In-vehicle privacy data management method
Patent Information
- Application Number
- CN202611309556.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-27
- Publication Date
- 2026-09-25
AI Technical Summary
[0004]本申请实施例的目的在于提供一种车载隐私数据管理方法,用以解决目前车载隐私数据管理依赖静态权限控制机制存在的隐私安全性低的问题
[0039]上述说明仅是本申请技术方案的概述,为了能够更清楚了解本申请的技术手段,而可依照说明书的内容予以实施,并且为了让本申请的上述和其它目的、特征和优点能够更明显易懂,以下特举本申请的具体实施方式。
Smart Images

Figure CN122824522A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle privacy and security management technology, and more specifically, to a method for managing vehicle privacy data. Background Technology
[0002] With the rapid development of intelligent connected vehicles, the amount of data collected and processed by in-vehicle systems every day has increased dramatically, covering various types such as vehicle location, speed, driving behavior, in-vehicle audio and video signals, user preferences, and application logs. A large amount of this data involves personal information and even sensitive data.
[0003] Currently, in-vehicle privacy data management primarily relies on static permission control mechanisms. For example, some vehicles use operating system-level permission models (such as declarative permissions in Android Automotive) to authorize applications to access sensitive data. However, these methods typically grant authorization when a user first installs or launches an application, and the permissions remain valid indefinitely, unable to be dynamically adjusted based on the vehicle's actual operating status. For instance, while driving, some third-party entertainment applications may continue to read high-precision location data or user preference data for personalized recommendations, increasing the risk of privacy breaches and potentially distracting the driver, posing safety hazards. Summary of the Invention
[0004] The purpose of this application is to provide a method for managing in-vehicle privacy data, in order to solve the problem of low privacy security that exists in current in-vehicle privacy data management methods that rely on static permission control mechanisms.
[0005] Firstly, this application provides a method for managing in-vehicle privacy data. The method includes: acquiring vehicle data stream information, wherein the vehicle data stream information includes multiple data streams, each data stream being obtained by attaching metadata tags to the monitored raw data stream to identify the data source and attributes; identifying the current operating scenario of the vehicle based on its current operating status; conducting a compliance assessment on each data stream in the vehicle data stream information based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, obtaining a compliance ruling result corresponding to each data stream; and performing corresponding data access control operations on the corresponding data stream and adjusting the data access permissions of the application to which the corresponding data stream belongs based on the compliance ruling result corresponding to each data stream.
[0006] The aforementioned in-vehicle privacy data management method captures the raw data stream and adds metadata tags identifying the source and attributes to generate complete vehicle data stream information. It then combines multi-dimensional parameters such as vehicle speed, gear position, battery level, and user identification to identify real-time vehicle operating scenarios. Multiple sets of locally pre-built compliance rules are matched to complete data stream compliance assessments one by one and generate exclusive adjudication results. Finally, data stream interception / release operations and application permission dynamic adjustments are executed simultaneously. The entire process runs locally offline without real-time cloud intervention. It relies on the underlying signed data stream to achieve fine-grained control of individual privacy data, abandons the static permanent authorization mode based on real-time scenario recognition, and incorporates a closed-loop logic of parallel assessment of multiple regulations and adjudication-linked permission adjustments. This allows for first achieving unobstructed interception and source tracing of the underlying sensor data stream, and then switching privacy control strategies in real-time according to vehicle operating conditions. It simultaneously considers multiple domestic and international compliance constraints, significantly reducing privacy leaks and driver distraction safety hazards during driving. Furthermore, the metadata uses immutable metadata tags, multiple regulations and rules are persistently stored locally, the assessment response speed remains at the millisecond level, and full-process privacy protection can be continuously completed even in offline environments.
[0007] In an optional implementation of the first aspect, based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, a compliance assessment is performed on each data stream in the vehicle data stream information to obtain a compliance ruling result corresponding to each data stream. This includes: based on the current operating scenario of the vehicle, selecting multiple target data compliance subsets from multiple preset data compliance rule sets, wherein each preset data compliance rule set includes multiple data compliance subsets, and each data compliance subset corresponds to a vehicle operating scenario; for each data stream, a compliance assessment is performed on the data stream based on the multiple target data compliance subsets to obtain independent ruling opinions corresponding to each target data compliance subset; and the multiple independent ruling opinions are integrated and adjudicated to generate a compliance ruling result corresponding to the data stream.
[0008] The above implementation method first filters irrelevant rules and selects multiple sets of target data compliance subsets that match the identified vehicle operation scenarios. Then, it sends a single data stream to all target subsets for parallel verification to generate multiple independent adjudication opinions. Finally, it integrates multiple opinions to eliminate rule conflicts and outputs a unique adjudication result. This compliance judgment logic of first filtering subsets, then conducting multi-dimensional parallel evaluation, and finally unifying arbitration can first filter invalid rules to reduce vehicle computing power consumption, then conduct multi-dimensional synchronous verification of data stream compliance risks from multiple data compliance rule sets, and finally unify the constraints and contradictions between multiple regulations. This can first improve the real-time evaluation efficiency of vehicle terminals, then comprehensively cover various privacy compliance risk points, and at the same time avoid the failure of control logic caused by conflicting instructions output by multiple rules in parallel.
[0009] In an optional implementation of the first aspect, multiple independent rulings are integrated and adjudicated to generate a compliance ruling result corresponding to the data stream, including: merging and arbitrating the independent rulings corresponding to each target data compliance subset according to the strictest matching principle to generate a compliance ruling result corresponding to the data stream, wherein the strictest matching principle means taking the most stringent value of each target data compliance subset for the same field, and the principle that any target data compliance subset that requires interception should be immediately intercepted.
[0010] In the above implementation method, after obtaining the independent ruling opinions generated by each target data compliance subset corresponding to a data stream, the merged arbitration operation is carried out strictly in accordance with the strictest matching principle to generate the compliance ruling result corresponding to the data stream. The entire operation follows the two layers of hard judgment criteria contained in the strictest matching principle. The first layer iterates through all independent ruling opinions for the same control field such as data retention period and cross-border transmission and selects the field value with the most stringent constraint in each target data compliance subset as the standard value. The second layer sets a risk priority fallback logic. As long as any independent ruling opinion raises a request to intercept the data stream, the final merged arbitration output result is directly locked as interception, thereby realizing merged arbitration.
[0011] In an optional implementation of the first aspect, the strictest matching principle includes: for the data retention period field, selecting the shortest period required by each target data compliance subset as the final retention period; for the cross-border data transfer field, when the data destination is a specific region and involves personal information, enforcing the outbound security assessment mechanism stipulated by the target region, and intercepting the data flow if the outbound security assessment fails.
[0012] In the above implementation method, this solution adopts the strictest matching principle to complete the merged arbitration when integrating multiple independent arbitration opinions. During the calculation process, the standard with the highest constraint strength within all subsets is uniformly selected for the same control fields such as data retention, cross-border, and authorization. At the same time, the highest priority is set for the interception command. As long as any subset outputs a prohibition on transmission opinion, the data stream is directly blocked. This can maximize the tightening of the control boundary of vehicle privacy data, simultaneously meet the mandatory constraints of multiple countries' regulations, unify the arbitration logic of vehicle models in multiple regions, and eliminate the need to modify the core arbitration procedure separately for different markets.
[0013] In an optional implementation of the first aspect, the independent rulings corresponding to each target data compliance subset are merged and arbitrated according to the strictest matching principle to generate a compliance ruling result for the data stream, including: if the independent ruling for any target data compliance subset is to prohibit transmission, then the final ruling result for the data stream is to block; if none of the target data compliance subsets require prohibition of transmission, but at least one target data compliance subset has an independent ruling that requires consent, then the final ruling result for the data stream is to allow transmission but require user confirmation; if all target data compliance subsets allow transmission without additional conditions, then the final ruling result is to directly allow transmission; if the data stream matches the preset low-risk rules in all target data compliance subsets, then the final ruling result is to only log without blocking.
