Vehicle computing resource settlement methods, devices, electronic equipment and storage media
By combining multi-dimensional computing power metrics with blockchain networks, the problem of inaccurate quantification and reliable evidence storage of vehicle computing power consumption has been solved. Fair settlement has been achieved in unstable networks and environments where multiple parties do not trust each other, ensuring accurate quantification and tamper-proof evidence storage of computing power consumption.
Patent Information
- Application Number
- CN202611028943.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-10
- Publication Date
- 2026-11-17
AI Technical Summary
Under the combined constraints of unstable in-vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, automakers, and tenants, the consumption of in-vehicle computing power cannot be accurately quantified, reliably stored, and fairly settled.
By establishing multi-dimensional vehicle-mounted computing power indicators, the operating status of hardware components is obtained using a trusted computing power metering module, computing power resource consumption is generated, and it is encapsulated into an encrypted computing power proof message. Settlement is then carried out using the settlement rules in the target blockchain network, achieving accurate quantification and trusted evidence storage.
It enables accurate quantification and fair settlement of vehicle computing power consumption in unstable networks and environments where multiple parties distrust each other, solves the problems of uncompensated hidden losses and data tampering, and ensures the immutability and transparency of settlement results.
Smart Images

Figure CN123123046A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle technology, and in particular to a method, apparatus, electronic device and storage medium for calculating computing resources for vehicles. Background Technology
[0002] With the rapid development of the intelligent connected vehicle industry, the computing power of in-vehicle computing platforms continues to rise. Currently, the computing power of a single chip in the autonomous driving domain controller of mainstream intelligent vehicles has reached more than 254 TOPS (Tera Operations Per Second), and the computing power of the cockpit domain chip NPU (Neural Processing Unit) has also exceeded 30 TOPS. However, according to traffic statistics, urban private cars are parked for more than 90% of the time every day. Therefore, the high-performance computing resources of vehicles are actually idle and wasted during the parking period.
[0003] In related technologies, for the parking computing power sharing scenario, the main approach is to receive computing power tasks and obtain revenue while the vehicle is charging, or to unify the management of the computing power of the parking charging vehicle with the cloud computing power.
[0004] However, the aforementioned vehicle computing power scheduling scheme faces the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, automakers, and tenants. As a result, the consumption of vehicle computing power cannot be accurately quantified, and reliable evidence storage and fair settlement cannot be achieved. These issues urgently need to be addressed. Summary of the Invention
[0005] This application provides a method, apparatus, electronic device, and storage medium for settling vehicle computing resources, in order to solve the technical problem that vehicle computing power consumption cannot be accurately quantified, reliably stored, and fairly settled under the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, vehicle manufacturers, and tenants.
[0006] The first aspect of this application provides a method for settling computing resources for vehicles, comprising the following steps: Based on multi-dimensional vehicle-mounted computing power metrics, a computing power resource measurement standard for the vehicle is established. Based on the trusted computing power metering module, the operating status of the hardware components corresponding to each vehicle-mounted computing power index is obtained, and the computing power resource consumption corresponding to each hardware component is generated according to the computing power resource metering standard. The computing power resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message and stored in the vehicle-side trusted execution environment. In response to the vehicle being connected to the network, the encrypted computing power proof message is sent to the target blockchain network, and the settlement is performed according to the computing power resource consumption based on the settlement rules in the target blockchain network to obtain the computing power resource settlement result of the vehicle.
[0007] According to one embodiment of this application, establishing a computing resource measurement standard for the vehicle based on multi-dimensional vehicle-mounted computing power metrics includes: The system obtains the integral value of the actual number of instruction cycles executed by the CPU (Central Processing Unit) within a preset task cycle, the integral value of the active percentage and running time of the GPU (Graphics Processing Unit) streaming multiprocessor, the integral value of the number of NPU operations, the integral value of the real-time throughput of memory read and write bandwidth, the total uplink and downlink data traffic of the network within the task cycle, and the difference between the real-time chip temperature and the ambient reference temperature. The computing resource measurement standard for the vehicle is established based on the integral value of the actual number of CPU instruction execution cycles within the preset task cycle and the corresponding first weight coefficient, the integral value of the GPU streaming multiprocessor activity ratio and running time and the corresponding second weight coefficient, the integral value of the NPU operation count and the corresponding third weight coefficient, the periodic integral value of the memory read / write bandwidth real-time throughput and the corresponding fourth weight coefficient, the total network uplink or downlink data traffic within the task cycle and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
[0008] According to one embodiment of this application, obtaining the operating status of the hardware component corresponding to each vehicle-mounted computing power index and generating the computing power resource consumption corresponding to each hardware component according to the computing power resource metering standard includes: Based on the trusted computing power measurement module, the multi-dimensional vehicle computing power indicators are directly read through memory-mapped input and output, and the operating status of the hardware components corresponding to each vehicle computing power indicator is obtained. The operating status of the hardware components is quantified to generate the computing resource consumption corresponding to each hardware component.
[0009] According to one embodiment of this application, the target blockchain network includes a first blockchain and a second blockchain, wherein the first blockchain is deployed on a roadside communication unit, a mobile edge computing node, and a cloud platform, and is used to write the task metadata and verification root of the encrypted computing power proof message; the second blockchain is used to receive the verification root of the first blockchain, so as to verify the task metadata according to the verification root.
[0010] According to one embodiment of this application, the step of using the settlement rules in the target blockchain network to settle according to the computing power resource consumption and obtain the computing power resource settlement result of the vehicle includes: The settlement rules in the target blockchain network are invoked to calculate the settlement amount of the computing power resources based on the amount of computing power resources consumed, and the settlement result of the computing power resources is generated.
[0011] The vehicle computing resource settlement method according to embodiments of this application establishes a vehicle computing resource measurement standard based on multi-dimensional vehicle computing power indicators. Based on a trusted computing power measurement module, it acquires the operating status of the hardware components corresponding to each vehicle computing power indicator and generates the computing resource consumption for each hardware component. The computing resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message, which is sent to the target blockchain network when the vehicle has network access. The settlement rules in the target blockchain network are used to settle the computing resource consumption, resulting in the vehicle's computing resource settlement result. This solves the technical problem of accurately quantifying, reliably storing, and fairly settling vehicle computing power consumption under the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, manufacturers, and tenants.
