t) random variable, XG2As a data group G2(t) random variable, then data set G1(t) and G2The associated entropy of (t) is:
Figure BDA0002295777050000032
wherein the content of the first and second substances,
Figure BDA0002295777050000033
to represent
Figure BDA0002295777050000034
And is
Figure BDA0002295777050000035
Probability of (2), then random variable
Figure BDA0002295777050000036
And
Figure BDA0002295777050000037
the mutual information is:
Figure BDA0002295777050000038
s34, use data group G1(t) and G2Redundant information R (G) between (t)1,G2) Quantized data set G1(t) and G2Correlation between (t):
Figure BDA0002295777050000039
s35, presence of S-l-Or s + l+Satisfy the requirement of
Figure BDA00022957770500000313
Or
Figure BDA00022957770500000314
But do not
Figure BDA00022957770500000315
Or
Figure BDA00022957770500000316
Will l-Or l+Definition data group Gs(t) a spatio-temporal neighborhood boundary; gs(t) a spatial neighborhood of
Figure BDA00022957770500000317
Wherein l-And l+Are all positive integers;
s36, the time neighborhood of a data set comprises data sets collected in the same road section but different time slots if t-l exists-Or t + l+Satisfy R (G)s(t),Gs(t-l-) Is not less than epsilon or R (G)s(t),Gs(t+l+) Is not less than epsilon, but R (G)s(t),Gs(t-l--1))<ε or R (G)s(t),Gs(t+l++1))<ε, title l-Or l+Is the spatio-temporal neighborhood boundary.
The method for detecting outliers in the spatio-temporal neighborhood of the abnormal data set by using the outlier detection algorithm in step S4 includes:
s41, forming space-time neighborhood of an abnormal data set by M data items, and recording as M ═ o1,o2,..pmCalculate the mean of the data in M:
Figure BDA00022957770500000310
s42, calculating the standard deviation of the data in M as
Figure BDA00022957770500000311
Wherein the content of the first and second substances,
Figure BDA00022957770500000312
s43, checking whether each data item o is in the interval [ A-k ]0·σ,A+k0·σ]Wherein k is0>1 is a predefined constant parameter; data items falling outside this interval are considered outliers, are considered spurious data injected by adversaries, and data providers are considered attackers.
The beneficial effect that this technical scheme can produce: the false data is preliminarily judged by utilizing the time-space correlation characteristics of traffic data in the real traffic environment, and the deviation degree of each data in the time-space neighborhood and the whole is quantized by combining a time-space outlier detection algorithm to find out the false data, so that the road traffic condition can be correctly and effectively evaluated, the traffic management behavior capability is improved, and thus, abundant and diverse upper-layer applications such as traffic safety, intelligent navigation and the like are realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a map useful in embodiments of the present invention;
FIG. 3 is a graph of the detection rate of the present invention and comparison algorithm at different vehicle densities;
FIG. 4 is a graph of the detection rate of the present invention and comparison algorithm at different aggressor ratios;
FIG. 5 shows the false detection rate of the present invention and the comparison algorithm at different aggressor ratios.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
The present invention uses discretized time and road space. The time is divided into several time slots of equal length, denoted as t 1,2, …. The road is divided into several road segments, denoted as s1, 2, …. The vehicle utilizes vehicle-mounted sensors to collect traffic data such as vehicle density, speed, flow and the like in the driving process, and shares the data through a DSRC (Dedicated Short-Range Communications) wireless communication technology so as to realize multi-type upper-layer application such as traffic safety, traffic management and the like. The general format of the message is o ═ t, (lon, lat), d, vid }, where t is the time of data acquisition, lon and lat represent the location (latitude and longitude) of data acquisition, d is the traffic data acquired by the vehicle, and vid is the vehicle id identifier. It is assumed that rsus (roadside units) are deployed in advance along the road and are responsible for collecting data reported by vehicles, the detection method provided by the invention is executed, the authenticity of the data reported by the vehicles is evaluated, and the false data injected by an attacker is detected.
As shown in fig. 1, an embodiment of the present invention provides a method for detecting false data based on spatio-temporal outliers in an internet of vehicles, which includes the following specific steps:
s1, data collection and grouping: the RSU deployed on the roadside in the Internet of vehicles is responsible for collecting data reported by vehicle nodes in a coverage area, and grouping the data according to the time and space attributes of the data to obtain a data group; taking the section s as an example, the RSU groups the received data into: gs(1),Gs(2),…Gs(T) represents data collected in a series of time slots 1,2, … T in the same road segment s.
