Edge processing method and device, equipment and storage medium

Abstract

The invention provides an edge processing method and device, equipment and a storage medium, relates to the technical field of communication, and is used for realizing edge credible verification of an artificial intelligence model so as to safely realize a service information processing process in a low-delay manner, avoid a service data leakage risk and effectively protect an artificial intelligence model application program. The method is applied to business side equipment, and comprises the following steps: sending a service request to edge computing equipment; the service request is used for requesting an information processing service of an artificial intelligence model on the edge computing device; receiving remote attestation information from the edge computing device; the remote certification information is certification information of a trusted execution environment instance which is registered in a remote certification server in advance and is deployed with an artificial intelligence model; sending a first attestation request to a remote attestation server; and sending an information processing request carrying the to-be-processed information to the edge computing equipment under the condition that a first passing response from the remote attestation server is received.

Description

Technical Field

[0001] The present application belongs to the field of communication technology, and in particular, relates to an edge processing method, device, equipment and storage medium. Background Art

[0002] With the rapid development of the fifth generation mobile communication technology (5G), multi-access edge computing (MEC) has been widely used. In addition, a large number of artificial intelligence (AI) models are deployed on MEC servers to perform reasoning services such as product quality inspection based on image recognition and traffic violation recognition, in order to reduce the reasoning response delay and reduce the data transmission bandwidth.

[0003] However, MEC servers are deployed at the edge of the network, and their infrastructure and usage rights belong to different owners. The network environment is complex, and there is a risk of privacy leakage for businesses involving sensitive data such as face recognition, location data, and production data. In addition, it is necessary to consider protecting third-party AI model applications deployed in the MEC environment. Summary of the invention

[0004] The present application provides an edge processing method, apparatus, device and storage medium for implementing edge trusted verification of artificial intelligence models, so as to implement business information processing processes safely and with low latency, avoid the risk of business data leakage, and effectively protect artificial intelligence model applications.

[0005] In order to achieve the above purpose, this application adopts the following technical solutions:

[0006] In a first aspect, an edge processing method is provided, which is applied to a business party device, including: sending a service request to an edge computing device. The service request is used to request an information processing service of an artificial intelligence model on the edge computing device. Receive remote attestation information from the edge computing device. The remote attestation information is the attestation information of a trusted execution environment instance deployed with an artificial intelligence model that is pre-registered in a remote attestation server. Send a first attestation request to the remote attestation server. The first attestation request is used to request verification of the remote attestation information. In the case of receiving a first passing response from the remote attestation server, send an information processing request carrying information to be processed to the edge computing device. The information processing request is used to request the artificial intelligence model to process the information to be processed.

[0007] Optionally, the remote attestation information includes the identifier of the trusted execution environment instance where the artificial intelligence model is deployed, the activation certificate of the trusted execution environment instance, and the private key signature corresponding to the activation certificate, the trusted execution environment software version number, the processor identifier, and the trusted computing base information.

[0008] Optionally, the method of sending an information processing request carrying information to be processed to an edge computing device includes: establishing a transport layer security protocol connection with the edge computing device based on an activation certificate of a trusted execution environment instance, and sending the information to be processed to the edge computing device through the transport layer security protocol connection.

[0009] Optionally, after sending an information processing request carrying information to be processed to the edge computing device, the method further includes: receiving an information processing result from the edge computing device.

[0010] Optionally, the artificial intelligence model is deployed on the edge computing device when the operator device determines that the remote attestation server has passed the verification of the remote attestation information. The operator device includes an operation support system, a multi-access edge computing orchestrator, multiple mobile edge platform managers, and a virtualized infrastructure management module. The operation support system is used to respond to the first model instantiation request from the service-side device and send a second model instantiation request to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is used to respond to the second model instantiation request, determine the edge computing device that meets the preset conditions, and send a third model instantiation request to the target mobile edge platform manager corresponding to the edge computing device in the multiple mobile edge platform managers. The preset conditions include having instantiation resources corresponding to the artificial intelligence model and having trusted execution environment capabilities. The target mobile edge platform manager is used to respond to the third model instantiation request and send a fourth model instantiation request to the virtualized infrastructure management module. The virtualized infrastructure management module is used to respond to the fourth model instantiation request, determine the instantiation resources of the artificial intelligence model on the edge computing device, and the remote attestation information of the trusted execution environment instance corresponding to the instantiation resources, and send a second attestation request to the remote attestation server. The second attestation request is used to request verification of whether the trusted execution environment instance is trustworthy. The instantiated resources include computing resources, storage resources, network resources, and trusted execution environment resources. The virtualization infrastructure management module is also used to instantiate the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model when receiving the second pass response from the remote attestation server.

