An artificial intelligence-based vehicle pseudonym replacement method

Abstract

The present application relates to the technical field of vehicle pseudonym replacement, and particularly relates to a vehicle pseudonym replacement method based on artificial intelligence, which comprises the following steps: determining whether pseudonym replacement is needed according to a replacement time interval value or performing secondary determination based on a quantity ratio reference value and a communication characteristic value; determining a target area type according to a to-be-replaced vehicle distribution reference value and a replacement allowance, and determining whether to directly perform pseudonym replacement or perform combined analysis according to the target area type; in the combined analysis, determining a related vehicle combination based on a position synchronization degree or an interaction reference value according to a vehicle analysis coefficient, determining whether to perform sequential setting of a virtual vehicle based on a quantity difference value or setting a virtual vehicle based on a historical attack frequency of a sub-area, and performing pseudonym replacement on to-be-replaced vehicles and virtual vehicles in the related vehicle combination. The present application can reduce the risk of vehicle location privacy leakage.

Description

Technical Field

[0001] This invention relates to the field of vehicle pseudonym replacement technology, and in particular to a vehicle pseudonym replacement method based on artificial intelligence. Background Technology

[0002] In the process of vehicle-to-everything (V2X) communication, vehicles need to perceive their own status and the surrounding environment in real time, and broadcast information such as their identity, current location, speed, and road conditions. However, if there is an attacker within the communication range, it is easy to cause information leakage. In order to protect vehicle location privacy, pseudonyms are used to replace real identities. However, due to problems such as the imbalance between the timing of replacement and security when vehicle density is low and the large system overhead during the pseudonym replacement process, the risk of vehicle location privacy leakage increases. Therefore, how to replace pseudonyms to reduce the risk of vehicle location privacy leakage is an urgent problem to be solved by those skilled in the art.

[0003] Chinese Patent Publication No. CN107396285A discloses a method, device, equipment, and storage medium for vehicle privacy protection. The method includes: acquiring the amount of privacy leakage of a vehicle within a preset pseudonym period; calculating the privacy leakage degree of the vehicle based on the amount of privacy leakage; when the privacy leakage degree reaches a privacy leakage degree threshold, selecting the next pseudonym from the vehicle pseudonym set for replacement and entering the next pseudonym period; otherwise, not replacing the pseudonym or updating the pseudonym period; when the preset pseudonym period ends, selecting the next pseudonym from the vehicle pseudonym set for replacement and entering the next pseudonym period; sending location-related information to an information management center using the pseudonym; the information management center verifies the legality of the pseudonym based on the corresponding certificate; when the verification is successful, accepting the information, thereby using the pseudonym to replace the vehicle's identity and severing the connection between the vehicle's identity information and its spatiotemporal information. However, the above solution has the following problems: it cannot adaptively change the pseudonym according to the actual situation of the area where the vehicle is located, making it difficult to cover the dynamic risks in traffic scenarios, resulting in a significant risk of vehicle location privacy leakage. Summary of the Invention

[0004] To address this, the present invention provides an artificial intelligence-based method for changing vehicle pseudonyms, which overcomes the problem in existing technologies that cannot adaptively change pseudonyms according to the actual conditions of the area where the vehicle is located, making it difficult to cover dynamic risks in traffic scenarios and resulting in a high risk of vehicle location privacy leakage.

[0005] To achieve the above objectives, the present invention provides a method for changing vehicle pseudonyms based on artificial intelligence, comprising:

[0006] The determination of whether pseudonym replacement is unnecessary is based on the replacement time interval value, or a secondary determination is made based on the quantity comparison reference value and the communication characterization value.

[0007] The target area type is determined based on the reference value of the distribution of vehicles to be replaced and the replacement allowance. Based on the target area type, it is determined whether to directly replace the pseudonyms or to perform a combination analysis.

[0008] In the combined analysis, the associated vehicle combinations are determined based on the vehicle analysis coefficient, either based on the location synchronization degree or based on the interaction reference value. The replacement priority coefficient corresponding to each associated vehicle combination is determined based on the number of combined vehicles. The virtual vehicles are set sequentially based on the quantity difference or set based on the historical attack frequency of the sub-region. The pseudonyms of the vehicles to be replaced and the virtual vehicles in the associated vehicle combinations are changed.

[0009] The reference vehicle quantity is determined based on the combined characterization value of the associated vehicle combination, and the quantity difference is determined based on the difference between the reference vehicle quantity and the reference real vehicle quantity.

[0010] Furthermore, for vehicles whose replacement time interval is less than the preset replacement time interval, there is no need to change the pseudonym.

[0011] Furthermore, for vehicles whose replacement time interval value is greater than or equal to the preset replacement time interval value, a secondary determination is made based on the quantity comparison reference value and the communication characterization value.

[0012] In the secondary judgment, for vehicles whose quantity comparison reference value is less than the preset quantity comparison reference value and whose communication characterization value is less than the preset communication characterization value, a determination is made on whether to change the pseudonym based on the cumulative danger value.

[0013] For vehicles whose quantity comparison reference value is greater than or equal to the preset quantity comparison reference value or whose communication characterization value is greater than or equal to the preset communication characterization value, a determination is made on whether to change the pseudonym based on the hazard assessment value.