[0012] A second aspect of this application provides a computing resource settlement device for vehicles, comprising: A construction module is used to establish a measurement standard for the computing power resources of the vehicle based on multi-dimensional vehicle-mounted computing power metrics. The acquisition module is used to acquire the operating status of the hardware components corresponding to each vehicle computing power index based on the trusted computing power metering module, generate the computing power resource consumption corresponding to each hardware component according to the computing power resource metering standard, and encapsulate the computing power resource consumption and vehicle identifier into an encrypted computing power proof message and store it in the vehicle-side trusted execution environment. The settlement module is used to respond to the vehicle being in a network-connected state by sending the encrypted computing power proof message to the target blockchain network, and using the settlement rules in the target blockchain network to settle the computing power resource consumption, thereby obtaining the computing power resource settlement result of the vehicle.
[0013] According to one embodiment of this application, the construction module includes: The first acquisition unit is used to acquire the integral value of the number of CPU actual instruction execution cycles, the integral value of the GPU streaming multiprocessor active ratio and running time, the integral value of the NPU operation count, the period integral value of the real-time throughput of memory read and write bandwidth, the total network uplink and downlink data traffic within the task cycle, and the difference between the chip real-time temperature and the ambient reference temperature. The construction unit is used to establish the computing resource measurement standard for the vehicle based on the integral value of the actual number of CPU instruction execution cycles within the preset task cycle and the corresponding first weight coefficient, the integral value of the active ratio of the GPU streaming multiprocessor and the running time and the corresponding second weight coefficient, the integral value of the NPU operation count and the corresponding third weight coefficient, the periodic integral value of the real-time throughput of memory read and write bandwidth and the corresponding fourth weight coefficient, the total uplink or downlink data traffic of the network within the task cycle and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
[0014] According to one embodiment of this application, the acquisition module includes: The second acquisition unit is used to directly read the multi-dimensional vehicle computing power indicators through memory-mapped input and output based on the trusted computing power measurement module, and to acquire the operating status of the hardware components corresponding to each vehicle computing power indicator. The generation unit is used to quantify the operating status of the hardware components and generate the computing resource consumption corresponding to each hardware component.
[0015] According to one embodiment of this application, the target blockchain network includes a first blockchain and a second blockchain, wherein the first blockchain is deployed on a roadside communication unit, a mobile edge computing node, and a cloud platform, and is used to write the task metadata and verification root of the encrypted computing power proof message; the second blockchain is used to receive the verification root of the first blockchain, so as to verify the task metadata according to the verification root.
[0016] According to one embodiment of this application, the settlement module includes: The settlement unit is used to invoke the settlement rules in the target blockchain network, calculate the settlement amount of the computing power resources based on the computing power resource consumption, and generate the computing power resource settlement result.
[0017] The vehicle computing resource settlement device according to the embodiments of this application establishes a vehicle computing resource measurement standard based on multi-dimensional vehicle computing power indicators. Based on a trusted computing power measurement module, it acquires the operating status of the hardware components corresponding to each vehicle computing power indicator and generates the computing resource consumption corresponding to each hardware component. The computing resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message, which is sent to the target blockchain network when the vehicle has network access. The settlement rules in the target blockchain network are used to settle the computing resource consumption, resulting in the vehicle's computing resource settlement result. This solves the technical problem of accurately quantifying, reliably storing, and fairly settling vehicle computing power consumption under the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, manufacturers, and tenants.
[0018] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle computing resource settlement method as described in the above embodiments.
[0019] A fourth aspect of this application provides a computer-readable storage medium storing computer instructions for causing the computer to execute the vehicle computing resource settlement method as described in the above embodiments.
[0020] A fifth aspect of this application provides a computer program product, including a computer program that is executed to implement the vehicle computing resource settlement method described in the above embodiments.
[0021] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0022] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a method for settling computing resources for a vehicle according to an embodiment of this application. Figure 2 This is a schematic diagram of a system architecture according to an embodiment of this application; Figure 3 This is a hardware sampling architecture diagram of a TMA (Trusted Measurement Agent) according to an embodiment of this application; Figure 4 This is a diagram of a layered blockchain evidence storage architecture according to an embodiment of this application; Figure 5 This is an example diagram of a vehicle computing resource settlement device according to an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation
[0023] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0024] The following description, with reference to the accompanying drawings, outlines a method, apparatus, electronic device, and storage medium for settling vehicle computing resources according to embodiments of this application. Addressing the issues mentioned in the background art, such as the inability to accurately quantify, reliably verify, and fairly settle vehicle computing power consumption under the combined constraints of unstable in-vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, manufacturers, and tenants, this application provides a method for settling vehicle computing power resources. In this method, a vehicle computing power resource measurement standard is established based on multi-dimensional in-vehicle computing power indicators. Based on a reliable computing power measurement module, the operating status of the hardware components corresponding to each in-vehicle computing power indicator is obtained, and the computing power resource consumption corresponding to each hardware component is generated. The computing power resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message. When the vehicle has network access, the encrypted computing power proof message is sent to the target blockchain network. The settlement rules in the target blockchain network are used to settle the computing power resource consumption, resulting in the vehicle's computing power resource settlement result. This solves the technical problem of accurately quantifying, reliably storing, and fairly settling vehicle computing power consumption under the combined constraints of unstable in-vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, car manufacturers, and tenants.
[0025] Specifically, before introducing the embodiments of this application, we will first introduce the relevant technologies and existing technical problems of this application.
[0026] With the rapid development of the intelligent connected vehicle industry, the computing power of in-vehicle computing platforms continues to rise. Currently, the single-chip computing power of autonomous driving domain controllers (such as NVIDIA Orin-X chips and Qualcomm Snapdragon RideFlex chips) in mainstream intelligent vehicles has reached over 254 TOPS, and the computing power of NPUs in cockpit domain chips (such as Qualcomm 8295) has also exceeded 30 TOPS. However, according to traffic statistics, urban private cars only travel for about 1.5 hours per day on average, with more than 90% of the time spent parked. During this parking period, the high-performance computing resources of in-vehicle vehicles are essentially idle and wasted.