[0011] Optionally, the operator device also includes a mobile edge platform. The mobile edge platform manager is also used to send a service configuration request to the edge computing platform. The edge computing platform is used to respond to the service configuration request, obtain remote attestation information, and send a third attestation request to the remote attestation server. The third attestation request is used to verify whether the trusted execution environment instance is trustworthy. The edge computing platform is also used to determine the configuration information of the artificial intelligence model when receiving a third pass response from the remote attestation server. The configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

[0012] Optionally, the edge computing platform is further used to send a service configuration response to the target mobile edge platform manager. The target mobile edge platform manager is further used to send a first model instantiation response to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is further used to send a second model instantiation response to the service party device through the operation support system.

[0013] In a second aspect, an edge processing device is provided, which is applied to a business party device, and includes: a sending module and a receiving module. The sending module is used to send a service request to the edge computing device. The service request is used to request the information processing service of the artificial intelligence model on the edge computing device. The receiving module is used to receive remote attestation information from the edge computing device. The remote attestation information is the attestation information of the trusted execution environment instance deployed with the artificial intelligence model, which is pre-registered in the remote attestation server. The sending module is also used to send a first attestation request to the remote attestation server. The first attestation request is used to request verification of the remote attestation information. The sending module is also used to send an information processing request carrying the information to be processed to the edge computing device when the receiving module receives a first passing response from the remote attestation server. The information processing request is used to request the artificial intelligence model to process the information to be processed.

[0014] Optionally, the remote attestation information includes the identifier of the trusted execution environment instance where the artificial intelligence model is deployed, the activation certificate of the trusted execution environment instance, and the private key signature corresponding to the activation certificate, the trusted execution environment software version number, the processor identifier, and the trusted computing base information.

[0015] Optionally, the sending module is specifically configured to: establish a transport layer security protocol connection with the edge computing device based on the activation certificate of the trusted execution environment instance, and send the information to be processed to the edge computing device through the transport layer security protocol connection.

[0016] Optionally, the receiving module is also used to receive information processing results from the edge computing device.

[0017] Optionally, the artificial intelligence model is deployed on the edge computing device when the operator device determines that the remote attestation server has passed the verification of the remote attestation information. The operator device includes an operation support system, a multi-access edge computing orchestrator, multiple mobile edge platform managers, and a virtualized infrastructure management module. The operation support system is used to respond to the first model instantiation request from the service-side device and send a second model instantiation request to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is used to respond to the second model instantiation request, determine the edge computing device that meets the preset conditions, and send a third model instantiation request to the target mobile edge platform manager corresponding to the edge computing device in the multiple mobile edge platform managers. The preset conditions include having instantiation resources corresponding to the artificial intelligence model and having trusted execution environment capabilities. The target mobile edge platform manager is used to respond to the third model instantiation request and send a fourth model instantiation request to the virtualized infrastructure management module. The virtualized infrastructure management module is used to respond to the fourth model instantiation request, determine the instantiation resources of the artificial intelligence model on the edge computing device, and the remote attestation information of the trusted execution environment instance corresponding to the instantiation resources, and send a second attestation request to the remote attestation server. The second attestation request is used to request verification of whether the trusted execution environment instance is trustworthy. The instantiated resources include computing resources, storage resources, network resources, and trusted execution environment resources. The virtualization infrastructure management module is also used to instantiate the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model when receiving the second pass response from the remote attestation server.

[0018] Optionally, the operator device also includes a mobile edge platform. The mobile edge platform manager is also used to send a service configuration request to the edge computing platform. The edge computing platform is used to respond to the service configuration request, obtain remote attestation information, and send a third attestation request to the remote attestation server. The third attestation request is used to verify whether the trusted execution environment instance is trustworthy. The edge computing platform is also used to determine the configuration information of the artificial intelligence model when receiving a third pass response from the remote attestation server. The configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

[0019] Optionally, the edge computing platform is further used to send a service configuration response to the target mobile edge platform manager. The target mobile edge platform manager is further used to send a first model instantiation response to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is further used to send a second model instantiation response to the service party device through the operation support system.

[0020] In a third aspect, a computer device is provided, comprising a memory and a processor; the memory is used to store computer execution instructions, and the processor is connected to the memory through a bus; when the computer device is running, the processor executes the computer execution instructions stored in the memory, so that the base station executes any optional edge processing method as described in the first aspect.

[0021] The computer device may be a network device, or a part of a network device, such as a chip system in a network device. The chip system is used to support the network device to implement the functions involved in the first aspect and any possible implementation thereof, for example, receiving, determining, and diverting the data and / or information involved in the above-mentioned edge processing method. The chip system includes a chip, and may also include other discrete devices or circuit structures.

[0022] In a fourth aspect, a computer-readable storage medium is provided, the computer-readable storage medium comprising computer execution instructions, and when the computer execution instructions are executed on a base station, the base station executes any optional edge processing method as described in the first aspect.

[0023] It should be noted that the above computer instructions may be stored in whole or in part on a first computer-readable storage medium, wherein the first computer-readable storage medium may be packaged together with the processor of the routing device, or may be packaged separately from the processor of the routing device, which is not limited in this application.