S2, data abnormity detection: calculating a data consistency factor C of the data group by using the RSU, judging whether the value of the data consistency factor C is obviously changed on a time axis based on a Bayesian decision and hypothesis test method, if so, marking the data group as an abnormal data group, and executing the step S3, otherwise, returning to the step S1;
the data anomaly detection method comprises the following steps:
s21, defining the data consistency as the data similarity degree in the data consistency factor quantization group, wherein the calculation method of the data consistency factor C of the data group comprises the following steps:
Figure BDA0002295777050000051
wherein G iss(t)＝{d1,d2,…dNThe data group of the traffic data collected by a group of N different vehicles in the road section s and the time slot T, where T is 1,2, … T, i, j is 1,2, …, N;
s22, establishing a hypothesis model according to a hypothesis testing principle:
Figure BDA0002295777050000052
wherein, thetaT＝{C1,…CTH is a vector of consistency factors for the current data set at time slots 1,2, … T, H0For the original hypothesis, represent CtDoes not significantly change in value of (A), H1For alternative assumptions, denote CtAt time slot t0A significant change occurs;
and S23, determining the probability that the value of the data consistency factor C is changed significantly in time t by using posterior probability according to the Bayes principle:
wherein q is ∈ {1, 2}, P (H)q|ΘT) As a likelihood function, P (H)q) Is a prior probability;
s24, judging the consistency factor C of the data at t-t according to the likelihood formula0A significant change occurred:
Figure BDA0002295777050000054
s25, data group Gs(t0) Is treated as an anomalous data set and a further detection process is performed to determine the spurious data injected by the attacker.
S3, establishing a space-time neighborhood: constructing a space-time field of the abnormal data set by using the associated entropy; data sets with data anomalies can be found using step S2, but it cannot be determined which data within the set was injected by an attacker. The invention utilizes historical traffic data to construct the space-time field of an abnormal data set, compares each data item with data in a space-time neighborhood, and discovers significantly deviated data items by using a space-time outlier algorithm. In a road scene, traffic data between two adjacent positions generally has strong space-time relevance, but the relevance degree is influenced by various factors such as geographical positions, traffic rules and the like. For example, in an expressway scene, two adjacent road sections in the front and the back present strong relevance, while in an urban scene, the two road sections in the front and the back divided by the intersection are close in geographic position, but are influenced by traffic lights, so that the vehicle motion relevance is weak. Spatio-temporal neighborhoods are used to represent geographical areas in road space where traffic data is highly correlated. In particular, data group Gs(t) data comprising time slot s and road segment t, wherein the time-space neighborhood comprises a direct time neighborhood, a direct space neighborhood and an indirect time-space neighborhood; the direct temporal neighborhood: data group G comprising same position and adjacent time slot collections(t-1)、Gs(t + 1); the direct spatial neighborhood: including data G of adjacent segments in the same time slots-1(t)、Gs+1(t); the indirect space-time neighborhood: including data sets that are both indirectly adjacent in time and space.
The method for constructing the space-time field of the abnormal data set by utilizing the associated entropy comprises the following steps:
s31, making X be a random variable, representing traffic data collected by vehicles, and discretizing the X value into Q small subintervals;
s32, data group Gs(t)＝(d1,d2,…dN) Data reported for N vehicles d1,d2,…dN，di∈[0,Q-1](ii) a Among the N' velocity values, the number of velocity values falling in the j-th subinterval is sjWherein j is more than or equal to 0 and less than or equal to Q-1, and the entropy of the random variable X is as follows:
Figure BDA0002295777050000061
s33, order XG1As a data group G1(t) random variable, XG2As a data group G2(t) random variable, then data set G1(t) and G2The associated entropy of (t) is:
Figure BDA0002295777050000062
wherein the content of the first and second substances,
Figure BDA0002295777050000063
to represent
Figure BDA0002295777050000064
And is
Figure BDA0002295777050000065
Probability of (2), then random variable
Figure BDA0002295777050000066
And
Figure BDA0002295777050000067
the mutual information is:
Figure BDA0002295777050000068
s34, use data group G1(t) and G2Redundant information R (G) between (t)1,G2) Quantized data set G1(t) and G2Correlation between (t):
Figure BDA0002295777050000069
s35, presence of S-l-Or s + l+Satisfy the requirement of
Figure BDA00022957770500000613
Or
Figure BDA00022957770500000614
But do not
Figure BDA00022957770500000615
Or
Figure BDA00022957770500000616
Wherein l-And l+Are all positive integers, are-Or l+Definition data group Gs(t) a spatio-temporal neighborhood boundary; gs(t) a spatial neighborhood of
Figure BDA00022957770500000617