[0027] In related technologies, vehicle computing power scheduling schemes are mainly concentrated in the following two categories: (1) Driving task offloading scheduling: such as vehicle-road cooperative computing offloading based on MEC (Multi-access Edge Computing), vehicle-to-vehicle computing collaboration based on D2D (Device-to-Device) communication, etc. These schemes regard vehicles as mobile computing nodes and focus on real-time task scheduling during driving; (2) Data center-style resource pooling scheduling: referencing cloud data center resource scheduling models (such as Kubernetes cluster scheduling, YARN resource manager, etc.), assuming that computing nodes are continuously online and resource supply is stable, lacking adaptation to the dynamic online and offline characteristics of mobile nodes.
[0028] However, the above technical solutions have the following technical problems: (1) There is a lack of multi-dimensional vehicle-side computing power consumption measurement models. The relevant technical solutions usually regard the vehicle as a whole computing node and use a single computing power indicator (such as TOPS, CPU utilization rate) for resource measurement. They fail to establish a refined consumption measurement model for the various hardware resources of the vehicle heterogeneous computing platform (CPU / GPU / NPU). Different tasks have significant differences in the consumption patterns and intensity of resources such as CPU, GPU, NPU, memory, and bandwidth. A single indicator cannot truly reflect the actual occupancy of various resources. (2) There is a lack of consideration for implicit costs such as hardware wear and tear. High-load computing tasks will cause the chip temperature to rise, accelerate chip aging, and increase the energy consumption of the vehicle thermal management system. The relevant technical solutions only consider the task execution time or computing power utilization rate in computing power pricing and revenue distribution. They do not include implicit costs such as accelerated hardware aging and cooling energy consumption in the billing model. This results in car owners bearing uncompensated hardware depreciation costs, making "computing power rental" an unfair transaction for car owners that is "profitable on paper but incurring hidden losses". This seriously restricts the sustainability of the computing power sharing model. (3) Lack of an immutable and verifiable computing power consumption evidence storage system. Vehicle computing power sharing involves three parties: car owners (computing power providers), car manufacturers (platform operators), and tenants (computing power users). The relevant technical solutions rely on centralized platforms to record computing power consumption data. However, as a transaction participant (usually charging platform service fees), the car manufacturer platform has the motivation and ability to unilaterally tamper with the data in its centralized database. In a scenario where the three parties do not trust each other, the centralized evidence storage method cannot establish a reliable trust foundation and cannot guarantee the public nature of the metering data. Positive; (4) Lack of lightweight on-chain and automatic settlement mechanism adapted to vehicle network environment. Vehicle computing power measurement requires high-frequency sampling (second level) to accurately reflect computing power consumption. If all sampled data is uploaded to the chain one by one, high gas costs will be generated. It is not economically feasible in short-term small-amount task scenarios. At the same time, there are intermittent network outages in the vehicle network environment (such as signal blind spots in underground parking lots, tunnels, etc.). The blockchain solutions of related technical solutions require nodes to be online continuously to conduct consensus and transaction submission, which does not match the actual network conditions of the vehicle environment, making it difficult to guarantee the efficiency and reliability of on-chain.
[0029] Therefore, based on the above-mentioned technical problems, this application constructs a multi-dimensional computing power model that includes chip thermal state compensation components and a hardware-level anti-tampering metering architecture based on a trusted execution environment to achieve accurate quantification and hidden loss compensation of vehicle-mounted heterogeneous computing power consumption. Furthermore, based on a layered blockchain notarization and smart contract automatic settlement mechanism, it completes the immutable notarization and fair automatic settlement of computing power consumption data under the conditions of unstable vehicle networks and mutual distrust among multiple parties.
[0030] Specifically, Figure 1This is a flowchart illustrating a method for settling computing resources for a vehicle, as provided in an embodiment of this application.
[0031] like Figure 1 As shown, the method for settling computing power resources for this vehicle includes the following steps: In step S101, a measurement standard for vehicle computing power resources is established based on multi-dimensional vehicle-mounted computing power metrics.
[0032] According to one embodiment of this application, a vehicle computing power resource measurement standard is established based on multi-dimensional vehicle computing power weight indicators, including: obtaining the integral value of the actual number of CPU instruction cycles executed within a preset task period, the integral value of the active percentage of GPU streaming multiprocessors and running time, the integral value of NPU operation count, the periodic integral value of the real-time throughput of memory read / write bandwidth, the total network uplink and downlink data traffic within the task period, and the difference between the chip real-time temperature and the ambient reference temperature; and establishing the vehicle computing power resource measurement standard based on the integral value of the actual number of CPU instruction cycles executed within the preset task period and the corresponding first weight coefficient, the integral value of the active percentage of GPU streaming multiprocessors and running time and the corresponding second weight coefficient, the integral value of NPU operation count and the corresponding third weight coefficient, the periodic integral value of the real-time throughput of memory read / write bandwidth and the corresponding fourth weight coefficient, the total network uplink or downlink data traffic within the task period and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
[0033] The preset task cycle can be set by those skilled in the art based on the task's execution time dimension, calculation and settlement cycle dimension, physical constraints of hardware sampling, etc., and is not specifically limited here.
[0034] Specifically, such as Figure 2 As shown, this application introduces a multi-dimensional hardware loss-aware computing power unit COU (ComputeOutputUnit) to break through the traditional single computing power statistical dimension and realize the normalized quantification of the consumption and hidden losses of multiple types of hardware resources.