[0014] The above implementation method sets up a fixed four-level progressive judgment process from high risk to low risk. Priority is given to directly blocking the transmission of prohibited opinions. If there is no blocking instruction but authorization is required, the user confirmation mode is determined. If there are no constraints from any rules, the transmission is allowed directly. If only low gray-scale rules are matched, only logs are retained. This layered and progressive judgment logic sets different handling paths for different risk data streams, thereby achieving immediate blocking of high-risk data, returning decision-making power to users for medium-risk data, allowing direct passage for no-risk data, and tolerating minor flaws while leaving traces. This balances privacy and security, regulatory compliance, and the user experience of in-vehicle applications.
[0015] In an optional implementation of the first aspect, based on the compliance ruling result corresponding to each data stream, a corresponding data access control operation is performed on the corresponding data stream, including: if the compliance ruling result is that transmission is allowed but user confirmation is required, controlling the in-vehicle human-machine interface to display data usage details to the user and waiting for an instruction response; if a user-selected instruction to allow transmission is received, then an allow transmission operation is performed on the corresponding data stream; if a user-selected instruction to deny transmission is received, then an interception and blocking operation is performed on the corresponding data stream.
[0016] The above implementation method, when the compliance ruling is that transmission is allowed but user confirmation is required, firstly displays the complete data usage details to the user through the human-machine interface and continuously waits for the user's instruction response, and then executes the release or blocking operation according to the user's input instruction to allow transmission or refuse transmission, respectively. This solves the loophole of non-compliance of traditional in-vehicle silent authorization. The human-machine interface combined with reminder display can be adapted to different driving and parking scenarios without interfering with driving. The complete user instruction response record can be directly used as regulatory compliance audit evidence. At the same time, the decision-making power for the handling of medium-risk data streams is completely handed over to the user, balancing privacy and security with the user's right to choose.
[0017] In an optional implementation of the first aspect, based on the compliance ruling result corresponding to each data stream, corresponding data access control operations are performed on the corresponding data stream, including: if the compliance ruling result is to only log without blocking, writing the relevant events of the corresponding data stream into the violation log and performing a transmission permission operation on the corresponding data stream.
[0018] The above implementation method allows data flow and log recording to proceed simultaneously, distinguishing between minor compliance flaws and major privacy risks. It will not interrupt the normal business functions of the application due to low-risk rule violations. Based on this handling logic, the original samples of low-risk violation events can be fully retained while ensuring the complete usability of the in-vehicle application. The event data written into the violation log can provide a data source for event similarity, taking into account both the user's smoothness of use and the data reserve needs for subsequent rule iterations.
[0019] In an optional implementation of the first aspect, the data access permissions of the application to which the corresponding data stream belongs are adjusted according to the compliance ruling result corresponding to each data stream, including: if the compliance ruling result corresponding to the data stream is blocking, the application to which the data stream belongs revoked or downgraded the access permissions of the corresponding data type of the data stream; if the compliance ruling result corresponding to the data stream is allowing transmission but requires user confirmation, the corresponding permissions are not granted before the user selects an allow transmission instruction, and the granted permissions are revoked after the user selects a refuse transmission instruction.
[0020] The above implementation method adjusts the corresponding data permissions of the application in sync with the two types of adjudication results: interception of a single data stream and pending confirmation. In the interception scenario, permissions are revoked / gradually downgraded. In the pending confirmation scenario, permissions are temporarily locked before authorization and permanently revoked after user rejection. This logic of linking data stream violations with application permissions achieves precise and targeted punishment. As a result, only the sensitive data corresponding to the violation is restricted, without interfering with the basic functions of the application. The pre-locking permission mechanism prevents the application background from collecting private data without authorization from the source, forming a two-layer risk control closed loop of data stream interception and application permission tightening.
[0021] In an optional implementation of the first aspect, the method further includes: freezing an application's access to vehicle data if the application generates multiple compliance rulings that block or obstruct the application within a preset time period and the cumulative number of such rulings exceeds a threshold.
[0022] The above implementation method accumulates the number of times the application blocks compliance rulings within a preset time period. Once the accumulated number exceeds a preset threshold, the application's access to all vehicle data is directly frozen. This underlying punishment logic based on sliding time window frequency statistics and over-limit strong freezing can distinguish between occasional single-time blocking and continuous, habitual privacy violations by applications. It can accurately lock long-term high-risk applications and completely cut off their data collection source. While a single data stream blocking only blocks a single request, this step achieves full-time, full-domain permission blocking.
[0023] In an alternative implementation of the first aspect, the method further includes: adjusting the data access permissions of the application according to the current operating scenario of the vehicle.
[0024] In an optional implementation of the first aspect, the data access permissions of the application are adjusted according to the current operating scenario of the vehicle, including: reducing the access permission level of the third-party application to the preset data type when the vehicle operating scenario changes from a parking state to a driving state; and restoring the access permission of the third-party application to the preset data type when the vehicle operating scenario changes from a driving state back to a parking state.
[0025] The two implementation methods described above proactively reduce the access permissions of third-party applications to preset highly sensitive data types when the vehicle operation scenario switches from a parked state to a driving state, and automatically restore the corresponding access permissions when the scenario switches back to a parked state. This underlying logic of adjusting permissions in a two-way gradient between the two core operating conditions of driving and parking is specifically matched to the differentiated management needs of the high-safety driving scenario and the low-risk parking scenario. Relying on this two-way switching permission adjustment step, it is possible to proactively restrict third-party applications from reading preset sensitive data such as high-precision positioning and audio and video during vehicle operation, eliminating the hidden dangers of distraction and privacy leakage risks while driving.
[0026] In an optional implementation of the first aspect, the method further includes: collecting historical compliance data for each application; calculating a data compliance score for each application based on the historical compliance data; and controlling access permissions for the corresponding application based on the data compliance score.
[0027] The above implementation method continuously collects historical compliance data for each application, calculates the corresponding data compliance score based on the historical compliance data, and uses the quantitative score as a standard to implement differentiated access control for applications. This can objectively and in the long term evaluate the data collection compliance capabilities of third-party applications. Based on the score-based differentiated control, control interference can be reduced for high-scoring and trustworthy applications, while sensitive data permissions can be tightened for low-scoring and high-risk applications, thus effectively purifying the overall privacy ecosystem of in-vehicle third-party applications in the long term.
[0028] In an optional implementation of the first aspect, a data compliance score is calculated for each application based on its historical compliance data, including: determining the violation frequency, data minimization degree, user authorization transparency, and historical complaint rate for each application based on its historical compliance data; and performing a weighted calculation on the violation frequency, data minimization degree, user authorization transparency, and historical complaint rate for each application to obtain a data compliance score for each application.
[0029] The above implementation method extracts four independent quantitative indicators from historical compliance data: violation frequency, data minimization, authorization transparency, and user complaint rate. Customizable weights are configured to perform weighted summation to generate a data compliance score, avoiding the distortion of scores based on a single indicator. All indicators are automatically extracted from logs without manual input, significantly reducing the compliance operation and maintenance costs for car companies. The entire scoring process is traceable and adaptable to regulatory verification.
[0030] In an optional implementation of the first aspect, access control is performed on the corresponding application based on the data compliance score of each application, including: determining the score range of the data compliance score of each application based on the data compliance score of each application; and performing access control on the corresponding application based on the access control configuration corresponding to the score range of the data compliance score of each application, wherein the access control configurations corresponding to different score ranges are different.
[0031] The above implementation method first performs interval classification and determination on the calculated application data compliance score, locks the preset score range to which the score belongs, and then retrieves the exclusive access permission configuration rules pre-bound to the range to complete the application permission allocation. It can open full data access capabilities for high-scoring trusted applications, restrict some sensitive data for medium-scoring applications, and tighten full-domain collection permissions for low-scoring applications, so as to achieve refined, hierarchical and flexible management and control.