S36, the time neighborhood of a data set comprises data sets collected in the same road section but different time slots if t-l exists-Or t + l+Satisfy R (G)s(t),Gs(t-l-) Is not less than epsilon or R (G)s(t),Gs(t+l+) Is not less than epsilon, but R (G)s(t),Gs(t-l--1))<ε or R (G)s(t),Gs(t+l++1))<ε, title l-Or l+Is the spatio-temporal neighborhood boundary.
S4, space-time outlier detection: and detecting outliers in the space-time neighborhood of the abnormal data set by using an outlier detection algorithm, and determining the false data injected by the attacker.
The method for detecting outliers in the spatio-temporal neighborhood of the abnormal data set by using the outlier detection algorithm comprises the following steps:
s41, forming space-time neighborhood of an abnormal data set by M data items, and recording as M ═ o1,o2,..omCalculate the mean of the data in M:
Figure BDA00022957770500000610
s42, calculating the standard deviation of the data in M as
Figure BDA00022957770500000611
Wherein the content of the first and second substances,
Figure BDA00022957770500000612
s43, checking whether each data item o is in the interval [ A-k ]0·σ,A+k0·σ]Wherein k is0>1 is a predefined constant parameter; data items falling outside this interval are considered outliers, are considered spurious data injected by adversaries, and data providers are considered attackers.
The effectiveness of the method is verified based on a simulation method, and compared with the existing method for analysis, the technical advantages of the method are evaluated.
(1) Simulation method
In the simulation experiment, traffic scenes with high reality are generated by using traffic simulation software SUMO, an electronic map provided by an OpenStreetMap project is used, and the size of a simulation area is 2000 m × 2000 m, as shown in FIG. 2. All roads are bidirectional, 3 lanes in each direction, and the position and direction of the vehicle entering the simulation area are randomly selected. The average density of the vehicles in the simulation area is 75 vehicles/kilometer, and the real-time running speed is reported to the RSU during the running process of the vehicles. The simulation was run 100 times and all simulation data were the average of 100 runs.
In order to verify the accuracy of the detection method, a collusion attack scene is set, and a false image of traffic jam is made by sending a forged low vehicle speed on the assumption that a plurality of collusion attackers exist. The attack position and time are randomly selected, the forged vehicle speed is randomly selected between [0,10] km/h, the vehicle speed is transmitted once in one time slot by each vehicle, and the proportion of an attacker in all vehicles is recorded as gamma.
The performance of the detection method was evaluated using the following metric. The Detection Rate (DR) is defined as the proportion of all the false data that is correctly detected, and the False Positive Rate (FPR) is defined as the probability of data being incorrectly detected, including the false data being detected as true data and the false data being detected as true data.
(2) Results and analysis
The influence of the vehicle density and the proportion of the attackers on the algorithm performance is examined. The simulation results are given in fig. 3 to 5. FIG. 3 shows the performance comparison evaluation of the method of the present invention and the LEAVE (local evaluation of attemperers and rating evaluators) protocol under different vehicle densities. The detection rate of the method is higher than LEAVE under all vehicle densities, and when the vehicle density is 80 vehicles/kilometer, the detection rate of the method is 98.3%, and the detection rate of a LEAVE protocol is 91.5%. In addition, the detection rate of the method is lower than that of a high-density vehicle scene when the vehicle density is low, and the detection rate is 92.0% when the vehicle density is 20 vehicles/kilometer. The detection rate was 99.9% when the vehicle density was 200 vehicles/km. The distance between the vehicles is large when the vehicle density is low, the space limitation between the vehicles is not obvious, the vehicles can run at an ideal speed, and the running speeds of the vehicles are different greatly due to the fact that different types of vehicles have different ideal speeds, so that data collected by the vehicles present large variance, and the difficulty is increased for correct identification.
Fig. 4 and 5 evaluate the collusion attack resistance of the inventive method and the LEAVE protocol. Fig. 3 shows the detection rate for different aggressor scales. From the figure, it can be seen that the detection rate of the method of the present invention at low aggressor ratio is similar to LEAVE. However, the detection rate of the LEAVE protocol rapidly decreases when the aggressor proportion increases, and 65.5% and 40% when the aggressor proportion reaches 30% and 40%, respectively. The method of the invention shows stronger collusion attack resistance. When the proportion of attackers reaches 25%, the detection rate of 96.5% can still be achieved. Meanwhile, the false detection rate is also obviously lower than LEAVE, and is 5.9% under the condition of 25% of attacker proportion.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (5)