[0035] Specifically, the multi-dimensional vehicle-mounted computing power metrics of this application mainly include, within a preset task cycle, the integral value C_cpu of the actual number of CPU instruction cycles executed, collected by the performance counter register of the SOC (System-on-Chip) hardware PMU (Performance Monitoring Unit); the integral value C_gpu of the GPU streaming multiprocessor active ratio and running time, collected by the GPU underlying hardware performance counter; the integral value C_npu of the NPU operation count, collected by the NPU hardware native counter (where the NPU operation count integral value is differentially weighted according to INT8 / FP16 / FP32 operation precision); the periodic integral value C_mem of the real-time throughput of memory read / write bandwidth, collected by the DDR (Double Data Rate) memory controller hardware register; the total network uplink and downlink data traffic C_bw within the task cycle, collected by the vehicle-mounted baseband chip hardware statistics module; and the integral value C_bw of the vehicle-mounted BMS (Battery Management System). The difference C_thermal between the real-time chip temperature and the ambient reference temperature, collected by the temperature sensor array of the Battery Management System (BMS), represents the change in hardware thermal state caused by computing load. Each hardware resource component corresponds to a dynamically calibrated weight coefficient. For example, the integral value of the actual number of CPU instruction cycles executed within a preset task cycle, C_cpu, corresponds to the first weight coefficient α; the integral value of the active percentage and running time of the GPU streaming multiprocessor, C_gpu, corresponds to the second weight coefficient β; the integral value of the number of NPU operations, C_npu, corresponds to the third weight coefficient γ; the periodic integral value of the real-time throughput of memory read / write bandwidth, C_mem, corresponds to the fourth weight coefficient δ; the total uplink or downlink data traffic within the task cycle, C_bw, corresponds to the fifth weight coefficient ε; and the difference between the real-time chip temperature and the ambient reference temperature, C_thermal, corresponds to the sixth weight coefficient ζ. The sum of the products of the multi-dimensional vehicle computing power weight index and its corresponding weight coefficient represents the computing power resource measurement standard COU, expressed as: COU=α·C_cpu+β·C_gpu+γ·C_npu+δ·C_mem+ε·C_bw+ζ·C_thermal; The technical definitions and hardware data sources for each vehicle-mounted computing power index are shown in Table 1. Table 1
[0036] It should be noted that the computing power resource measurement standard COU is predefined and is a physical dimension normalized measurement indicator for computing power resource consumption. It is only used for technical aspects such as computing power traceability, loss quantification and data storage, and is not tied to any commercial pricing, fee rate or revenue sharing rules.
[0037] In this application, the difference between the chip's real-time temperature and the ambient reference temperature, C_thermal, is a thermal loss compensation item and a technical dimension specific to vehicle computing power measurement. Its corresponding sixth weight coefficient ζ is calibrated by both the Arrhenius chip aging model and the vehicle thermal management COP energy efficiency coefficient. This transforms the implicit losses caused by computing load, such as chip temperature rise, cooling energy consumption, and accelerated hardware aging, into quantifiable computing power measurement indicators. This technically addresses the loss deficiency problem of traditional measurement models. In other words, the Arrhenius chip aging model is used to map the chip temperature rise caused by computing load to an equivalent hardware depreciation cost coefficient, and the COP energy efficiency coefficient is used to map the temperature rise to an equivalent cooling energy consumption cost coefficient. The sixth weight coefficient ζ is the weighted sum of the hardware depreciation cost coefficient and the cooling energy consumption cost coefficient.
[0038] It should be noted that the weight coefficients corresponding to the multi-dimensional vehicle computing power metrics in this application embodiment, namely the first weight coefficient α, the second weight coefficient β, the third weight coefficient γ, the fourth weight coefficient δ, the fifth weight coefficient ε, and the sixth weight coefficient ζ, can be dynamically calibrated in the cloud according to different vehicle chip models and different computing power task types (model inference, model training), thereby achieving the uniformity and comparability of COU measurement results under different hardware platforms and different task scenarios. Specifically, for inference tasks, the NPU operation count integral value C_npu can be increased, corresponding to the third weight coefficient γ; for training tasks, the GPU streaming multiprocessor active ratio and running time integral value C_gpu can be increased, corresponding to the second weight coefficient β.
[0039] In step S102, based on the trusted computing power metering module, the operating status of the hardware components corresponding to each vehicle-mounted computing power index is obtained, and the computing power resource consumption corresponding to each hardware component is generated according to the computing power resource metering standard. The computing power resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message and stored in the vehicle-side trusted execution environment.
[0040] According to one embodiment of this application, the operating status of the hardware component corresponding to each vehicle computing power indicator is obtained, and the computing power resource consumption corresponding to each hardware component is generated according to the computing power resource metering standard, including: based on the trusted computing power metering module, multi-dimensional vehicle computing power indicators are directly read through memory-mapped input and output, and the operating status of the hardware component corresponding to each vehicle computing power indicator is obtained; the operating status of the hardware component is quantified, and the computing power resource consumption corresponding to each hardware component is generated respectively.
[0041] Specifically, such as Figure 3As shown, in this embodiment of the application, a trusted computing power metering module TMA is first deployed in the vehicle-side trusted execution environment (TEE) (such as ARMTrustZoneSecureWorld (an independent and isolated runtime environment) or an independent HSM (HardwareSecurityModule)). The TEE is isolated from the ordinary execution environment to prevent malicious software or users with root privileges from tampering with or spying on the TMA code and data.
[0042] TMA does not rely on the / proc virtual file system interface provided by the operating system (which is easily tampered with by LD_PRELOADHook or kernel modules). Instead, it directly accesses the register address space of each hardware component through memory-mapped I / O (Input / Output) (e.g., MMIO (Memory-Mapped I / O)) and reads the vehicle computing power indicators of each hardware component. Specifically, it directly reads the SOC hardware PMU performance counter register, GPU underlying hardware performance counter, NPU hardware native counter, DDR memory controller hardware register, vehicle baseband chip hardware statistics module, and hardware-level counters such as the vehicle BMS temperature sensor array through isolated CAN (Controller Area Network) / LIN (Local Interconnect Network) bus interfaces. Then, it obtains the operating status data of each hardware component, such as CPU, GPU, NPU, memory, network, and temperature sensor, instead of forwarding through the vehicle entertainment system or gateway, thus avoiding data tampering during transmission.