[0014] Furthermore, the target region types include:

[0015] A target area whose vehicle distribution reference value is greater than or equal to the preset vehicle distribution reference value and whose replacement allowance is greater than or equal to the preset replacement allowance;

[0016] The second type of target area is where the reference value for the distribution of vehicles to be replaced is less than the preset reference value for the distribution of vehicles to be replaced or the replacement allowance is less than the preset replacement allowance.

[0017] Furthermore, for vehicles to be replaced in the first type of target area, pseudonyms are directly replaced; for vehicles to be replaced in the second type of target area, combined analysis is performed.

[0018] Furthermore, for vehicles to be replaced in the second type of target area where the vehicle analysis coefficient is less than the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the position synchronization degree.

[0019] For vehicles to be replaced in the second type of target area where the vehicle analysis coefficient is greater than or equal to the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the interactive reference value.

[0020] The vehicle analysis coefficients are determined based on vehicle density and congestion coefficients.

[0021] Furthermore, the replacement priority coefficient corresponding to each associated vehicle combination is determined based on the number of combined vehicles.

[0022] The replacement priority coefficient for a single associated vehicle combination is positively correlated with the number of vehicles in that combination.

[0023] Furthermore, the number of reference vehicles is determined based on the combined characterization value of the associated vehicle combinations;

[0024] The number of reference vehicles corresponding to a single associated vehicle combination is positively correlated with the combination characteristic value corresponding to that associated vehicle combination.

[0025] Furthermore, for associated vehicle combinations where the quantity difference is greater than or equal to a preset quantity difference, virtual vehicles are sequentially set;

[0026] For associated vehicle combinations where the quantity difference is less than a preset quantity difference, the system determines whether to increase the number of virtual vehicles based on the combination area comparison value, and sets the number of virtual vehicles based on the historical attack frequency of the sub-region.

[0027] Furthermore, for associated vehicle combinations whose combined area comparison value is less than the preset combined area comparison value, the number of virtual vehicles is increased based on the comparison deviation value;

[0028] For a group of associated vehicles whose comparison deviation value is greater than or equal to a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a first preset ratio.

[0029] For a group of associated vehicles whose comparison deviation value is less than a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a second preset ratio.

[0030] The first preset ratio is less than the second preset ratio.

[0031] Compared with the prior art, the beneficial effects of the present invention are that, in the technical solution of the present invention, the usage time of the vehicle pseudonym is effectively reflected by changing the time interval value and the resulting basic privacy leakage risk. Then, based on the time interval value, it is determined whether the pseudonym needs to be changed or whether a secondary judgment is made based on the quantity comparison reference value and communication characterization value. This makes the choice of judgment method more in line with the actual application scenario, which helps to avoid the waste of resources or insufficient privacy protection caused by frequent pseudonym changes in a short period of time. Thus, while ensuring the privacy and security of vehicle location, the judgment efficiency and resource utilization rationality are improved, and a dynamic balance between privacy protection and system overhead is achieved.

[0032] Furthermore, this invention effectively reflects the spatial distribution quality of vehicles to be replaced and the risk controllability of pseudonym replacement within the target area by using the reference value of vehicle distribution and the replacement allowance. Then, the target area type is determined based on the reference value of vehicle distribution and the replacement allowance. Based on the target area type, direct pseudonym replacement or combined analysis can be adaptively selected, which helps to deeply adapt the pseudonym replacement strategy to the actual characteristics of the area. Thus, while ensuring the privacy and security of vehicle location, a dynamic balance is achieved between privacy protection accuracy, replacement efficiency and system resource consumption, improving the practicality and adaptability of the solution in different traffic scenarios.

[0033] Furthermore, this invention effectively reflects the combined impact of vehicle density and traffic congestion on the risk of pseudonym replacement through vehicle analysis coefficients. Then, based on the vehicle analysis coefficients, it adaptively selects to determine the associated vehicle combination based on location synchronization or interaction reference value, which helps to deeply match the construction logic of the associated combination with the actual traffic characteristics of the area, thereby improving the vehicle privacy protection effect.

[0034] Furthermore, the present invention determines the sequential setting of virtual vehicles based on the quantity difference or the setting of virtual vehicles based on the historical attack frequency of sub-regions. This helps to accurately match the configuration strategy of virtual vehicles with the actual needs of the anonymity set. When the quantity difference is large, the scale can be quickly supplemented by sequential setting to avoid privacy leakage due to insufficient quantity. When the quantity difference is small, virtual vehicles are focused on high-risk areas to resist attacks. This avoids the waste of resources caused by blind configuration and can improve the support effect of virtual vehicles for privacy protection in different scenarios, ensuring the security and practicality of the anonymity set of the second type of target area. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the vehicle pseudonym replacement method based on artificial intelligence according to the present invention;

[0036] Figure 2 This is a flowchart illustrating the present invention for determining whether pseudonym replacement is unnecessary based on the replacement time interval value or for secondary determination based on quantity comparison reference value and communication characterization value.

[0037] Figure 3 This is a flowchart illustrating how the target area type is determined based on the distribution reference value of the vehicles to be replaced and the replacement allowance in this invention.

[0038] Figure 4 This is a flowchart illustrating how the present invention sets virtual vehicles sequentially based on quantity differences or sets virtual vehicles based on the historical attack frequency of a sub-region. Detailed Implementation

[0039] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0040] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0041] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0042] Please see Figures 1 to 4 As shown, this invention provides a vehicle pseudonym replacement method based on artificial intelligence, comprising:

[0043] The determination of whether pseudonym replacement is unnecessary is based on the replacement time interval value, or a secondary determination is made based on the quantity comparison reference value and the communication characterization value.