[0032] In an optional implementation of the first aspect, the method further includes: statistically analyzing violation events within a preset time period; performing cluster analysis on the violation events within the preset time period to identify frequently occurring target violation events; converting the identified target violation events into a custom data compliance candidate set; transmitting the custom data compliance candidate set to the cloud for review; receiving a rule update package approved by the cloud, wherein the rule update package is generated based on the content of the custom data compliance candidate set; and synchronizing the update rule corresponding to the rule update package to multiple data compliance rule sets.
[0033] The above implementation method periodically summarizes all vehicle violation logs, uses clustering algorithms to mine high-frequency common violations to generate custom rule candidates, uploads them to the cloud for multi-layer compliance verification, and after approval, receives the update package and synchronizes it to the local full set of compliance rules. This solves the problem of the lag in traditional manual rule updates, which takes several months. Automatically generating candidate sets reduces the workload of manual development, cloud verification ensures that new rules comply with regulations, and the rule package feedback enables all vehicles to continuously fill privacy protection vulnerabilities.
[0034] Secondly, this application also provides an in-vehicle privacy data management device, which includes: an acquisition module, an identification module, a compliance adjudication module, and a control adjustment module; the acquisition module is used to acquire vehicle data stream information, wherein the vehicle data stream information includes multiple data streams, each data stream is obtained by attaching metadata tags for identifying the data source and attributes to the monitored raw data stream; the identification module is used to identify the current operating scenario of the vehicle based on the current operating status of the vehicle; the compliance adjudication module is used to perform compliance assessment on each data stream in the vehicle data stream information based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, and obtain the compliance adjudication result corresponding to each data stream; the control adjustment module is used to perform corresponding data access control operations on the corresponding data stream and adjust the data access permissions of the application to which the corresponding data stream belongs based on the compliance adjudication result corresponding to each data stream.
[0035] The aforementioned in-vehicle privacy data management device captures the raw data stream and adds metadata tags identifying the source and attributes to generate complete vehicle data stream information. It combines multi-dimensional parameters such as vehicle speed, gear position, battery level, and user identification to identify real-time vehicle operating scenarios. It matches multiple sets of locally pre-built compliance rules to complete data stream compliance assessments one by one and generates a unique ruling. Finally, it synchronously executes data stream interception / release operations and dynamically adjusts application permissions. The entire process runs locally offline without real-time cloud intervention. It relies on the underlying signed data stream to achieve fine-grained control of individual privacy data, abandons the static permanent authorization mode based on real-time scenario recognition, and incorporates a closed-loop logic of parallel assessment of multiple regulations and linked adjudication for permission adjustments. This allows for first achieving unobstructed interception and source tracing of the underlying sensor data stream, and then switching privacy control strategies in real-time according to vehicle operating conditions. It simultaneously considers multiple domestic and international compliance constraints, significantly reducing privacy leaks and driver distraction safety hazards during driving. Furthermore, the metadata uses immutable metadata tags, multiple regulations and rules are persistently stored locally, the assessment response speed remains at the millisecond level, and it can continuously complete the entire process of privacy protection even in offline environments.
[0036] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the method described in the first aspect and any optional embodiment of the first aspect.
[0037] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the method described in the first aspect or any optional embodiment of the first aspect.
[0038] Fifthly, the present invention provides a computer program product, including a computer program / instructions, which, when executed by a processor, perform the method described in the first aspect and any optional embodiment of the first aspect.
[0039] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0040] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0041] Figure 1 A first flowchart of the in-vehicle privacy data management method provided in this application embodiment; Figure 2 This is a second flowchart of the in-vehicle privacy data management method provided in an embodiment of this application; Figure 3 A third flowchart illustrating the in-vehicle privacy data management method provided in this application embodiment; Figure 4 A fourth flowchart of the in-vehicle privacy data management method provided in an embodiment of this application; Figure 5 The fifth flowchart of the in-vehicle privacy data management method provided in the embodiments of this application; Figure 6 The sixth flowchart of the in-vehicle privacy data management method provided in the embodiments of this application; Figure 7 The seventh flowchart of the in-vehicle privacy data management method provided in the embodiments of this application; Figure 8 The eighth flowchart of the in-vehicle privacy data management method provided in the embodiments of this application; Figure 9 This is a schematic diagram of the structure of the in-vehicle privacy data management device provided in the embodiments of this application; Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0042] Icons: 900 - Acquisition module; 910 - Identification module; 920 - Compliance adjudication module; 930 - Control and adjustment module; 10 - Electronic device; 1001 - Processor; 1002 - Memory; 1003 - Communication bus. Detailed Implementation
[0043] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.
[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.
[0045] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.
[0046] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0047] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.
[0048] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).
[0049] In the description of the embodiments of this application, the technical terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.
[0050] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.
[0051] With the rapid development of intelligent connected vehicles, the amount of data collected and processed by in-vehicle systems every day has increased dramatically, covering various types such as vehicle location, speed, driving behavior, in-vehicle audio and video signals, user preferences, and application logs. A large amount of this data involves personal information and even sensitive data.
[0052] Currently, in-vehicle privacy data management primarily relies on static permission control mechanisms. For example, some vehicles use operating system-level permission models (such as declarative permissions in Android Automotive) to authorize applications to access sensitive data. However, these methods typically grant authorization when a user first installs or launches an application, and the permissions remain valid indefinitely, unable to be dynamically adjusted based on the vehicle's actual operating status. For instance, while driving, some third-party entertainment applications may continue to read high-precision location data or user preference data for personalized recommendations, increasing the risk of privacy breaches and potentially distracting the driver, posing safety hazards.
[0053] To address the aforementioned issues, this application designs an in-vehicle privacy data management method. It captures the raw data stream and adds metadata tags identifying the source and attributes to generate complete vehicle data stream information. Combining multi-dimensional parameters such as vehicle speed, gear position, battery level, and user identifier, it identifies real-time vehicle operating scenarios. Multiple sets of locally pre-built compliance rules are matched to complete data stream compliance assessments one by one and generate exclusive adjudication results. Finally, it synchronously executes data stream interception / release operations and dynamically adjusts application permissions. The entire process runs locally offline without real-time cloud intervention. It relies on the underlying signed data stream to achieve fine-grained control of individual privacy data, abandons the static permanent authorization mode based on real-time scenario recognition, and incorporates a closed-loop logic of parallel assessment of multiple regulations and adjudication-linked permission adjustments. This allows for first achieving unobstructed interception and source tracing of the underlying sensor data stream, and then switching privacy control strategies in real-time according to vehicle operating conditions. It simultaneously considers multiple domestic and international compliance constraints, significantly reducing privacy leaks and driver distraction safety hazards during driving. Furthermore, the metadata uses immutable metadata tags, multiple regulations and rules are persistently stored locally, the assessment response speed remains at the millisecond level, and it can continuously complete the entire process of privacy protection even in offline environments.
[0054] Based on the above ideas, this application first provides a method for managing in-vehicle privacy data. This method is applied to a computing device, which may include a computer, a server, a vehicle controller, and a smart cockpit running Android Automotive or a similar in-vehicle operating system, etc. Figure 1 As shown, this in-vehicle privacy data management method can be implemented in the following ways: Step S100: Obtain vehicle data stream information.
[0055] Step S110: Identify the current operating scenario of the vehicle based on its current operating status.
[0056] Step S120: Based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, conduct a compliance assessment on each data stream in the vehicle data stream information to obtain the compliance ruling result corresponding to each data stream.
[0057] Step S130: Based on the compliance ruling result corresponding to each data stream, perform the corresponding data access control operation on the corresponding data stream and adjust the data access permissions of the application to which the corresponding data stream belongs.
[0058] In the above implementation, the vehicle data stream information includes the entire set of data generated, collected and transmitted during the operation of the vehicle system. It can be composed of multiple data streams, and each data stream is obtained by attaching metadata tags to the monitored raw data stream to identify the data source and attributes.
[0059] The raw data stream refers to the raw data messages and content that are natively output by vehicle sensors, vehicle systems, and third-party applications without being labeled or compliantly filtered. Specifically, it can be obtained through sensors in the Vehicle HAL layer. This raw data stream may include at least two items from vehicle location, vehicle speed, user driving behavior, in-vehicle camera and microphone, user application preferences, and vehicle system logs.