[0043] Secondly, this application quantifies the operating status data of each hardware component according to the computing power resource measurement standard COU, and generates the computing power resource consumption corresponding to each hardware component. The quantification process mainly includes: performing time integration on the CPU's operating status data to generate the CPU's actual instruction execution cycle count integral value C_cpu; performing time integration on the GPU's operating status data to generate the GPU streaming multiprocessor active ratio and running time integral value C_gpu; performing cumulative statistics on the NPU's operating status data to generate the NPU operation count integral value C_npu; performing time integration on the memory controller's operating status data to generate the memory read / write bandwidth real-time throughput cycle integral value C_mem; performing cumulative statistics on the network communication module's operating status data to generate the total network uplink and downlink data traffic C_bw within the task cycle; and performing difference calculation on the temperature sensor's operating status data to generate the difference between the chip's real-time temperature and the ambient reference temperature C_thermal.
[0044] Next, TMA collects the computing power resource consumption corresponding to each of the above hardware components at a fixed sampling interval (e.g., 1 second), and encrypts the computing power resource consumption corresponding to each hardware component with the sampling timestamp, vehicle VIN (Vehicle Identification Number) (de-identified hash), tenant ID (Identifier) (de-identified hash), and task ID, and encapsulates them together into an encrypted computing power proof message, which is signed using the ECDSAsecp256k1 private key built into HSM, and stored in the local secure storage area protected by the vehicle-side trusted execution environment (TEE), and uploaded in batches to the cloud settlement platform and blockchain nodes when the network is available.
[0045] In step S103, in response to the vehicle being in a network-connected state, an encrypted computing power proof message is sent to the target blockchain network, and the settlement is performed based on the computing power resource consumption according to the settlement rules in the target blockchain network to obtain the computing power resource settlement result of the vehicle.
[0046] According to one embodiment of this application, the target blockchain network includes a first blockchain and a second blockchain, wherein the first blockchain is deployed on a roadside communication unit, a mobile edge computing node, and a cloud platform, and is used to write task metadata and verification roots of encrypted computing power proof messages; the second blockchain is used to receive the verification roots of the first blockchain, so as to verify the task metadata according to the verification roots.
[0047] Specifically, vehicle-mounted computing power measurement requires high-frequency sampling (second-level). Uploading all sampled data to the blockchain would incur high gas costs. Furthermore, vehicle networks are prone to intermittent outages (e.g., signal blind spots in parking garages, tunnels), necessitating an evidence storage architecture that supports outage caching and batch uploading upon network reconnection. Therefore, as... Figure 4 As shown, this application designs a two-layer evidence storage architecture of "high-frequency writing to the side chain + final anchoring to the main chain".
[0048] Specifically, the target blockchain network in this application embodiment includes a first blockchain (i.e., a sidechain) and a second blockchain (i.e., a main chain). The sidechain is a lightweight consortium blockchain (e.g., Hyperledger Fabric or a vehicle-mounted sidechain based on Tendermint). The nodes are deployed on roadside communication units, mobile edge computing nodes (e.g., edge MEC (Multi-Access Edge Computing) servers) and cloud platforms (e.g., cloud-based vehicle enterprise settlement platforms). The sidechain undertakes the following functions: (1) High-frequency data writing: When the network is available, the vehicle uploads the encrypted computing power proof message to the sidechain, and the sidechain parses and extracts the task metadata; (2) Batch Merkleization: The sidechain constructs a Merkle tree for the task metadata within a certain time window (e.g., 100 transactions) or a preset number (e.g., up to N transactions), and calculates the verification root, i.e., the Merkle tree root hash; (3) Lightweight consensus: An improved Raft or HotStuff consensus mechanism is adopted to adapt to the vehicle network latency (e.g., 50-500ms) and achieve data consistency between sidechain nodes.
[0049] It should be noted that the sidechain does not store the complete cryptographic power proof message, but only the task metadata and the Merkle tree root hash, in order to reduce storage overhead.
[0050] The main chain is a mature public chain or consortium chain such as Ethereum or Hyperledger Fabric, and mainly undertakes the following functions: (1) Final immutability of evidence: receiving the Merkle root hash periodically anchored by the side chain, writing it into the main chain block, and using the computing power or consensus of the main chain to ensure the final immutability of the data; (2) Tiered on-chain strategy: for low-frequency large-value tasks, such as a single COU ≥ 10,000 (corresponding to more than 4 hours of GPU intensive training), the single on-chain method is adopted to directly submit the computing power proof to the main chain; for high-frequency small-value tasks, such as a single COU < 100, the aggregated batch on-chain method is adopted to accumulate several tasks (e.g., 100 tasks) on the side chain. The Merkle root hash accumulated on the sidechain is submitted to the main chain all at once, which greatly reduces on-chain gas consumption. In other words, when a car owner, tenant, or auditor needs to verify whether the metering data of a certain task is true and has not been tampered with, the verifier first obtains the task metadata and its Merkle path proof (i.e., the sibling node hash sequence) corresponding to the task from the sidechain. Then, the verifier uses the hash value of the task metadata and the Merkle path proof to calculate upwards level by level to obtain the root hash. The calculated root hash is then compared with the publicly stored Merkle root hash on the main chain. If the two are consistent, it proves that the task metadata does indeed belong to the corresponding batch and has not been tampered with.
[0051] According to one embodiment of this application, the settlement of computing resources for a vehicle is obtained by using the settlement rules in the target blockchain network and settling the amount of computing resources consumed. The settlement results include: calling the settlement rules in the target blockchain network, calculating the settlement amount of computing resources based on the amount of computing resources consumed, and generating the computing resources settlement results.
[0052] Specifically, this application solidifies the encrypted computing power proof message into immutable on-chain evidence data and opens up multi-party verification interfaces, enabling car manufacturers, tenants, and auditors to independently complete data verification without relying on a centralized platform. It mainly focuses on the technical aspects of data solidification, signature verification, and traceability and rights confirmation, eliminating all commercial settlement logic and retaining only patentable evidence storage and signature verification technology processes to support trusted data verification by the three parties without trust endorsement.