1. A false data detection method based on space-time outliers in the Internet of vehicles is characterized by comprising the following steps:
s1, data collection and grouping: the RSU deployed on the roadside in the Internet of vehicles is responsible for collecting data reported by vehicle nodes in a coverage area, and grouping the data according to the time and space attributes of the data to obtain a data group;
s2, data abnormity detection: calculating a data consistency factor C of the data group by using the RSU, judging whether the value of the data consistency factor C is obviously changed on a time axis based on a Bayesian decision and hypothesis test method, if so, marking the data group as an abnormal data group, and executing the step S3, otherwise, returning to the step S1;
s3, establishing a space-time neighborhood: constructing a space-time field of the abnormal data set by using the associated entropy;
s4, space-time outlier detection: and detecting outliers in the space-time neighborhood of the abnormal data set by using an outlier detection algorithm, and determining the false data injected by the attacker.
2. The method for detecting false data based on spatio-temporal outliers in car networking according to claim 1, wherein the method for detecting data anomaly in step S2 is:
s21, the calculation method of the data consistency factor C of the data group is as follows:
Figure FDA0002295777040000011
wherein G iss(t)＝{d1,d2,…dNThe data group of the traffic data collected by a group of N different vehicles in the road section s and the time slot T, where T is 1,2, … T, i, j is 1,2, …, N;
s22, establishing a hypothesis model according to a hypothesis testing principle:
Figure FDA0002295777040000012
wherein, thetaT＝{C1,…CTIs the consistency factor vector, H, of the current data set at time slots 1,2, … T0Is represented by CtDoes not significantly change in value of (A), H1Is represented by CtAt time slot t0A significant change occurs;
and S23, determining the probability that the value of the data consistency factor C is changed significantly in time t by using posterior probability according to the Bayes principle:
Figure FDA0002295777040000013
wherein q is ∈ {1, 2}, P (H)q|ΘT) As a likelihood function, P (H)q) Is a prior probability;
s24, judging the consistency factor C of the data at t-t according to the likelihood formula0A significant change occurred:
Figure FDA0002295777040000014
s25, data group Gs(t0) And (4) regarding the data as an abnormal data set, constructing a space-time neighborhood, and determining the false data injected by an attacker.
3. The method for detecting false data based on spatio-temporal outliers in the internet of vehicles according to claim 1, wherein the spatio-temporal neighborhood comprises a direct temporal neighborhood, a direct spatial neighborhood and an indirect spatio-temporal neighborhood; the direct temporal neighborhood: data group G comprising same position and adjacent time slot collections(t-1)、Gs(t + 1); the direct spatial neighborhood: including data G of adjacent segments in the same time slots-1(t)、Gs+1(t); the indirect space-time neighborhood: including data sets that are both indirectly adjacent in time and space.
4. The method for detecting the spatiotemporal outlier-based false data in the internet of vehicles according to claim 1, wherein the method for constructing the spatiotemporal domain of the abnormal data set by using the associated entropy in the step S3 is as follows:
s31, making X be a random variable, representing traffic data collected by vehicles, and discretizing the X value into Q small subintervals;
s32, data group Gs(t)＝(d1,d2,…dN) Data reported for N vehicles d1,d2,…dNWherein d isi∈[0,Q-1](ii) a Among the N' velocity values, the number of velocity values falling in the j-th subinterval is sjWherein j is more than or equal to 0 and less than or equal to Q-1, and the entropy of the random variable X is as follows:
Figure FDA0002295777040000021
s33, order XG1As a data group G1(t) random variable, XG2As a data group G2(t) random variable, then data set G1(t) and G2The associated entropy of (t) is:
Figure FDA0002295777040000022
wherein the content of the first and second substances,
Figure FDA0002295777040000023
to represent
Figure FDA0002295777040000024
And is
Figure FDA0002295777040000025
Probability of (2), then random variable
Figure FDA0002295777040000026
And
Figure FDA0002295777040000027
the mutual information is:
Figure FDA0002295777040000028
s34, use data group G1(t) and G2Redundant information R (G) between (t)1,G2) Quantized data set G1(t) and G2Correlation between (t):
Figure FDA0002295777040000029
s35, presence of S-l-Or s + l+Satisfy R (G)s(t),
Figure FDA00022957770400000213
Or R (G)s(t),
Figure FDA00022957770400000214
But R (G)s(t),
Figure FDA00022957770400000215
Or R (G)s(t),
Figure FDA00022957770400000216
Will l-Or l+Definition data group Gs(t) a spatio-temporal neighborhood boundary; gs(t) a spatial neighborhood of
Figure FDA00022957770400000217
Wherein l-And l+Are all positive integers;
s36, the time neighborhood of a data set comprises data sets collected in the same road section but different time slots if t-l exists-Or t + l+Satisfy R (G)s(t),Gs(t-l-) Is not less than epsilon or R (G)s(t),Gs(t+l+) Is not less than epsilon, but R (G)s(t),Gs(t-l--1))<ε or R (G)s(t),Gs(t+l++1))<ε, title l-Or l+Is the spatio-temporal neighborhood boundary.
5. The method for detecting spatio-temporal outlier-based false data in Internet of vehicles according to claim 1, wherein the method for detecting outliers in the spatio-temporal neighborhood of the outlier data set by using the outlier detection algorithm in step S4 comprises the following steps:
s41, forming space-time neighborhood of an abnormal data set by M data items, and recording as M ═ o1,o2,..omCalculate the mean of the data in M:
Figure FDA00022957770400000210
s42, calculating the standard deviation of the data in M as
Figure FDA00022957770400000211
Wherein the content of the first and second substances,
Figure FDA00022957770400000212
s43, checking whether each data item o is in the interval [ A-k ]0·σ,A+k0·σ]Wherein k is0>1 is a predefined constant parameter; data items falling outside this interval are considered outliers, are considered spurious data injected by adversaries, and data providers are considered attackers.
CN201911200558.3A 2019-11-29 2019-11-29 False data detection method based on space-time outliers in Internet of vehicles Active CN110992690B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911200558.3A CN110992690B (en) 2019-11-29 2019-11-29 False data detection method based on space-time outliers in Internet of vehicles