[0060] Metadata tags represent identification information attached to the raw data stream, specifically used to identify the data source and attributes. They can be generated using a TEE hardware root of trust to create an HMAC immutable signature. The tags are stored in a structured format as a quadruple (data attribute type, data source application, data purpose, lifespan). This metadata signature remains transparent and readable throughout the entire transmission chain, from data stream acquisition and data stream adjudication to application adjustments and cloud uploads.
[0061] The vehicle operation scenario refers to the vehicle's working state as determined by the vehicle state machine based on real-time operating parameters. It includes at least four categories: driving state scenario, parking state scenario, charging state scenario, and vehicle sharing state scenario. Among them, the vehicle operation scenario can be determined based on at least two of the following conditions from the vehicle state machine: vehicle speed, gear, battery status, seat occupancy, and user identifier.
[0062] Data compliance rule set: A collection of multiple pre-set in-vehicle data security compliance management rules, such as the EU General Data Protection Regulation (GDPR) rule set, the California Consumer Privacy Act (CCPA) rule set, the Chinese Personal Information Protection Law rule set, and the OEM custom rule set, which are four local persistent rule sets.
[0063] Compliance assessment: This is a calculation process that verifies the compliance status of each individual data stream by combining the current vehicle operation scenario with a set of preset data compliance rules, and outputs a compliance ruling result.
[0064] Compliance ruling results: After the compliance assessment is completed, standardized ruling results are output, which are divided into four categories: interception, allow transmission but require user confirmation, allow transmission directly, and log only without blocking.
[0065] The above-described implementation method of this solution first performs the step of acquiring vehicle data stream information, monitors various raw data streams on the vehicle in real time, and attaches metadata tags to each raw data stream to identify the data source and attributes, thereby generating vehicle data stream information containing multiple tagged data streams. In this way, each raw data stream is attached with metadata tags containing attributes such as "data type, data source application, data usage purpose, and data validity period" when it enters the system, so that all subsequent processing steps can be judged based on the complete context, and each data stream has a traceable identity.
[0066] Simultaneously, the system continuously collects real-time vehicle operating status data such as vehicle speed, gear position, battery status, and user identification. Based on the collected status parameters, it identifies the current operating scenario of the vehicle. The system then retrieves multiple pre-set data compliance rule sets locally, using the identified current vehicle operating scenario as a matching condition to independently conduct compliance assessments on each data stream within the vehicle data stream information and generate compliance ruling results for each data stream. Finally, the system relies on the exclusive compliance ruling results of each data stream to simultaneously complete two-layer linkage operations: one layer performs data access control operations matching the ruling type for that data stream, and the other layer simultaneously adjusts the data access permissions corresponding to the application to which this data stream belongs.
[0067] This solution provides a vehicle privacy data management method that captures raw data streams and adds metadata tags identifying the source and attributes to generate complete vehicle data stream information. It combines multi-dimensional parameters such as vehicle speed, gear position, battery level, and user identification to identify real-time vehicle operating scenarios. Multiple sets of locally pre-built compliance rules are matched to complete data stream compliance assessments one by one and generate personalized adjudication results. Finally, data stream interception / release operations and dynamic adjustment of application permissions are executed synchronously. The entire process runs locally offline without real-time cloud intervention. It relies on underlying signed data streams to achieve fine-grained control of individual privacy data, abandons the static permanent authorization mode based on real-time scenario recognition, and incorporates a closed-loop logic of parallel assessment of multiple regulations and linked adjudication and permission adjustment. This allows for first achieving unobstructed interception and source tracing of underlying sensor data streams, and then switching privacy control strategies in real-time according to vehicle operating conditions. It simultaneously considers multiple domestic and international compliance constraints, significantly reducing privacy leaks and driver distraction safety hazards during driving. Furthermore, the metadata uses immutable metadata tags, multiple regulations and rules are persistently stored locally, assessment response speeds are maintained at millisecond levels, and end-to-end privacy protection can be continuously completed even in offline environments.
[0068] In an optional implementation of this embodiment, as a possible approach, step S120 can be specifically implemented in the following manner: Figure 2 As shown, it includes: Step S200: Based on the current operating scenario of the vehicle, select multiple target data compliance subsets from a set of preset data compliance rules.
[0069] Step S210: For each data stream, conduct a compliance assessment of the data stream based on multiple target data compliance subsets to obtain independent adjudication opinions corresponding to each target data compliance subset.
[0070] Step S220: Integrate and adjudicate multiple independent adjudication opinions to generate a compliance adjudication result corresponding to the data stream.
[0071] In the above implementation method, the target data compliance subset represents a subset of rules selected from multiple preset data compliance rule sets that match the current vehicle operation scenario. Each data compliance rule set contains multiple data compliance subsets, and each data compliance subset corresponds to a vehicle operation scenario.
[0072] An independent decision opinion represents the decision vector output after a single data stream is evaluated by a single target data compliance subset. In this context, an independent decision opinion is output for each of the multiple target data compliance subsets.
[0073] An integrated ruling refers to the process of summarizing multiple independent rulings on the same data stream and making an overall ruling based on the strictest matching principle.
[0074] In the implementation method described above, when making compliance decisions, the system first reads and identifies the current operating scenario of the vehicle and then filters out multiple target data compliance subsets adapted to the current operating conditions from a set of preset data compliance rules based on this scenario. Subsequently, for each individual data stream in the vehicle data stream information, the data stream is sequentially sent to all the filtered target data compliance subsets to perform compliance assessment calculations. Each target data compliance subset will output an independent decision opinion specific to that data stream, resulting in multiple independent decision opinions. Finally, the system performs a unified integration decision calculation on all independent decision opinions corresponding to the same data stream, eliminating the judgment conflicts caused by multiple subset assessments and generating a unique compliance decision result corresponding to that data stream.
[0075] The multiple data compliance rule sets designed in this solution may include the EU General Data Protection Regulation (GDPR) rule base, the California Consumer Privacy Act (CCPA) rule base, the Chinese Personal Information Protection Law rule base, and the OEM-customized rule base. All four rule bases are stored locally in the Privacy Compliance Service process on the vehicle's infotainment system. The initial version is pre-installed at the factory, and parallel queries and rule decisions are all completed locally, thereby ensuring millisecond-level decision latency and supporting offline operation when the network is disconnected.
[0076] The above-described implementation method of this solution first filters irrelevant rules and selects multiple sets of target data compliance subsets that match the identified vehicle operation scenarios. Then, it sends a single data stream to all target subsets for parallel verification to generate multiple independent adjudication opinions. Finally, it integrates multiple opinions to eliminate rule conflicts and outputs a unique adjudication result. This compliance judgment logic of first filtering subsets, then conducting multi-dimensional parallel evaluation, and finally unifying arbitration can first filter invalid rules to reduce vehicle computing power consumption, then conduct multi-dimensional synchronous verification of data stream compliance risks from multiple data compliance rule sets, and finally unify the constraints and contradictions between multiple regulations. This can first improve the real-time evaluation efficiency of vehicle terminals, then comprehensively cover various privacy compliance risk points, and at the same time avoid the failure of control logic caused by conflicting instructions output by multiple rules in parallel.
[0077] In an optional implementation of this embodiment, as a possible approach, the solution for step S220 can achieve the integration decision in the following manner: Figure 3 As shown, it includes: Step S300: Merge and arbitrate the independent rulings corresponding to each target data compliance subset according to the strictest matching principle to generate the compliance ruling result corresponding to the data stream.
[0078] In the above implementation, the strictest matching principle means taking the most stringent value from each target data compliance subset for the same field, and immediately blocking any target data compliance subset that requires blocking.
[0079] Consolidated arbitration refers to the process of comparing, screening, and merging multiple independent arbitration opinions according to the strictest matching principle.