[0053] Specifically, the on-chain evidence storage triggering in this application embodiment covers three full scenarios: normal task completion, abnormal exit interruption, and cross-node task continuation. It mainly includes: when the computing task on the vehicle is completed normally, the on-chain evidence storage of the encrypted computing power proof message is triggered; when the computing task is abnormally interrupted due to the vehicle leaving the parking position, the on-chain evidence storage of the encrypted computing power proof message executed before the interruption is triggered; when the computing task is migrated to another computing node to continue execution, the on-chain evidence storage of the encrypted computing power proof message executed before the migration is triggered.
[0054] The on-chain evidence data relies on a dual verification mechanism to ensure its immutability: this mainly includes ECDSA (Elliptic Curve Digital Signature Algorithm) signature verification and Merkle root hash verification. The ECDSA signature verification verifies the credibility of the source of the encrypted computing power proof message, confirming that the message was indeed generated by the target vehicle's HSM signature, thus preventing data forgery. The Merkle root hash verification verifies the integrity of the batch of task metadata, confirming that the record does indeed belong to the corresponding batch and has not been tampered with.
[0055] Furthermore, after the encrypted computing power proof message is stored on the blockchain, multi-party independent verification is achieved through the following methods: The data structure stored on the target blockchain network includes at least the task identifier, the Merkle root hash of the batch of task metadata, the total computing power resource consumption, the vehicle identifier hash, the tenant identifier hash, the task start and end time, the heat loss compensation value, and the digital signature of the trusted computing power metering module; the vehicle enterprise node, the tenant node, and the third-party audit node respectively obtain the stored data from the target blockchain network, verify the source credibility of the encrypted computing power proof message through ECDSA signature verification, and then verify the integrity of the batch of task metadata through Merkle root hash verification, thereby achieving multi-party independent verification without relying on a centralized platform, and realizing decentralized trusted rights confirmation.
[0056] Furthermore, the settlement rules of this application are smart contracts deployed in the target blockchain network. After the encrypted computing power proof message is stored on the chain, the smart contract automatically triggers the settlement process. The settlement process includes: reading the computing power resource consumption of each hardware component in the encrypted computing power proof message stored on the chain, calculating the settlement amount according to the tiered fee rate model associated with the vehicle parking time, and distributing the settlement amount to multiple accounts according to the preset revenue sharing ratio to generate the vehicle's computing power resource settlement result. At the same time, the car manufacturer's settlement platform provides a web-based reconciliation dashboard for car owners and tenants. The reconciliation dashboard can query the computing power resource consumption details of each task, the corresponding on-chain transaction hash, and the settlement amount, allowing car owners and tenants to reconcile accounts in real time and achieve transparent accounts.
[0057] Furthermore, when any party—the vehicle owner, tenant, or vehicle manufacturer—submits an appeal against the settlement result, the arbitration node retrieves the original encrypted computing power proof message signed by the trusted computing power metering module from decentralized storage, re-verifies the ECDSA signature, confirms that the message indeed originates from the target vehicle and has not been tampered with, and then independently recalculates the computing power resource consumption according to the COU standard and compares it with the on-chain stored value. The arbitration node verifies the correctness of the computing power resource consumption calculation process using zero-knowledge proofs. During the verification process, the original sensor data and location information of the vehicle are not obtained to protect the vehicle owner's privacy. Finally, the arbitration judgment is written into the target blockchain network, forming an irrefutable final ruling.
[0058] Among them, the multi-party accounts include vehicle smart contract wallets (rather than traditional EOA (Externally Owned Account) wallets). The vehicle smart contract wallets are implemented based on the ERC-4337 (Ethereum Request for Comments) account abstraction standard. The vehicle smart contract wallets support gasless transactions (transaction gas fees are paid by the car manufacturer's platform) and support the SessionKey short-term authorization mechanism, which means that the vehicle can complete multiple settlement confirmations without repeated signatures within the preset authorization period. At the same time, it supports multi-currency settlement of stablecoins, reducing the threshold for users.
[0059] Therefore, based on the complete closed loop described above, all data from each stage leaves an immutable record on the target blockchain, and no party can unilaterally deny or tamper with the measurement and settlement results, fundamentally solving the trust issue among car owners, car manufacturers, and tenants.
[0060] In summary, this application can achieve the following beneficial effects: (1) For the first time, “hardware loss” is included in the billing model: through the heat loss compensation term ζ·C_thermal of COU, the price of computing power transaction reflects the real cost of chip aging and cooling energy consumption, which solves the fairness problem that “hidden loss” in computing power rental is borne unilaterally by the vehicle owner. It is the core breakthrough of the commercialization of vehicle computing power from technical feasibility to economic closed loop. (2) Anti-tampering metering architecture based on TEE + hardware PMC (Performance Monitoring Counter): The metering architecture based on TEE + hardware PMC direct sampling, combined with HSM's ECDSAsecp256k1 hardware signature and AES (Advanced Encryption Standard)-256-GCM (Galois / Counter Mode) hardware encryption, eliminates the possibility of software-level metering forgery from the physical layer, system layer and cryptographic layer. The metering credibility is far superior to the existing cloud service provider's scheme based on the host OS (Operating System) statistics. (3) Blockchain notarization enables mutual trust among the three parties: OEM (Original Equipment Manufacturer), vehicle owner, and tenant do not need to trust any centralized institution. All metering data is automatically executed through on-chain smart contracts, ensuring that bills are non-repudiable and responsibilities are clear. (4) Layered blockchain evidence storage architecture (side chain + main chain) adapted to vehicle intermittent network: The layered chain architecture (side chain batch + main chain summary) and payment channel solution reduce the on-chain transaction frequency by 2-3 orders of magnitude. The side chain undertakes high-frequency writing (generating a computing power proof every 1 second on the vehicle side), and the main chain undertakes the final anchor that is tamper-proof. Through the hierarchical strategy of "large-value tasks are uploaded to the chain in a single transaction and small-value tasks are uploaded to the chain in batches", it ensures efficient and reliable operation under the conditions of vehicle 5G intermittent network (no signal in the underground parking lot - network connection after driving out). It achieves the best balance among security (main chain is tamper-proof), cost (high-frequency writing goes through the side chain), and maturity (Hyperledger / Tendermint has been mass-produced and verified). (5) Promote the marketization and standardization of vehicle computing power: COU, as a unified unit of computing power, is expected to become the industry standard unit of measurement in the field of vehicle edge computing power trading, similar to "vCPU·hour" in the field of cloud computing. (6) Tiered fee rate and multi-party automatic revenue sharing smart contract associated with vehicle parking window: This application embeds a tiered fee rate model (nighttime off-peak × 0.6, daytime peak × 1.0, emergency task × 1.5) linked to parking time (peak / valley / flat) in the smart contract, as well as a multi-party automatic revenue sharing logic of 60% for car owner / 20% for car company / 10% for dispatch / 10% for maintenance. This mechanism encodes the complex multi-dimensional pricing and revenue sharing rules into immutable on-chain code, fundamentally eliminating the dispute risk and operating cost of manual reconciliation in the traditional model; (7) Privacy-protected dispute arbitration mechanism based on ZKP (Zero-Knowledge Proof): This application introduces ZKP technology in the dispute arbitration process. The arbitration node can verify the correctness of the COU measurement calculation based solely on ZKP proof without obtaining the vehicle's original sensor data (GPS (Global Positioning System) location, SOC change curve, temperature sensor matrix). This mechanism protects the privacy of vehicle owners while ensuring the auditability and non-repudiation of the measurement results.