[0044] The target area type is determined based on the reference value of the distribution of vehicles to be replaced and the replacement allowance. Based on the target area type, it is determined whether to directly replace the pseudonyms or to perform a combination analysis.

[0045] In the combined analysis, the associated vehicle combinations are determined based on the vehicle analysis coefficient, either based on the location synchronization degree or based on the interaction reference value. The replacement priority coefficient corresponding to each associated vehicle combination is determined based on the number of combined vehicles. The virtual vehicles are set sequentially based on the quantity difference or set based on the historical attack frequency of the sub-region. The pseudonyms of the vehicles to be replaced and the virtual vehicles in the associated vehicle combinations are changed.

[0046] The reference vehicle quantity is determined based on the combined characterization value of the associated vehicle combination, and the quantity difference is determined based on the difference between the reference vehicle quantity and the reference real vehicle quantity.

[0047] The application scenario of this invention is vehicle pseudonym replacement. This invention has several historical records, each of which records at least one vehicle pseudonym replacement process, including quantity comparison reference values, communication characterization values, vehicle distribution reference values, and replacement allowance. Each historical record also has a corresponding qualification mark, which records whether the vehicle pseudonym replacement process meets the user's needs. The qualification mark can be recorded manually. It is understood that the user can determine whether the vehicle pseudonym replacement process meets the requirements based on self-defined indicators. Self-defined indicators can be, but are not limited to, the maximum pseudonym replacement time, which will not be elaborated here. The maximum pseudonym replacement time is the maximum value among the times required for each vehicle to undergo pseudonym replacement.

[0048] The target area is the geographical region where the "old kana → new kana" switch needs to be completed synchronously. The virtual vehicle is a fake node injected by RSU / edge cloud that only broadcasts and does not drive. It is renamed synchronously with the real vehicle. The virtual vehicle needs to rely on the "three-level legal identity system" to achieve secure access. The regional transportation authority issues a root certificate and then issues a "virtual vehicle management certificate" to the specific RSU / edge cloud. Finally, the RSU / edge cloud generates a sub-certificate based on the management certificate, which contains the virtual vehicle's unique ID, the RSU number to which it belongs, the scenario-specific validity period, and "broadcast only" permissions. This ensures that it can pass the system's dedicated verification branch, so that it is neither judged as an illegal node and discarded, nor mixed into the real vehicle identity system and caused management chaos. The virtual vehicle does not participate in any V2X control message broadcasting and is only used for synchronous kana broadcasting required for privacy protection. The above is the content that is easy for those skilled in the art to understand and will not be elaborated further.

[0049] Specifically, for vehicles whose replacement time interval is less than the preset replacement time interval, there is no need to change the pseudonym.

[0050] Specifically, the present invention sets a continuous cyclic monitoring cycle, and at the end of each monitoring cycle, the time interval value is determined to be changed. The duration of the monitoring cycle can be set according to the user's needs. The greater the user's need for monitoring accuracy, the shorter the duration of the monitoring cycle. One monitoring cycle value is provided, which is 5 minutes.

[0051] For a single vehicle, the time when the vehicle last changed its pseudonym is recorded as the reference time. The change time interval for a single vehicle is the duration between the reference time and the end time of the current monitoring period, in minutes.

[0052] The user can determine the preset replacement time interval value according to the actual application scenario. The greater the user's need for improved accuracy in protecting vehicle location privacy, the smaller the preset replacement time interval value should be. One preset replacement time interval value is provided, which is 10 minutes.

[0053] Specifically, for vehicles whose replacement time interval value is greater than or equal to the preset replacement time interval value, a secondary determination is made based on the quantity comparison reference value and the communication characterization value.

[0054] In the secondary judgment, for vehicles whose quantity comparison reference value is less than the preset quantity comparison reference value and whose communication characterization value is less than the preset communication characterization value, a determination is made on whether to change the pseudonym based on the cumulative danger value.

[0055] For vehicles whose quantity comparison reference value is greater than or equal to the preset quantity comparison reference value or whose communication characterization value is greater than or equal to the preset communication characterization value, a determination is made on whether to change the pseudonym based on the hazard assessment value.

[0056] Specifically, the reference value for the number comparison of a single vehicle is the average value of the number comparison values ​​at each time point in the current monitoring period;

[0057] In this invention, the time point is set by the user. A method for setting the time point is provided, which takes the start time of a single monitoring cycle as the start point and sets an interval point every 30 seconds. The start point and each interval point are recorded as time points.

[0058] The quantity comparison value of a single vehicle at a single point in time is the ratio of the standard quantity to the number of vehicles appearing in the reference range corresponding to that vehicle at that point in time; the standard quantity is 12 vehicles, and the reference range corresponding to a single vehicle at a single point in time is a circle with the position of that vehicle at that point in time as the center and a preset length as the radius, the preset length being 50m;

[0059] The communication characteristic value for a single vehicle = number of communications / communication number threshold × number of communications weighting coefficient + number of communication nodes / number of communication nodes threshold × number of communication nodes weighting coefficient. The number of communications is the total number of times the vehicle broadcasts the reference information corresponding to the vehicle to other nodes in the current monitoring period. The number of communication nodes is the number of reference nodes that receive the reference information broadcast by the vehicle in the current monitoring period. The threshold for the number of communications is 10 times, the threshold for the number of communication nodes is 16, the number of communications weighting coefficient is 0.6, and the number of communication nodes weighting coefficient is 0.4.