[0080] Specifically, the strictest matching principle includes: for the data retention period field, selecting the shortest period required by each target data compliance subset as the final retention period; for the cross-border data transfer field, when the data destination is a specific region and involves personal information, enforcing the outbound security assessment mechanism stipulated by the target region, and intercepting the data flow if the outbound security assessment fails. In addition to the fields in this example, strictest matching can also be included for other fields, such as fields specifically targeted for advertising, where the final field determination is based on the adjudication results (pass / block) of each target data compliance subset.
[0081] The above-described implementation scheme, after obtaining the independent adjudication opinions generated from each compliant subset of target data corresponding to a data stream, strictly follows the strictest matching principle to conduct a merged arbitration operation to generate the compliance adjudication result corresponding to the data stream. The entire operation follows the two layers of hard judgment criteria contained in the strictest matching principle. The first layer iterates through all independent adjudication opinions for the same control field such as data retention period and cross-border transmission, and selects the field value with the most stringent constraint in each compliant subset of target data as the standard value. The second layer sets a risk-priority fallback logic. As long as any independent adjudication opinion raises a request to intercept the data stream, the final merged arbitration output result is directly locked as interception, thereby realizing merged arbitration.
[0082] As a concrete example, taking the "Precise targeting for advertising" field, GDPR rulings prohibit its use for behavioral advertising, Personal Information Protection Law rulings require separate consent, CCPA rulings require opt-out, and custom rule rulings prohibit precise targeting for advertising when the vehicle is in a driving scenario. Using the strictest matching principle, due to the existence of rule base requirements for blocking, the combined arbitration result is blocking. Taking the "Data Retention Duration" field as another example, the four databases provide values of 30 days, 12 months, 6 months, and 7 days respectively. Following the strictest matching principle, the combined arbitration result takes the shortest 7 days as the final retention duration.
[0083] In the above implementation method, this solution adopts the strictest matching principle to complete the merged arbitration when integrating multiple independent arbitration opinions. During the calculation process, the standard with the highest constraint strength within all subsets is uniformly selected for the same control fields such as data retention, cross-border, and authorization. At the same time, the highest priority is set for the interception command. As long as any subset outputs a prohibition on transmission opinion, the data stream is directly blocked. This can maximize the tightening of the control boundary of vehicle privacy data, simultaneously meet the mandatory constraints of multiple countries' regulations, unify the arbitration logic of vehicle models in multiple regions, and eliminate the need to modify the core arbitration procedure separately for different markets.
[0084] In an optional implementation of this embodiment, as a specific implementation method, the specific decision for step S300 can be achieved in the following manner, such as... Figure 4 As shown, it includes: Step S400: If the independent ruling opinion corresponding to any compliant subset of target data is to prohibit transmission, then the final ruling result corresponding to the data stream is determined to be interception.
[0085] Step S410: If none of the target data compliance subsets require a ban on transmission, but there is an independent ruling opinion for at least one target data compliance subset that requires consent, then the final ruling result for the data stream is determined to be that transmission is permitted but user confirmation is required.
[0086] Step S420: If all compliant subsets of the target data are allowed without additional conditions, the final decision is to directly allow the transmission.
[0087] Step S430: If the data stream hits the preset low-risk rules in all target data compliance subsets, the final decision is to only log and not block it.
[0088] In the above implementation, this solution completes the merging and arbitration of multiple independent rulings and executes a four-layer decision logic. The first layer prioritizes traversing all independent rulings. If any independent ruling indicates a prohibition on transmission, the final decision for that data stream is directly locked as an interception. The second layer, assuming no independent rulings prohibit transmission, continues to check if at least one independent ruling indicates that consent is required. If so, the final decision is determined to be that transmission is allowed but user confirmation is required. The third layer, when there are no prohibitions on transmission or user consent requirements in any of the target data compliance subsets, directly determines the final decision to be that transmission is directly allowed. The fourth layer, only when the conditions of the previous three layers are not met, and the data stream hits the preset low-risk rules within all target data compliance subsets, determines the final decision to be that only logging is done and no blocking occurs.
[0089] The implementation method described above in this solution sets up a fixed four-level progressive judgment process from high risk to low risk. Priority is given to directly blocking the transmission of prohibited opinions. If there is no blocking instruction but authorization is required, the user confirmation mode is determined. If there are no constraints from any rules, the transmission is allowed directly. If only low gray-scale rules are matched, only logs are retained. This layered and progressive judgment logic sets differentiated handling paths for different risk data streams, thereby achieving immediate blocking of high-risk data, returning decision-making power to users for medium-risk data, allowing direct passage for no-risk data, and tolerating minor flaws while leaving traces. This balances privacy and security, regulatory compliance, and the user experience of in-vehicle applications.
[0090] In an optional implementation of this embodiment, for step S130, this scheme can perform different data access control operations under different adjudication results.
[0091] As one possible implementation, this solution, when the compliance ruling allows transmission but requires user confirmation, controls the in-vehicle human-machine interface to display data usage details to the user and wait for a command response. If a user-selected allow transmission command is received, the corresponding data stream is allowed to transmit; if a user-selected deny transmission command is received, the corresponding data stream is blocked. Specifically, the in-vehicle human-machine interface can be an in-vehicle central control HMI privacy dashboard. Specifically, the in-vehicle central control HMI privacy dashboard can be controlled to pop up a window and simultaneously broadcast a voice message to the user, fully displaying the data usage details corresponding to this data stream. The interface continuously listens for user command responses, and the system continuously recognizes user input commands. If a user-issued allow transmission command is received, the normal allow transmission operation is performed for this data stream; if a user-issued deny transmission command is received, the blocking operation for this data stream is immediately performed.
[0092] The above-described implementation method of this solution, when the compliance ruling allows transmission but requires user confirmation, first displays the complete data usage details to the user through a human-machine interface and continuously waits for the user's command response. Then, based on the user's input of a command to allow or deny transmission, it executes the release or blocking operation respectively. This solves the loophole of non-compliance of traditional in-vehicle silent authorization. The human-machine interface, combined with reminder displays, can adapt to different driving and parking scenarios without interfering with driving. The complete user command response record can be directly used as regulatory compliance audit evidence. At the same time, it completely entrusts the decision-making power for the handling of medium-risk data streams to the user, balancing privacy and security with the user's right to choose.
[0093] As another possible implementation, if the compliance ruling is to only log without blocking, this solution writes the relevant events of the corresponding data stream into the violation log and performs a transmission permission operation on the corresponding data stream. In this way, the data stream is allowed and the log is recorded simultaneously, distinguishing between minor compliance flaws and major privacy risks. Normal business functions of the application will not be interrupted due to low-risk rule violations. Based on this handling logic, the original samples of low-risk violation events can be fully retained while ensuring the complete usability of the in-vehicle application. The event data retained in the violation log can provide a data source for event similarity, taking into account both the user's smoothness of use and the data reserve needs for subsequent rule iterations.
[0094] As another possible implementation, if the compliance ruling result is immediate blocking, this solution will directly block the corresponding data stream; if the compliance ruling result is direct permission to transmit, this solution will directly transmit the corresponding data stream.
[0095] In an optional implementation of this embodiment, for step S130 described above, which involves adjusting the data access permissions of the application to which the corresponding data stream belongs based on the compliance ruling result corresponding to each data stream, this solution can execute different adjustment strategies under different ruling results.
[0096] As one possible implementation, if the compliance ruling for the data stream is blocking, this solution directly revoks or downgrades the application's access rights to the corresponding data type of the data stream. If the compliance ruling for the data stream is allowing transmission but requires user confirmation, the solution manages permissions in two stages: before receiving a user's instruction to allow transmission, the application is temporarily not granted the corresponding data access rights; once a user's instruction to refuse transmission is received, the application's existing access rights to the same type of data are immediately revoked.