[0061] The vehicle computing resource settlement method according to embodiments of this application establishes a vehicle computing resource measurement standard based on multi-dimensional vehicle computing power indicators. Based on a trusted computing power measurement module, it acquires the operating status of the hardware components corresponding to each vehicle computing power indicator and generates the computing resource consumption for each hardware component. The computing resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message, which is sent to the target blockchain network when the vehicle has network access. The settlement rules in the target blockchain network are used to settle the computing resource consumption, resulting in the vehicle's computing resource settlement result. This solves the technical problem of accurately quantifying, reliably storing, and fairly settling vehicle computing power consumption under the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, manufacturers, and tenants.
[0062] Next, referring to the accompanying drawings, a vehicle computing resource settlement device according to an embodiment of this application is described.
[0063] Figure 5 This is a block diagram of a vehicle computing resource settlement device according to an embodiment of this application.
[0064] like Figure 5 As shown, the computing power resource settlement device 10 of the vehicle includes: a construction module 100, an acquisition module 200 and a settlement module 300.
[0065] Among them, the construction module 100 is used to establish a measurement standard for vehicle computing power resources based on multi-dimensional vehicle computing power indicators. The acquisition module 200 is used to acquire the operating status of the hardware components corresponding to each vehicle computing power index based on the trusted computing power metering module, generate the computing power resource consumption corresponding to each hardware component according to the computing power resource metering standard, and encapsulate the computing power resource consumption and vehicle identifier into an encrypted computing power proof message and store it in the vehicle-side trusted execution environment. The settlement module 300 is used to send an encrypted computing power proof message to the target blockchain network in response to the vehicle being in a networked state, and to use the settlement rules in the target blockchain network to settle the computing power resources based on the consumption of computing power resources, so as to obtain the computing power resource settlement result of the vehicle.
[0066] According to one embodiment of this application, the construction module 100 includes: The first acquisition unit is used to acquire the integral value of the number of CPU actual instruction execution cycles, the integral value of the GPU streaming multiprocessor active ratio and running time, the integral value of the NPU operation count, the period integral value of the real-time throughput of memory read and write bandwidth, the total network uplink and downlink data traffic within the task cycle, and the difference between the chip real-time temperature and the ambient reference temperature. The construction unit is used to establish a vehicle computing resource measurement standard based on the integral value of the actual number of CPU instruction cycles executed within a preset task cycle and the corresponding first weight coefficient, the integral value of the active ratio of GPU streaming multiprocessor and running time and the corresponding second weight coefficient, the integral value of NPU operation count and the corresponding third weight coefficient, the periodic integral value of memory read / write bandwidth real-time throughput and the corresponding fourth weight coefficient, the total network uplink or downlink data traffic within the task cycle and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
[0067] According to one embodiment of this application, the acquisition module 200 includes: The second acquisition unit is used to directly read multi-dimensional vehicle computing power indicators through memory-mapped input and output based on the trusted computing power metering module, and to obtain the operating status of the hardware components corresponding to each vehicle computing power indicator. The generation unit is used to quantify the operating status of hardware components and generate the computing resource consumption corresponding to each hardware component.
[0068] According to one embodiment of this application, the target blockchain network includes a first blockchain and a second blockchain, wherein the first blockchain is deployed on a roadside communication unit, a mobile edge computing node, and a cloud platform, and is used to write task metadata and verification roots of encrypted computing power proof messages; the second blockchain is used to receive the verification roots of the first blockchain, so as to verify the task metadata according to the verification roots.
[0069] According to one embodiment of this application, the settlement module 300 includes: The settlement unit is used to invoke the settlement rules in the target blockchain network, calculate the settlement amount of computing resources based on the amount of computing resources consumed, and generate the settlement result of computing resources.
[0070] The vehicle computing resource settlement device according to the embodiments of this application establishes a vehicle computing resource measurement standard based on multi-dimensional vehicle computing power indicators. Based on a trusted computing power measurement module, it acquires the operating status of the hardware components corresponding to each vehicle computing power indicator and generates the computing resource consumption corresponding to each hardware component. The computing resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message, which is sent to the target blockchain network when the vehicle has network access. The settlement rules in the target blockchain network are used to settle the computing resource consumption, resulting in the vehicle's computing resource settlement result. This solves the technical problem of accurately quantifying, reliably storing, and fairly settling vehicle computing power consumption under the combined constraints of unstable vehicle networks, uncertain vehicle parking status, and mutual distrust among vehicle owners, manufacturers, and tenants.
[0071] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 601, the processor 602, and the computer program stored on the memory 601 and capable of running on the processor 602.
[0072] When the processor 602 executes the program, it implements the vehicle computing resource settlement method provided in the above embodiments.
[0073] Furthermore, electronic devices also include: Communication interface 603 is used for communication between memory 601 and processor 602.
[0074] The memory 601 is used to store computer programs that can run on the processor 602.