[0060] Reference information includes the vehicle's location, speed, and pseudonym; reference nodes include, but are not limited to, other vehicles and RSUs.

[0061] Users can determine the preset quantity comparison reference value and preset communication characterization value according to the actual application scenario. The greater the user's requirement for the accuracy of vehicle location privacy protection, the smaller the preset quantity comparison reference value and preset communication characterization value should be. A method for determining the preset quantity comparison reference value and preset communication characterization value is provided. The historical records of whether pseudonym replacement is performed based on the risk assessment value are detected. The average value of the vehicle quantity comparison reference value and the average value of the communication characterization value corresponding to the historical records that meet the user's needs are respectively recorded as the preset quantity comparison reference value and preset communication characterization value.

[0062] In determining whether to change a pseudonym based on the accumulated hazard value, pseudonyms are changed for vehicles with an accumulated hazard value greater than or equal to a preset accumulated hazard value, while pseudonyms are not changed for vehicles with an accumulated hazard value less than the preset accumulated hazard value.

[0063] In determining whether to change a pseudonym based on the hazard assessment value, a pseudonym is changed for vehicles with a hazard assessment value greater than or equal to the preset hazard assessment value, while no pseudonym is required for vehicles with a hazard assessment value less than the preset hazard assessment value.

[0064] The cumulative risk value for a single vehicle is the sum of the risk assessment values ​​for each reference period. The monitoring periods between the monitoring period where the reference time for that vehicle is located and the current monitoring period, as well as the current monitoring period, are recorded as the reference periods.

[0065] The risk assessment value for a single vehicle = (quantity comparison reference value / preset quantity comparison reference value) × first weighting coefficient + (communication characterization value / preset communication characterization value) × second weighting coefficient, where both the first and second weighting coefficients are 0.5.

[0066] Users can determine the values ​​of the preset hazard accumulation value and the preset hazard assessment value according to the actual application scenario. The greater the user's requirement for the accuracy of vehicle location privacy protection, the smaller the preset hazard accumulation value and the preset hazard assessment value will be. One preset hazard accumulation value and preset hazard assessment value are provided: the preset hazard accumulation value is 1.7 and the preset hazard assessment value is 0.6.

[0067] Understandably, the replacement time interval effectively reflects the length of time since the last alias change: when the replacement time interval is less than the preset replacement time interval, it means that the last alias change was recent and the current alias usage time has not exceeded the safe range. The risk of location privacy leakage due to long-term alias use is low, so there is no need to change the alias.

[0068] When the replacement time interval is greater than or equal to the preset replacement time interval, it indicates that the vehicle's last pseudonym replacement was a long time ago, and the current pseudonym usage duration is close to or exceeds the safety threshold. Relying solely on the time dimension is insufficient to eliminate the risk of privacy leakage, and further judgment is needed based on the vehicle's actual operating status. Therefore, a secondary judgment is made based on the quantity comparison reference value and the communication characterization value. In the secondary judgment, when the quantity comparison reference value is less than the preset quantity comparison reference value and the communication characterization value is less than the preset communication characterization value, it indicates that the vehicle's current location has a high vehicle density, and the communication status is within the normal range, which is a low real-time risk state. However, it is necessary to pay attention to the risk accumulation effect of historical cycles to avoid a single low risk accumulating into a high risk. Therefore, a judgment is made on whether to replace the pseudonym based on the danger accumulation value. When the quantity comparison reference value is greater than or equal to the preset quantity comparison reference value or the communication characterization value is greater than or equal to the preset communication characterization value, it indicates that the vehicle's current location has an excessively low vehicle density, which may pose a risk of privacy tracking, or the communication status may be abnormal, potentially increasing the probability of information leakage. This is a medium-to-high real-time risk state, and priority should be given to responding to the current risk to prevent its expansion. Therefore, a judgment is made on whether to replace the pseudonym based on the danger assessment value.

[0069] Specifically, the target region types include:

[0070] A target area whose vehicle distribution reference value is greater than or equal to the preset vehicle distribution reference value and whose replacement allowance is greater than or equal to the preset replacement allowance;

[0071] The second type of target area is where the reference value for the distribution of vehicles to be replaced is less than the preset reference value for the distribution of vehicles to be replaced or the replacement allowance is less than the preset replacement allowance.

[0072] Specifically, the circumcircle of the target area is denoted as the reference circle, and the reference circle is divided into m sectors along 360°. The greater the calculation accuracy of the user's reference value for the distribution of vehicles to be replaced, the larger the value of m. One possible value for m is 20. The reference value for the distribution of vehicles to be replaced = the number of sector areas containing vehicles to be replaced in the target area / the total number of sector areas × the first coefficient + the area of ​​the smallest circle that can contain all vehicles to be replaced in the target area / the area of ​​the target area × the second coefficient. Both the first and second coefficients are 0.5. It should be noted that when determining the sector areas containing vehicles to be replaced, if there is a vehicle to be replaced at the center of the circle, the vehicle to be replaced at the center of the circle is not included in any sector area.