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911200558.3A CN110992690B (en) 2019-11-29 2019-11-29 False data detection method based on space-time outliers in Internet of vehicles

Publications (2)

Publication Number Publication Date
CN110992690A true CN110992690A (en) 2020-04-10
CN110992690B CN110992690B (en) 2021-09-17

Family

ID=70088311

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911200558.3A Active CN110992690B (en) 2019-11-29 2019-11-29 False data detection method based on space-time outliers in Internet of vehicles

Country Status (1)

Country Link
CN (1) CN110992690B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113255593A (en) * 2021-06-25 2021-08-13 北京市水利自动化研究所 Sensor information anomaly detection method facing space-time analysis model
CN115311838A (en) * 2022-07-22 2022-11-08 重庆大学 Method for evaluating vehicle collaborative consistency in tunnel entrance area

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198296A (en) * 2013-03-07 2013-07-10 中国科学技术大学 Method and device of video abnormal behavior detection based on Bayes surprise degree calculation
CN105654735A (en) * 2016-03-24 2016-06-08 安徽四创电子股份有限公司 Rapid fake-licensed car recognizing method based on outlier analysis algorithm
CN107247954A (en) * 2017-06-16 2017-10-13 山东省计算中心（国家超级计算济南中心） A kind of image outlier detection method based on deep neural network
CN108400973A (en) * 2018-02-02 2018-08-14 中原工学院 Spoofing detection method based on traffic flow model in vehicular ad hoc network
US20190097827A1 (en) * 2013-01-18 2019-03-28 Irobot Corporation Mobile robot providing environmental mapping for household environmental control
US20190172345A1 (en) * 2017-12-04 2019-06-06 Institute For Information Industry System and method for detecting dangerous vehicle
CN109902612A (en) * 2019-02-22 2019-06-18 北京工业大学 A kind of monitor video method for detecting abnormality based on unsupervised learning
US20190213874A1 (en) * 2018-01-05 2019-07-11 Here Global B.V. Multi-modal traffic detection