[0075] The memory 601 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.
[0076] If the memory 601, processor 602, and communication interface 603 are implemented independently, then the communication interface 603, memory 601, and processor 602 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0077] Optionally, in a specific implementation, if the memory 601, processor 602, and communication interface 603 are integrated on a single chip, then the memory 601, processor 602, and communication interface 603 can communicate with each other through an internal interface.
[0078] The processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.
[0079] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for calculating vehicle computing resources.
[0080] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for calculating vehicle computing resources.
[0081] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0083] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0084] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0085] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0086] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium. When executed, the program includes one or a combination of the steps of the method embodiments.
[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0088] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A method for settling computing power resources for vehicles, characterized in that, Includes the following steps: Based on multi-dimensional vehicle-mounted computing power metrics, a computing power resource measurement standard for the vehicle is established. Based on the trusted computing power metering module, the operating status of the hardware components corresponding to each vehicle computing power index is obtained, and the computing power resource consumption corresponding to each hardware component is generated according to the computing power resource metering standard. The computing power resource consumption and vehicle identifier are encapsulated into an encrypted computing power proof message and stored in the vehicle-side trusted execution environment. In response to the vehicle being connected to the network, the encrypted computing power proof message is sent to the target blockchain network, and the settlement is performed according to the computing power resource consumption based on the settlement rules in the target blockchain network to obtain the computing power resource settlement result of the vehicle.
2. The method according to claim 1, characterized in that, The establishment of a computing resource measurement standard for the vehicle based on multi-dimensional vehicle-mounted computing power metrics includes: The system obtains the integral value of the number of CPU instruction cycles actually executed within a preset task cycle, the integral value of the GPU streaming multiprocessor activity ratio and running time, the integral value of the NPU operation count, the periodic integral value of the real-time throughput of memory read and write bandwidth, the total network uplink and downlink data traffic within the task cycle, and the difference between the chip's real-time temperature and the ambient reference temperature. The computing resource measurement standard for the vehicle is established based on the integral value of the actual number of CPU instruction execution cycles within the preset task cycle and the corresponding first weight coefficient, the integral value of the GPU streaming multiprocessor activity ratio and running time and the corresponding second weight coefficient, the integral value of the NPU operation count and the corresponding third weight coefficient, the periodic integral value of the memory read / write bandwidth real-time throughput and the corresponding fourth weight coefficient, the total network uplink or downlink data traffic within the task cycle and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
3. The method according to claim 1, characterized in that, The step of obtaining the operating status of the hardware components corresponding to each vehicle-mounted computing power index and generating the computing power resource consumption corresponding to each hardware component according to the computing power resource metering standard includes: Based on the trusted computing power measurement module, the multi-dimensional vehicle computing power indicators are directly read through memory-mapped input and output, and the operating status of the hardware components corresponding to each vehicle computing power indicator is obtained. The operating status of the hardware components is quantified to generate the computing resource consumption corresponding to each hardware component.
4. The method according to claim 1, characterized in that, The target blockchain network includes a first blockchain and a second blockchain. The first blockchain is deployed on a roadside communication unit, a mobile edge computing node, and a cloud platform to write task metadata and verification roots into the encrypted computing power proof message. The second blockchain is used to receive the verification roots from the first blockchain to verify the task metadata based on the verification roots.
5. The method according to claim 1, characterized in that, The step of using the settlement rules in the target blockchain network to settle accounts based on the consumption of computing resources, and obtaining the computing resource settlement result for the vehicle, includes: The settlement rules in the target blockchain network are invoked to calculate the settlement amount of the computing power resources based on the amount of computing power resources consumed, and the settlement result of the computing power resources is generated.
6. A computing resource settlement device for vehicles, characterized in that, include: A construction module is used to establish a measurement standard for the computing power resources of the vehicle based on multi-dimensional vehicle-mounted computing power metrics. The acquisition module is used to acquire the operating status of the hardware components corresponding to each vehicle computing power index based on the trusted computing power metering module, generate the computing power resource consumption corresponding to each hardware component according to the computing power resource metering standard, and encapsulate the computing power resource consumption and vehicle identifier into an encrypted computing power proof message and store it in the vehicle-side trusted execution environment. The settlement module is used to respond to the vehicle being in a network-connected state by sending the encrypted computing power proof message to the target blockchain network, and using the settlement rules in the target blockchain network to settle the computing power resource consumption, thereby obtaining the computing power resource settlement result of the vehicle.
7. The apparatus according to claim 6, characterized in that, The building module includes: The acquisition unit is used to acquire the integral value of the number of CPU instruction cycles actually executed, the integral value of the GPU streaming multiprocessor active percentage and running time, the integral value of the NPU operation count, the period integral value of the real-time throughput of memory read and write bandwidth, the total network uplink and downlink data traffic within the task cycle, and the difference between the chip real-time temperature and the ambient reference temperature. The construction unit is used to establish the computing resource measurement standard for the vehicle based on the integral value of the actual number of CPU instruction execution cycles within the preset task cycle and the corresponding first weight coefficient, the integral value of the active ratio of the GPU streaming multiprocessor and the running time and the corresponding second weight coefficient, the integral value of the NPU operation count and the corresponding third weight coefficient, the periodic integral value of the real-time throughput of memory read and write bandwidth and the corresponding fourth weight coefficient, the total uplink or downlink data traffic of the network within the task cycle and the corresponding fifth weight coefficient, and the difference between the chip real-time temperature and the ambient reference temperature and the corresponding sixth weight coefficient.
8. The apparatus according to claim 6, characterized in that, The acquisition module includes: The acquisition unit is used to directly read the multi-dimensional vehicle-mounted computing power indicators through memory-mapped input and output based on the trusted computing power measurement module, and to acquire the operating status of the hardware components corresponding to each vehicle-mounted computing power indicator. The generation unit is used to quantify the operating status of the hardware components and generate the computing resource consumption corresponding to each hardware component.
9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, the processor executing the program to implement the vehicle computing resource settlement method as described in any one of claims 1-5.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the computing resource settlement method for vehicles as described in any one of claims 1-5.