[0073] Replacement allowance = (number of vehicles to be replaced / average number of vehicles to be replaced corresponding to the historical records that can meet user needs) × third weighting coefficient + (1 - communication vehicle number threshold / average number of communication vehicle number thresholds corresponding to the historical records that can meet user needs) × fourth weighting coefficient, where the third weighting coefficient is 0.6 and the fourth weighting coefficient is 0.4.

[0074] The threshold for the number of communication vehicles is the sum of the number of communication vehicles corresponding to each vehicle to be replaced in the target area. The number of communication vehicles corresponding to a single vehicle to be replaced is the number of reference vehicles that receive the reference information broadcast by the vehicle to be replaced in the current monitoring period. The reference vehicles corresponding to a single vehicle to be replaced are other vehicles in the target area besides the vehicle to be replaced.

[0075] The user can determine the preset values ​​of the vehicle distribution reference value to be replaced and the preset replacement allowance based on the actual application scenario. The higher the user's requirements for the accuracy and security of vehicle privacy protection, the larger the preset values ​​of the vehicle distribution reference value to be replaced and the preset replacement allowance. A method for determining the preset values ​​of the vehicle distribution reference value to be replaced and the preset replacement allowance is provided. The average value of the vehicle distribution reference value to be replaced and the average value of the replacement allowance corresponding to each type of target area in the historical record that can meet the user's needs are detected and recorded as the preset vehicle distribution reference value to be replaced and the preset replacement allowance, respectively.

[0076] Understandably, the reference value for the distribution of vehicles to be replaced effectively reflects the rationality of the spatial distribution of vehicles to be replaced within the target area. The higher the value, the more uniform the vehicle distribution and the more sufficient the coverage, and the stronger the spatial obfuscation capability of the anonymity set. The substitution allowance effectively reflects the controllability of the risk of pseudonym replacement within the target area. The substitution allowance is determined by the threshold of the number of vehicles to be replaced and the number of communication vehicles. The threshold of the number of vehicles to be replaced and the number of communication vehicles effectively reflects the identity obfuscation basis and the intensity of the associated risk of pseudonym replacement within the target area. When the number of vehicles to be replaced in the target area is small, it indicates that the size of the anonymity set in the target area is insufficient, and it is easy for external observers to identify the identity through "minority comparison". When the threshold of the number of communication vehicles is large, it indicates that the topology of the connection between vehicles in the target area is dense and the risk of eavesdropping is high. Direct pseudonym replacement will lead to a sharp increase in the associated risk, and external attackers can easily match the old and new identities one by one.

[0077] Specifically, for vehicles to be replaced in the first type of target area, pseudonyms are directly changed; for vehicles to be replaced in the second type of target area, combined analysis is performed.

[0078] Specifically, when directly changing the pseudonym for a vehicle to be replaced in a target area, each vehicle to be replaced in the target area requests a new pseudonym from the certificate issuing authority for the pseudonym change, or selects a new pseudonym from its own pseudonym pool and notifies the certificate issuing authority to retrieve the expired pseudonym. This is a common technical method used by those skilled in the art, and will not be elaborated on in detail.

[0079] Specifically, for vehicles to be replaced in the second type of target area where the vehicle analysis coefficient is less than the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the position synchronization degree.

[0080] For vehicles to be replaced in the second type of target area where the vehicle analysis coefficient is greater than or equal to the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the interactive reference value.

[0081] The vehicle analysis coefficients are determined based on vehicle density and congestion coefficients.

[0082] Specifically, the vehicle analysis coefficient = vehicle density / average vehicle density corresponding to historical records that can meet user needs × density weight coefficient + congestion coefficient / average congestion coefficient corresponding to historical records that can meet user needs × congestion weight coefficient, where both density weight coefficient and congestion weight coefficient are 0.5.

[0083] Vehicle density is the ratio of the number of vehicles in the target area to the area of ​​the target area. Congestion coefficient is the ratio of the speed threshold to the average speed of each vehicle in the target area at the end of the current monitoring period. The speed threshold is 80 km / h.

[0084] The user can determine the value of the preset vehicle analysis coefficient according to the actual application scenario. The smaller the value of the preset vehicle analysis coefficient, the greater the user's need to determine the associated vehicle combination based on the interactive reference value. One preset vehicle analysis coefficient value is provided, which is 0.6.

[0085] For each vehicle to be replaced in the target area, a combination analysis is performed. When performing a combination analysis on a single vehicle to be replaced, the vehicle to be replaced is recorded as the target vehicle to be replaced. Other vehicles to be replaced in the target area that are not recorded in the associated vehicle combination are recorded as reference vehicles to be replaced. The target vehicle to be replaced and the reference vehicles to be replaced that meet the preset conditions are recorded into an associated vehicle combination. The combination analysis continues for vehicles to be replaced that are not recorded into an associated vehicle combination until all vehicles to be replaced are recorded into an associated vehicle combination, at which point the combination analysis stops.

[0086] When determining the associated vehicle combination based on the position synchronization degree, the preset condition is that the position synchronization degree with the target vehicle to be replaced is greater than the preset position synchronization degree.

[0087] When determining the associated vehicle combination based on the interaction reference value, the preset condition is that the interaction reference value with the target vehicle to be replaced is greater than the preset interaction reference value.

[0088] For any two vehicles to be replaced, the position synchronization degree is the ratio of the distance threshold to the average distance. The average distance is the average of the distance reference values ​​corresponding to each time point in the current monitoring period. The distance threshold is 10m. The distance reference value corresponding to a single time point is the distance between the two vehicles to be replaced at that time point.