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190097827A1 (en) * 2013-01-18 2019-03-28 Irobot Corporation Mobile robot providing environmental mapping for household environmental control
CN103198296A (en) * 2013-03-07 2013-07-10 中国科学技术大学 Method and device of video abnormal behavior detection based on Bayes surprise degree calculation
CN105654735A (en) * 2016-03-24 2016-06-08 安徽四创电子股份有限公司 Rapid fake-licensed car recognizing method based on outlier analysis algorithm
CN107247954A (en) * 2017-06-16 2017-10-13 山东省计算中心（国家超级计算济南中心） A kind of image outlier detection method based on deep neural network
US20190172345A1 (en) * 2017-12-04 2019-06-06 Institute For Information Industry System and method for detecting dangerous vehicle
US20190213874A1 (en) * 2018-01-05 2019-07-11 Here Global B.V. Multi-modal traffic detection
CN108400973A (en) * 2018-02-02 2018-08-14 中原工学院 Spoofing detection method based on traffic flow model in vehicular ad hoc network
CN109902612A (en) * 2019-02-22 2019-06-18 北京工业大学 A kind of monitor video method for detecting abnormality based on unsupervised learning

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
H. C. MANDHARE: "A comparative study of cluster based outlier detection, distance", 《 2017 INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTING AND CONTROL SYSTEMS(ICICCS)》 *
刘伎昭: "车载自组织网络安全关键技术研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113255593A (en) * 2021-06-25 2021-08-13 北京市水利自动化研究所 Sensor information anomaly detection method facing space-time analysis model
CN113255593B (en) * 2021-06-25 2021-09-24 北京市水利自动化研究所 Sensor information anomaly detection method facing space-time analysis model
CN115311838A (en) * 2022-07-22 2022-11-08 重庆大学 Method for evaluating vehicle collaborative consistency in tunnel entrance area
CN115311838B (en) * 2022-07-22 2023-09-26 重庆大学 Vehicle cooperative consistency evaluation method for tunnel entrance area

Also Published As

Publication number Publication date
CN110992690B (en) 2021-09-17

Similar Documents

Publication Publication Date Title
Chen et al. A short-term traffic prediction model in the vehicular cyber–physical systems
KR101493360B1 (en) Method of vehicle driving managing through detection state change of around cars and system for it
US20240042995A1 (en) Vehicle operation using a dynamic occupancy grid
Kamel et al. A misbehavior authority system for sybil attack detection in c-its
CN104851295A (en) Method and system for acquiring road condition information
CN110992690B (en) False data detection method based on space-time outliers in Internet of vehicles
Watta et al. Vehicle position and context detection using V2V communication
Fox et al. Crowdsourcing undersampled vehicular sensor data for pothole detection
CN112712729B (en) Method and system for predicting motion trajectory
CN110445788A (en) The trust evaluation system and method for content oriented under a kind of vehicular ad hoc network environment
Kumar et al. Study on road traffic congestion: A review
Sabour et al. Deepflow: Abnormal traffic flow detection using siamese networks
Wang et al. Detecting urban traffic congestion with single vehicle
Wang et al. Traffic speed estimation based on multi-source GPS data and mixture model
Ahmad et al. Microscopic congestion detection protocol in VANETs
Raghu et al. Efficient dead reckoning approach for localization prediction in VANETs
He et al. Crowd-sensing enhanced parking patrol using sharing bikes’ trajectories
CN114495505B (en) Method, device, medium and server for predicting passing duration of congestion road section
CN117178309A (en) Method for creating a map with collision probability
CN117148829A (en) Method and system for autonomous vehicles and non-transitory storage medium
Alzahrani et al. Robust misbehavior detection scheme for vehicular network
CN111507564B (en) Urban road alarm message reliability assessment method integrating space-time correlation
Abuelela et al. Automatic incident detection in VANETs: a bayesian approach
Chetouane et al. Traffic congestion detection: solutions, open issues and challenges
Bhanja et al. Dynamic trafile congestion detection in VANETS using a Fuzzy rule-based system and K-means clustering

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