[0097] The above-described implementation method of this solution adjusts the corresponding data permissions of the application in sync with the two types of adjudication results: single data stream interception and pending confirmation. In the interception scenario, permissions are revoked / gradually downgraded. In the pending confirmation scenario, permissions are temporarily locked before authorization and permanently revoked after user rejection. This logic of linking data stream violations with application permissions achieves precise and targeted punishment. As a result, only the sensitive data corresponding to the violation is restricted, without interfering with the basic functions of the application. The pre-locking permission mechanism prevents the application background from collecting private data without authorization from the source, forming a two-layer risk control closed loop of data stream interception and application permission tightening.
[0098] In an optional implementation of this embodiment, the solution can also adjust the access permissions of applications in the following way: when an application generates multiple compliance rulings that block access within a preset time period and the cumulative number exceeds a threshold, its access permissions to vehicle data are frozen. Here, the preset time period represents a sliding time window for counting the number of application blocking rulings, and the cumulative number threshold represents the critical number of blocking decisions required to determine high-frequency violations and trigger permission freezing; both values can be adaptively adjusted according to the actual application scenario.
[0099] Specifically, this solution can continuously count the cumulative number of interception and blocking compliance rulings generated by each application within a preset time period, and continuously compare the cumulative number of counts with the preset threshold in real time. When the cumulative number of interceptions and blockings of an application exceeds the preset threshold, the system will immediately freeze all vehicle data access permissions of that application and cut off all its underlying data collection channels.
[0100] The above-described implementation method of this solution accumulates the number of times the application blocks compliance rulings within a preset time period. Once the accumulated number exceeds a preset threshold, the application's access to all vehicle data is directly frozen. This underlying punishment logic, based on frequency statistics of sliding time windows and strong freezing of over-limit applications, can distinguish between occasional single-time blocking and continuous, habitual privacy violations. It can accurately identify long-term high-risk applications and completely cut off their data collection source. While single data stream blocking only blocks a single request, this step achieves full-time, full-domain access blocking.
[0101] In an optional implementation of this embodiment, the solution can also adjust the access permissions of the application in the following way: adjust the data access permissions of the application according to the current operating scenario of the vehicle. That is, when the vehicle state machine recognizes that the operating scenario has changed (such as switching from parking to driving), the permission set of each application is recalculated according to the new scenario, and unnecessary permissions that are irrelevant to the current scenario are deprived.
[0102] Furthermore, as a specific implementation, this solution reduces the access permission level of third-party applications to preset data types when the vehicle operation scenario switches from a parked state to a driving state; and restores the access permission of third-party applications to preset data types when the vehicle operation scenario switches back from a driving state to a parked state. For example, in the driving state, the data reading permission of the entertainment application is revoked, and the precise location information permission of the location application is downgraded to a coarse location information permission with an accuracy of approximately 5 kilometers.
[0103] The above-described implementation of this solution proactively reduces the access permissions of third-party applications to preset highly sensitive data types when the vehicle operation scenario switches from a parked state to a driving state, and automatically restores the corresponding access permissions when the scenario switches back to a parked state. This underlying logic of bidirectional gradient adjustment of permissions between the two core operating conditions of driving and parking is specifically matched to the differentiated management needs of the high-safety driving scenario and the low-risk parking scenario. Relying on this bidirectional switching permission adjustment step, it can proactively restrict third-party applications from reading preset sensitive data such as high-precision positioning and audio and video during vehicle operation, eliminating the hidden dangers of distraction and privacy leakage risks while driving.
[0104] In an optional implementation of this embodiment, as one possible approach, this solution adds a complete chain of long-term quantitative control, such as... Figure 5 As shown, it includes: Step S500: Compile historical compliance data for each application.
[0105] Step S510: Calculate the data compliance score for each application based on its historical compliance data.
[0106] Step S520: Based on the data compliance score corresponding to each application, implement access control for the corresponding application.
[0107] In the above implementation, historical compliance data represents a collection of compliance decision results data, such as interception, logging, direct approval, or requiring user consent, within a previously preset historical time period.
[0108] The data compliance score represents a quantitative score of application credibility calculated based on a weighted average of historical compliance data.
[0109] The above-described implementation method of this solution obtains historical compliance data by continuously collecting compliance decision data such as interception, logging, direct access, or user consent required for each application within a preset historical time period. Then, this solution performs a quantitative calculation of a dedicated data compliance score for each application's historical compliance data, and uses the calculated data compliance score of each application as the basis for judgment to perform differentiated data access permission control operations on the corresponding applications.
[0110] The above-described implementation method of this solution continuously collects historical compliance data for each application, calculates the corresponding data compliance score based on the historical compliance data, and uses the quantitative score as a standard to implement differentiated access control for applications. This can objectively and in the long term evaluate the data collection compliance capabilities of third-party applications. Based on the score-based differentiated control, control interference can be reduced for high-scoring and trustworthy applications, while sensitive data permissions can be tightened for low-scoring and high-risk applications, thus effectively purifying the overall privacy ecosystem of in-vehicle third-party applications in the long term.
[0111] In an optional implementation of this embodiment, as one possible implementation, such as Figure 6 As shown, this solution can calculate the data compliance score in the following ways: Step S600: Based on the historical compliance data of each application, determine the violation frequency, data minimization level, user authorization transparency, and historical complaint rate for each application.
[0112] Step S610: Calculate the data compliance score for each application by weighting the violation frequency, data minimization level, user authorization transparency, and historical complaint rate.
[0113] In the above implementation, the violation frequency represents the ratio of application interception / logging violation events to the total number of access requests per unit time. Specifically, this violation rate can be calculated based on the number of times the application was intercepted or logged within 30 days compared to the total number of data access requests during the same period.
[0114] Data minimization is calculated by inversely normalizing the excess proportion of the fields and precision actually requested and used by the application relative to the minimum set of fields necessary to fulfill its declared purpose; the less over-requested, the higher the score.
[0115] User authorization transparency is an indicator of the ease and clarity of authorization notification and revocation operations. Specifically, it can be obtained by normalizing and weighting two factors: notification granularity (whether notification is given on a per-purpose or per-field basis) and revocation convenience (number of steps required for revocation, whether one-click revocation is supported). By default, each factor accounts for half.
[0116] The historical complaint rate can be obtained by normalizing the number of complaints from car owners against the app over a 90-day period.
[0117] Based on the above methods used to calculate the violation frequency, data minimization level, user authorization transparency, and historical complaint rate for each application, this solution can perform a weighted calculation on these factors to obtain a data compliance score for each application. Specifically, the weighted calculation formula for this solution can be as follows: ; in, Indicates the frequency of violations. The first weight corresponding to the frequency of violations; Indicates the degree of data minimization. The second weight represents the degree to which the data is minimized; Indicates transparency in user authorization. This indicates the third weight corresponding to the transparency of user authorization; Indicates the historical complaint rate. This indicates the fourth weight corresponding to the historical complaint rate. .
[0118] The above-described implementation method of this solution extracts four independent quantitative indicators from historical compliance data: violation frequency, data minimization, authorization transparency, and user complaint rate. Customizable weights are configured to perform weighted summation to generate a data compliance score, avoiding the distortion of scores based on a single indicator. All indicators are automatically extracted from logs without manual input, significantly reducing the compliance operation and maintenance costs for car companies. The entire scoring process is traceable and adaptable to regulatory verification.
[0119] In an optional implementation of this embodiment, step S520 can be implemented in the following way: Figure 7 As shown, it includes: Step S700: Determine the score range for the data compliance score of each application based on the data compliance score corresponding to each application.
[0120] Step S710: Based on the access permission configuration corresponding to the data compliance score range of each application, control the access permissions of the corresponding application.
[0121] In the above implementation, the access permission configurations corresponding to different score ranges are different. For example, as a specific example, this solution sets five score ranges: A+, A, B, C, and D. Among them, a score of not less than 0.9 is A+, not less than 0.8 is A, not less than 0.65 is B, not less than 0.5 is C, and less than 0.5 is D. The access permission configurations corresponding to the score ranges are as follows: A+ and A are unrestricted; Level B disables sensitive permissions such as precise positioning, camera, and microphone, and only retains necessary permissions; Level C only allows the minimum permissions to complete core functions and prohibits background data collection; Level D freezes its access to Vehicle HAL data and requires rectification and re-evaluation.