[0089] For any two vehicles, denoted as vehicle 1 and vehicle 2 respectively, the number of times vehicle 1 broadcasts the reference information corresponding to vehicle 1 to vehicle 2 and the number of times vehicle 2 broadcasts the reference information corresponding to vehicle 2 to vehicle 1 are recorded as the number of interactions. The interaction coefficient = number of interactions / sum of the number of communications corresponding to the two vehicles to be replaced.

[0090] Communication frequency similarity = 1 - the absolute value of the difference in the number of communications between the two vehicles to be replaced / the larger value of the number of communications between the two vehicles to be replaced; Interaction reference value = communication frequency similarity / communication frequency similarity threshold × frequency weight coefficient + (1 - interaction coefficient / interaction coefficient threshold) × interaction weight coefficient, where the frequency weight coefficient is 0.7 and the interaction weight coefficient is 0.3; the communication frequency similarity threshold is 0.65 and the interaction coefficient threshold is 0.37;

[0091] Understandably, communication frequency similarity can accurately reflect the similarity of the number of communications between vehicles, while the interaction coefficient can effectively measure the eavesdropping risk caused by information exchange between vehicles. Based on the interaction reference value constructed by the two, vehicles with "consistent communication rhythm and low eavesdropping risk" can be accurately screened and grouped into the same association group, ultimately blocking the identity mapping of attackers from two core dimensions: First, the screening based on communication frequency similarity can ensure that the broadcast intervals and periodic patterns of vehicles within the group are highly synchronized, avoiding the formation of individual characteristic markers due to "differences in communication rhythm". Second, the screening based on the interaction coefficient can strictly control the intensity of information exchange between vehicles within the group, ensuring that although the vehicles have synchronized communication rhythms, the frequency at which they receive each other's broadcasts is low and the correlation is weak, and no "strong interactive binding relationship" is formed.

[0092] Users can determine the preset position synchronization degree and preset interaction reference value according to the actual application scenario. The higher the user's requirements for the accuracy and security of vehicle privacy protection, the larger the preset position synchronization degree and preset interaction reference value will be. One preset position synchronization degree and preset interaction reference value are provided: preset position synchronization degree is 0.72 and preset interaction reference value is 0.59.

[0093] Understandably, vehicle analysis coefficients effectively reflect the combined impact of vehicle density and traffic conditions within the target area on the risk of pseudonym replacement association. When the vehicle analysis coefficient is less than the preset coefficient, it indicates that the vehicle density in the target area is low and the congestion is mild. The spatial relationship between vehicles has a more significant impact on privacy protection. In this case, the synchronization of vehicle trajectories is the core of constructing an effective anonymity set. Combinations of vehicles with synchronized locations can enhance spatial obfuscation capabilities. Therefore, combinations of associated vehicles are determined based on location synchronization. When the vehicle analysis coefficient is greater than or equal to the preset coefficient, it indicates that the vehicle density in the target area is high and the congestion is severe. The communication frequency between vehicles is higher and the association is closer. The communication characteristics have a more prominent impact on privacy protection. In this case, combinations should be selected based on the consistency of communication rhythm and low risk of eavesdropping to avoid identity association due to dense interactions. Therefore, combinations of associated vehicles are determined based on interaction reference values.

[0094] Specifically, the replacement priority coefficient for each associated vehicle combination is determined based on the number of vehicles in the combination.

[0095] The replacement priority coefficient for a single associated vehicle combination is positively correlated with the number of vehicles in that combination.

[0096] Specifically, the replacement priority coefficient for a single associated vehicle combination is equal to the number of vehicles in that combination multiplied by the replacement coefficient, where the replacement coefficient is 1. The number of vehicles in a single associated vehicle combination is the total number of vehicles to be replaced in that combination.

[0097] When changing pseudonyms for each associated vehicle combination in a single Class II target area, the higher the priority coefficient of the associated vehicle combination, the higher the priority of the pseudonym change. When changing the name of a single associated vehicle combination, pseudonyms are changed for each vehicle to be changed in the associated vehicle combination and each virtual vehicle corresponding to the associated vehicle combination at the same time. Each vehicle requests a new pseudonym from the certificate authority for pseudonym change, or selects a new pseudonym from its own pseudonym pool, and notifies the certificate authority to retrieve the expired pseudonym.

[0098] Specifically, the number of reference vehicles is determined based on the combined characterization value of the associated vehicle combination;

[0099] The number of reference vehicles corresponding to a single associated vehicle combination is positively correlated with the combination characteristic value corresponding to that associated vehicle combination.

[0100] Specifically, the combined representation value = (1 - number of combined vehicles / average number of combined vehicles corresponding to each associated vehicle combination in the historical records that can meet user needs) × quantity weight coefficient + (switching time reference value / average switching time reference value corresponding to each associated vehicle combination in the historical records that can meet user needs) × time weight coefficient, with the quantity weight coefficient being 0.4 and the time weight coefficient being 0.6.

[0101] The reference value for the switching time of a single associated vehicle combination is the minimum value of the switching time of each vehicle to be replaced in that associated vehicle combination; the switching time of a single vehicle to be replaced is the ratio of the edge length to the speed of the vehicle at the end of the current monitoring cycle. A straight line is drawn along the direction of the vehicle's movement, with the starting point being the location of the vehicle at the end of the current monitoring cycle. The intersection of this straight line with the edge of the target area is the ending point, and the length of the line connecting the starting point and the ending point is the edge length.

[0102] It is understandable that the number of combined vehicles reflects the privacy obfuscation capability of the associated combination itself, and the switching time reference value reflects the urgency of the timeliness of the virtual vehicle demand of the associated vehicle combination. Therefore, based on the combination characteristic value determined by the two, the number of virtual vehicles can be accurately matched, which maximizes the privacy obfuscation effect and adapts to the efficiency requirements of dynamic collaboration.

[0103] The number of reference vehicles corresponding to a single associated vehicle combination is n, where n is the smallest integer greater than or equal to n0, and n0 is the product of the combination representation value corresponding to the associated vehicle combination and the quantity threshold, which is 12 vehicles.

[0104] Specifically, for associated vehicle combinations whose quantity difference is greater than or equal to a preset quantity difference, virtual vehicles are sequentially set;

[0105] For associated vehicle combinations where the quantity difference is less than a preset quantity difference, the system determines whether to increase the number of virtual vehicles based on the combination area comparison value, and sets the number of virtual vehicles based on the historical attack frequency of the sub-region.

[0106] Specifically, in the sequential setting of virtual vehicles, a virtual vehicle is set at any position within each reference radius where no vehicle to be replaced exists.

[0107] The quantity difference corresponding to a single associated vehicle combination is the difference between the reference vehicle quantity and the reference actual vehicle quantity;

[0108] A vehicle to be replaced is randomly selected. The radius of the line connecting the center of the circumcircle of the target area and the selected vehicle to be replaced is used as the starting radius. This starting radius is then rotated clockwise at a preset angle to obtain several rotating radii. The starting radius and the rotating radii are recorded as reference radii. The number of reference radii containing a vehicle to be replaced is recorded as the reference number of actual vehicles. It should be noted that if a vehicle to be replaced is located exactly at the center of the circumcircle of the target area, that vehicle is not included in the above statistics. The preset angle is the ratio of 360° to the number of reference vehicles.

[0109] The user can determine the value of the preset quantity difference according to the actual application scenario. The smaller the preset quantity difference value, the greater the user's need to set the virtual vehicles in sequence. One preset quantity difference value is provided, with a preset quantity difference of 6.

[0110] The combined area comparison value is the ratio of the area of ​​the smallest circle that can contain each vehicle to be replaced in the associated combination to the area of ​​the circumscribed circle of the target area.

[0111] When setting up virtual vehicles based on the historical attack frequency of a sub-region, sub-regions are selected in descending order of historical attack frequency until the number of selected sub-regions equals the number of virtual vehicles. A virtual vehicle is then set up at any location in each selected sub-region.

[0112] Divide the outer circle of the target area where the associated vehicle combination is located into n sector regions, each sector region being a sub-region, where n equals the number of reference vehicles;

[0113] The historical attack frequency corresponding to a single sub-region is the number of historical monitoring periods with abnormal requests. The historical monitoring period is each monitoring period before the current monitoring period. An abnormal request is when a suspicious vehicle initiates a query for any vehicle pseudonym in the sub-region during the current monitoring period and the number of queries is greater than 2. A suspicious vehicle is a vehicle marked as malicious by RSU.

[0114] Understandably, when the quantity difference is greater than or equal to the preset quantity difference, it indicates that there are many reference radii without vehicles to be replaced, resulting in a large gap in the anonymity set size. This directly affects the basic effectiveness of privacy protection and makes the system susceptible to accurate identification due to insufficient quantity. Therefore, virtual vehicles are set sequentially, and the gap is quickly filled by supplementing according to the reference radius to ensure that the anonymity set size meets the target. When the quantity difference is less than the preset quantity difference, it indicates that there are few reference radii without vehicles to be replaced, and the basic size of the anonymity set is close to the target. However, the deployment quality of virtual vehicles needs to be further optimized. Therefore, the number of virtual vehicles is adjusted based on the combined area comparison value. The combined area comparison value effectively reflects the spatial clustering degree of vehicles to be replaced within the associated vehicle combination, thereby determining whether it is necessary to increase the number of virtual vehicles to disperse clustering risks. Virtual vehicles are also set based on the historical attack frequency of sub-regions, and virtual vehicles are deployed in high-risk areas to improve the vehicle privacy protection effect.

[0115] Specifically, for associated vehicle combinations whose combined area comparison value is less than the preset combined area comparison value, the number of virtual vehicles is increased based on the comparison deviation value.

[0116] For a group of associated vehicles whose comparison deviation value is greater than or equal to a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a first preset ratio.

[0117] For a group of associated vehicles whose comparison deviation value is less than a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a second preset ratio.

[0118] The first preset ratio is less than the second preset ratio.

[0119] Specifically, for associated vehicle combinations whose combined area comparison value is greater than or equal to the preset combined area comparison value, there is no need to increase the number of virtual vehicles; the number of virtual vehicles is directly set to the quantity difference.

[0120] Users can determine the preset combined area comparison value and preset comparison deviation value according to the actual application scenario. The higher the user's requirement for improving the accuracy and security of vehicle privacy protection, the larger the preset combined area comparison value and preset comparison deviation value will be. One preset combined area comparison value and preset comparison deviation value are provided: preset combined area comparison value is 0.56 and preset comparison deviation value is 0.5.