[0122] Based on the above, this solution can determine the corresponding access permission configuration based on the score range of the data compliance score for each application, and then adjust the access permissions of the corresponding applications based on the access permission configuration.
[0123] The above implementation method of this solution first classifies the calculated application data compliance score into intervals to determine the preset score range to which the score belongs, and then retrieves the exclusive access permission configuration rules pre-bound to the interval to complete the application permission allocation. It can open full data access capabilities for high-scoring trusted applications, restrict some sensitive data for medium-scoring applications, and tighten full-domain collection permissions for low-scoring applications, so as to achieve refined, hierarchical and flexible management and control.
[0124] In an optional implementation of this embodiment, the computing device of this solution can also interact and communicate with the cloud to construct a complete set of rules that autonomously iterate in a closed loop, such as... Figure 8 As shown, it can be implemented in the following ways, including: Step S800: Count the violations within the preset time period.
[0125] Step S810: Perform cluster analysis on the violations within the preset time period to identify the target violations that occur frequently.
[0126] Step S820: Transform the identified target violation events into a custom data compliance candidate set.
[0127] Step S830: Transmit the custom data compliance candidate set to the cloud for review.
[0128] Step S840: Receive the rule update package approved by the cloud.
[0129] Step S850: Synchronize the update rules corresponding to the rule update package to multiple data compliance rule sets.
[0130] In the above implementation, this solution can first count the violations within a preset time period, where the violation refers to the violation recorded in the log as described above.
[0131] This solution then encodes each violation event into a feature vector using a four-tuple of application identifier (appId), data type (dataType), scene, and purpose. Density clustering is performed using either DBSCAN or HDBSCAN algorithms (neighborhood radius eps and minimum sample size minSamples are taken as exemplary empirical values, or adaptively determined by HDBSCAN). A confidence score is calculated for each stable cluster, obtained by multiplying the similarity ratio of samples within the cluster by the cluster's stability. Only when the number of samples in a cluster is not less than the first threshold (default 20) and the confidence score is higher than the second threshold (default 0.85) is the common pattern of that cluster converted into a custom data compliance candidate. This custom data compliance candidate is represented in JSON and includes fields such as triggering conditions (matching patterns of appId, dataType, scene, and purpose), suggested actions (block, permit-with-consent, or log-only), applicable legal provisions, source cluster statistics (sample size and confidence score), and the intended scope of application, thereby avoiding false alarms.
[0132] The custom data compliance candidates for all clusters corresponding to the violation times obtained from the above statistics are collected to form a custom data compliance candidate set. This custom data compliance candidate set is then submitted to the cloud-based compliance rule center for review. The cloud review may include four items: regulatory consistency verification, verifying whether the candidate conflicts with the current provisions and existing rules of GDPR, CCPA, and the Personal Information Protection Act; redundancy and conflict verification, verifying whether the candidate has been covered by existing rules or contradicts them; privacy impact and false blocking assessment, replaying candidate rules with historical data in shadow mode, requiring that their false blocking rate be lower than the set upper limit; and sample sufficiency review, verifying the sample size and confidence level across vehicle fleets.
[0133] The review is considered successful only if all four of the above items are passed (with manual compliance review as necessary). Otherwise, the review is rejected and a rejection code and reason (rule conflict, rule redundancy, excessively high false interception rate or insufficient sample) are sent back to the vehicle. The vehicle can then make corrections and resubmit.
[0134] The approved rule candidates are then pushed out in a gray-scale manner via the cloud and then fully fed back to all vehicles via OTA, so that all vehicles receive the rule update package. The update rules corresponding to the rule update package are then synchronized to the custom rule library of multiple data compliance rule sets, thereby realizing the custom rule update.
[0135] The above-described implementation method of this solution periodically summarizes all vehicle violation logs, uses clustering algorithms to mine high-frequency common violations to generate custom rule candidates, uploads them to the cloud for multi-layer compliance verification, and after approval, receives the update package and synchronizes it to the local set of all compliance rules. This solves the problem of the lag in traditional manual rule updates, which can take several months. Automatically generating candidate sets reduces the workload of manual development, cloud verification ensures that new rules comply with regulations, and the return of rule packages enables all vehicles to continuously fill privacy protection vulnerabilities.
[0136] To facilitate understanding of this solution, the in-vehicle privacy data management method designed in this solution is illustrated through the following specific embodiment: Example: When a user is driving on city roads, a third-party music application attempts to obtain the vehicle's high-precision location (accuracy of 1 meter) through Car Service for location-based advertising. In this case, the in-vehicle privacy data management method designed in this solution can achieve privacy data management in the following way: The Vehicle HAL monitoring node intercepted the data stream, appended a quadruple (dataType=PRECISE_LOCATION, sourceApp=com.example.music, purpose=AD_PERSONALIZATION, ttl=NONE) to the data stream, and signed it with a TEE HMAC.
[0137] The vehicle is considered to be in "driving mode" based on three factors: vehicle speed of 60 km / h, gear in D, and occupancy of both the driver's seat and the passenger seat.
[0138] Parallel query of multiple regulatory rules engines: GDPR ruling "Explicit consent required + not allowed for behavioral advertising", CCPA ruling "Op-out required", Personal Information Protection Act ruling "Separate consent required", Custom rule ruling "Precise location is prohibited for advertising while driving".
[0139] The strictest matching principle was applied, resulting in a combined ruling of "Blocked, reason: Precise location targeting while driving for advertising violates GDPR, PIPL, and custom rules."
[0140] Output the interception command and notify the third-party music application via Binder IPC: the reason for rejection and the appeal portal.
[0141] Reduce the precise location permissions of third-party music apps to coarse location permissions (accuracy 5 km), only for driving scenarios; automatically restore when parked.
[0142] The daily violation count for third-party music apps will be increased by 1, and the data compliance score will be updated to Grade B (formerly Grade A).
[0143] The privacy dashboard displays in real time: "2026-04-29 14:32 A precise location reading request was blocked. Application: Music X. Reason: Advertising purposes while driving." Users can revoke all permissions for this application with one click.
[0144] Figure 9 A schematic structural block diagram of an in-vehicle privacy data management device provided in this application is presented. It should be understood that the device is disposed on the computing device described above, and the device is connected to... Figures 1 to 8 The method embodiment executed by the computing device corresponds to the method described above, and can perform the steps involved in the aforementioned method. The specific functions of the device can be found in the description above, and detailed descriptions are omitted here to avoid repetition. The device includes at least one software functional module that can be stored in memory or embedded in the device's operating system (OS) in the form of software or firmware. Specifically, the device includes: an acquisition module 900, an identification module 910, a compliance adjudication module 920, and a control adjustment module 930; the acquisition module 900 is used to acquire vehicle data stream information, wherein the vehicle data stream information includes multiple data streams, each data stream is obtained by attaching metadata tags to the monitored raw data stream to identify the data source and attributes; the identification module 910 is used to identify the current operating scenario of the vehicle based on the current operating status of the vehicle; the compliance adjudication module 920 is used to perform compliance assessment on each data stream in the vehicle data stream information based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, and obtain the compliance adjudication result corresponding to each data stream; the control adjustment module 930 is used to perform corresponding data access control operations on the corresponding data stream and adjust the data access permissions of the application to which the corresponding data stream belongs based on the compliance adjudication result corresponding to each data stream.