[0121] The comparison deviation value is the ratio of the number of reference real vehicles to the number of reference vehicles;

[0122] When determining the increase in the number of virtual vehicles based on the first preset ratio, the increase in the number of virtual vehicles = the first preset ratio × (the number of reference vehicles - the difference in number); when determining the increase in the number of virtual vehicles based on the second preset ratio, the increase in the number of virtual vehicles = the second preset ratio × (the number of reference vehicles - the difference in number); the first preset ratio is 0.4, and the second preset ratio is 0.7.

[0123] The smallest integer greater than or equal to the increase value of virtual vehicles is denoted as a0, and the number of virtual vehicles = a0 + the difference in number.

[0124] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for changing vehicle pseudonyms based on artificial intelligence, characterized in that, include: The determination of whether pseudonym replacement is unnecessary is based on the replacement time interval value, or a secondary determination is made based on the quantity comparison reference value and the communication characterization value. The target area type is determined based on the reference value of the distribution of vehicles to be replaced and the replacement allowance. Based on the target area type, it is determined whether to directly replace the pseudonyms or to perform a combination analysis. In the combined analysis, the associated vehicle combinations are determined based on the vehicle analysis coefficient, either based on the location synchronization degree or based on the interaction reference value. The replacement priority coefficient corresponding to each associated vehicle combination is determined based on the number of combined vehicles. The virtual vehicles are set sequentially based on the quantity difference or set based on the historical attack frequency of the sub-region. The pseudonyms of the vehicles to be replaced and the virtual vehicles in the associated vehicle combinations are changed. The reference vehicle count is determined based on the combined representation value of the associated vehicle combination, and the count difference is determined based on the difference between the reference vehicle count and the reference actual vehicle count.

2. The vehicle pseudonym replacement method based on artificial intelligence according to claim 1, characterized in that, For vehicles whose replacement time interval is less than the preset replacement time interval, there is no need to change the pseudonym.

3. The vehicle pseudonym replacement method based on artificial intelligence according to claim 2, characterized in that, For vehicles whose replacement time interval value is greater than or equal to the preset replacement time interval value, a secondary determination is made based on the quantity comparison reference value and the communication characterization value. In the secondary judgment, for vehicles whose quantity comparison reference value is less than the preset quantity comparison reference value and whose communication characterization value is less than the preset communication characterization value, a determination is made on whether to change the pseudonym based on the cumulative danger value. For vehicles whose quantity comparison reference value is greater than or equal to the preset quantity comparison reference value or whose communication characterization value is greater than or equal to the preset communication characterization value, a determination is made on whether to change the pseudonym based on the hazard assessment value.

4. The vehicle pseudonym replacement method based on artificial intelligence according to claim 1, characterized in that, The target region types include: A target area whose vehicle distribution reference value is greater than or equal to the preset vehicle distribution reference value and whose replacement allowance is greater than or equal to the preset replacement allowance; The second type of target area is where the reference value for the distribution of vehicles to be replaced is less than the preset reference value for the distribution of vehicles to be replaced or the replacement allowance is less than the preset replacement allowance.

5. The vehicle pseudonym replacement method based on artificial intelligence according to claim 4, characterized in that, For vehicles to be replaced in the first type of target area, pseudonyms are directly replaced; for vehicles to be replaced in the second type of target area, combined analysis is performed.

6. The method for changing vehicle pseudonyms based on artificial intelligence according to claim 1, characterized in that, For vehicles to be replaced in Class II target areas where the vehicle analysis coefficient is less than the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the position synchronization degree. For vehicles to be replaced in the second type of target area where the vehicle analysis coefficient is greater than or equal to the preset vehicle analysis coefficient, the associated vehicle combination is determined based on the interactive reference value. The vehicle analysis coefficients are determined based on vehicle density and congestion coefficients.

7. The vehicle pseudonym replacement method based on artificial intelligence according to claim 6, characterized in that, The replacement priority coefficient for each associated vehicle combination is determined based on the number of vehicles in the combination. The replacement priority coefficient for a single associated vehicle combination is positively correlated with the number of vehicles in that combination.

8. The method for changing vehicle pseudonyms based on artificial intelligence according to claim 1, characterized in that, The number of reference vehicles is determined based on the combined characterization value of the associated vehicle groups; The number of reference vehicles corresponding to a single associated vehicle combination is positively correlated with the combination characteristic value corresponding to that associated vehicle combination.

9. The method for changing vehicle pseudonyms based on artificial intelligence according to claim 1, characterized in that, For associated vehicle combinations where the quantity difference is greater than or equal to a preset quantity difference, virtual vehicles are sequentially set; For associated vehicle combinations where the quantity difference is less than a preset quantity difference, the system determines whether to increase the number of virtual vehicles based on the combination area comparison value, and sets the number of virtual vehicles based on the historical attack frequency of the sub-region.

10. The vehicle pseudonym replacement method based on artificial intelligence according to claim 9, characterized in that, For associated vehicle combinations whose combined area comparison value is less than the preset combined area comparison value, the number of virtual vehicles is increased based on the comparison deviation value. For a group of associated vehicles whose comparison deviation value is greater than or equal to a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a first preset ratio. For a group of associated vehicles whose comparison deviation value is less than a preset comparison deviation value, the increase in the number of virtual vehicles is determined based on a second preset ratio. The first preset ratio is less than the second preset ratio.

Images