[0145] The aforementioned in-vehicle privacy data management device captures the raw data stream and adds metadata tags identifying the source and attributes to generate complete vehicle data stream information. It combines multi-dimensional parameters such as vehicle speed, gear position, battery level, and user identification to identify real-time vehicle operating scenarios. It matches multiple sets of locally pre-built compliance rules to complete data stream compliance assessments one by one and generates a unique ruling. Finally, it synchronously executes data stream interception / release operations and dynamically adjusts application permissions. The entire process runs locally offline without real-time cloud intervention. It relies on the underlying signed data stream to achieve fine-grained control of individual privacy data, abandons the static permanent authorization mode based on real-time scenario recognition, and incorporates a closed-loop logic of parallel assessment of multiple regulations and linked adjudication for permission adjustments. This allows for first achieving unobstructed interception and source tracing of the underlying sensor data stream, and then switching privacy control strategies in real-time according to vehicle operating conditions. It simultaneously considers multiple domestic and international compliance constraints, significantly reducing privacy leaks and driver distraction safety hazards during driving. Furthermore, the metadata uses immutable metadata tags, multiple regulations and rules are persistently stored locally, the assessment response speed remains at the millisecond level, and it can continuously complete the entire process of privacy protection even in offline environments.
[0146] According to some embodiments of this application, such as Figure 10 As shown, this application provides an electronic device 10, including: a processor 1001 and a memory 1002. The processor 1001 and the memory 1002 are interconnected and communicate with each other through a communication bus 1003 and / or other forms of connection mechanism (not shown). The memory 1002 stores a computer program executable by the processor 1001. When the computing device is running, the processor 1001 executes the computer program to execute any of the optional implementation methods of the aforementioned client execution, such as steps S100 to S130: acquiring vehicle data stream information; identifying the current operating scenario of the vehicle based on the current operating status of the vehicle; performing compliance assessment on each data stream in the vehicle data stream information based on the current operating scenario of the vehicle and multiple preset data compliance rule sets to obtain the compliance ruling result corresponding to each data stream; and performing corresponding data access control operations on the corresponding data stream and adjusting the data access permissions of the application to which the corresponding data stream belongs based on the compliance ruling result corresponding to each data stream.
[0147] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the method in any of the aforementioned optional implementations.
[0148] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0149] This application provides a computer program product that, when run on a computer, causes the computer to perform a method in any of the optional implementations.
[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no structural conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. A method for managing in-vehicle privacy data, characterized in that, The method includes: Obtain vehicle data stream information, wherein the vehicle data stream information includes multiple data streams, each data stream is obtained by attaching metadata tags to the monitored raw data stream to identify the data source and attributes; Identify the current operating scenario of the vehicle based on its current operating status; Based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, a compliance assessment is performed on each data stream in the vehicle data stream information to obtain the compliance ruling result corresponding to each data stream; Based on the compliance ruling for each data stream, perform corresponding data access control operations on the data stream and adjust the data access permissions of the application to which the data stream belongs.
2. The method according to claim 1, characterized in that, The process involves evaluating the compliance of each data stream in the vehicle data stream information based on the current operating scenario of the vehicle and multiple preset data compliance rule sets, to obtain a compliance ruling result for each data stream, including: Based on the current operating scenario of the vehicle, multiple target data compliance subsets are selected from multiple preset data compliance rule sets. Each preset data compliance rule set includes multiple data compliance subsets, and each data compliance subset corresponds to a vehicle operating scenario. For each data stream, a compliance assessment is performed on the data stream based on the multiple target data compliance subsets to obtain independent adjudication opinions corresponding to each target data compliance subset; Multiple independent rulings are integrated and adjudicated to generate a compliance ruling result corresponding to the data stream.
3. The method according to claim 2, characterized in that, The process of integrating and adjudicating multiple independent opinions to generate a compliance adjudication result corresponding to the data stream includes: The independent adjudication opinions corresponding to each target data compliance subset are merged and arbitrated according to the strictest matching principle to generate the compliance adjudication result corresponding to the data stream. The strictest matching principle means taking the most stringent value of each target data compliance subset for the same field, and the principle that any target data compliance subset that requires interception should be intercepted immediately.
4. The method according to claim 3, characterized in that, The strictest matching principle includes: For the data retention period field, the shortest period required by each target data compliance subset is selected as the final retention period; For cross-border data transfer fields, when the data destination is located in a specific region and involves personal information, the outbound security assessment mechanism stipulated by the target region shall be enforced. If the outbound security assessment fails, the data flow shall be intercepted.
5. The method according to claim 3, characterized in that, The step of merging and arbitrating the independent rulings corresponding to each target data compliance subset according to the strictest matching principle to generate the compliance ruling result corresponding to the data stream includes: If the independent ruling opinion corresponding to any compliant subset of target data is to prohibit transmission, then the final ruling result corresponding to the data stream is determined to be interception; If none of the target data compliance subsets require a ban on transmission, but there is an independent ruling for at least one target data compliance subset that requires consent, then the final ruling for the data stream is determined to be that transmission is permitted but user confirmation is required. If all compliant subsets of the target data are permitted without additional conditions, the final decision is to directly allow the transmission. If the data stream matches the preset low-risk rules in all target data compliance subsets, the final decision is to only log the data and not block it.
6. The method according to claim 5, characterized in that, Based on the compliance ruling for each data stream, perform corresponding data access control operations on the corresponding data stream, including: If the compliance ruling allows transmission but requires user confirmation, the system controls the in-vehicle human-machine interface to display data usage details to the user and wait for a command response. If a user-selected permission to transmit instruction is received, then the permission to transmit operation is performed on the corresponding data stream; If a user-selected rejection instruction is received, an interception and blocking operation will be performed on the corresponding data stream.
7. The method according to claim 5, characterized in that, Based on the compliance ruling for each data stream, perform corresponding data access control operations on the corresponding data stream, including: If the compliance ruling is to log only without blocking, the relevant events of the corresponding data stream will be written to the violation log, and the corresponding data stream will be allowed to transmit.
8. The method according to claim 5, characterized in that, Based on the compliance ruling for each data stream, adjust the data access permissions of the application to which the corresponding data stream belongs, including: If the compliance ruling for the data stream is blocking, the application to which the data stream belongs shall revoke or downgrade its access rights to the corresponding data type of the data stream. If the compliance ruling for the data stream is that transmission is permitted but user confirmation is required, the corresponding permission will not be granted before the user selects the permission to transmit instruction, and the granted permission will be revoked after the user selects the denial to transmit instruction.
9. The method according to claim 5, characterized in that, The method further includes: If an application generates multiple compliance rulings that block or obstruct compliance within a preset time period and the cumulative number of such rulings exceeds a threshold, its access to vehicle data will be frozen.
10. The method according to claim 1, characterized in that, The method further includes: Adjust the application's data access permissions based on the current operating scenario of the vehicle.
11. The method according to claim 10, characterized in that, The adjustment of application data access permissions based on the current operating scenario of the vehicle includes: When the vehicle operation scenario switches from a parked state to a driving state, the access permission level of third-party applications to preset data types is reduced; When the vehicle operation scenario switches from driving mode to parking mode, restore the third-party application's access permissions to preset data types.
12. The method according to claim 1, characterized in that, The method further includes: Compile historical compliance data for each application; Calculate the data compliance score for each application based on its historical compliance data; Access control is applied to each application based on its data compliance score.
13. The method according to claim 12, characterized in that, The process of calculating a data compliance score for each application based on its historical compliance data includes: Based on the historical compliance data of each application, determine the violation frequency, data minimization level, user authorization transparency, and historical complaint rate for each application; The data compliance score for each application is calculated by weighting the frequency of violations, the degree of data minimization, the transparency of user authorization, and the historical complaint rate.
14. The method according to claim 12, characterized in that, The method of controlling access permissions for each application based on its data compliance score includes: Based on the data compliance score for each application, determine the score range for the data compliance score for each application; Access control is applied to each application based on the access permission configuration corresponding to the score range of its data compliance score. The access permission configurations are different for different score ranges.
15. The method according to claim 1, characterized in that, The method further includes: Statistics on violations within a preset time period; Cluster analysis is performed on the violations within the preset time period to identify frequently occurring target violations; Transform identified target violation events into a custom data compliance candidate set; The custom data compliance candidate set is transmitted to the cloud for review. Receive a rule update package that has been approved by the cloud, wherein the rule update package is generated based on the content of the custom data compliance candidate set; The update rules corresponding to the rule update package are synchronized to multiple data compliance rule sets.