[0024] In this application, the name of the above-mentioned routing device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0025] These and other aspects of the present application will become more apparent from the following description.

[0026] The technical solution provided by this application brings at least the following beneficial effects:

[0027] Based on any of the above aspects, in this application, before sending the information to be processed to the edge computing device, the business party device can first obtain the remote attestation information of the trusted execution environment instance where the artificial intelligence model is deployed from the edge computing device, and verify the remote attestation information through interaction with the remote attestation server to achieve edge trusted verification of the artificial intelligence model. Furthermore, if the remote attestation information is verified, the business party device can send the information to be processed to the edge computing device for processing, so as to implement the business information processing process safely and with low latency, and avoid the risk of business data leakage. In addition, the artificial intelligence model is deployed in a trusted execution environment instance, which can effectively protect the artificial intelligence model application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the structure of an edge processing system provided in an embodiment of the present application;

[0029] Figure 2 A schematic diagram of the hardware structure of a service party device provided in an embodiment of the present application;

[0030] Figure 3 A schematic diagram of a flow chart of an edge processing method provided in an embodiment of the present application;

[0031] Figure 4 A deployment flow chart of an artificial intelligence model provided in an embodiment of the present application;

[0032] Figure 5 A schematic diagram of a flow chart of another edge processing method provided in an embodiment of the present application;

[0033] Figure 6 A schematic diagram of a flow chart of another edge processing method provided in an embodiment of the present application;

[0034] Figure 7 A schematic diagram of a flow chart of another edge processing method provided in an embodiment of the present application;

[0035] Figure 8 A schematic diagram of the structure of an edge processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0038] In order to facilitate a clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, words such as "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. Those skilled in the art can understand that words such as "first" and "second" do not limit the quantity and order of execution.

[0039] In addition, the terms "including" and "having" in the embodiments, claims and drawings of the present application are not exclusive. For example, a process, method, system, product or device including a series of steps or modules is not limited to the listed steps or modules, and may also include steps or modules that are not listed.

[0040] In order to facilitate the understanding of the present application, the relevant elements involved in the present application are now described.

[0041] 1. Confidential computing technology-trusted execution environment (TEE).

[0042] A trusted execution environment is an execution environment that requires authorized program code to execute within the processor, and the data used by the program code cannot be read or tampered with by code outside the TEE.

[0043] TEE is a secure area within a central processing unit (CPU) or graphics processing unit (GPU) that runs in an independent environment and in parallel with the operating system. The CPU or GPU ensures that the confidentiality and integrity of the code and data in the TEE are protected. Trusted applications running in the TEE can access the full functionality of the device's main processor and memory, while hardware isolation protects these applications from other applications running in the main operating system. Therefore, some data-sensitive data processing services can be performed in the TEE. For example, data processing services such as secret key generation and storage, payment account authentication, privacy data processing, and algorithm protection. Optionally, the hardware corresponding to the TEE can be deployed in a general-purpose X86 server or in some network devices.

[0044] 2. Multi-access edge computing (MEC).

[0045] MEC is a system that provides network services and cloud computing capabilities at the edge of the access network that includes one or more access technologies. MEC has the advantage of being close to users and can better support data processing services with low latency and high privacy protection requirements. MEC receives user-side requirements through the MEC orchestrator (MEC orchestrator, MEO) network element, and uniformly orchestrates MEC infrastructure as a service (IaaS) resources to manage the life cycle of MEC applications (applications, APPs). The virtualization infrastructure management module (virtualizationinfrastructure manager, VIM) is used to manage the IaaS resources of the local data center, including resource reporting and resource allocation, as well as virtual machine (VM) and Docker pull-up, logout and other operations.

[0046] 3. Artificial intelligence (AI) reasoning.

[0047] In machine learning, AI inference is the process of using a trained model to run live data to make predictions or solve tasks. Inference is the process by which an AI model generates output by applying its training data knowledge to previously unseen data. The goal of AI inference is to compute and output actionable results.

[0048] Services after various AI model training, such as image recognition, video recognition, voice recognition, and judgment and decision-making, are ultimately provided in the form of AI reasoning services, such as traffic violation recognition, product quality inspection, and autonomous driving.

[0049] The following is a brief introduction to the application background of this application.

[0050] With the deployment of 5G, especially the deployment of 5G private networks, the demand for data processing at the edge of the network has increased. On the one hand, there are low latency requirements for specific services, requiring services to provide agile and rapid responses. On the other hand, there are data management policy requirements, which require business data to be processed within the park. MEC can provide network services and cloud computing capabilities in data centers close to the edge of the network, and is gradually being used by video and data services (over the top, OTT) and operators to build the necessary infrastructure for the edge business ecosystem.

[0051] As AI technology matures, a large number of AI models are deployed in MEC to perform reasoning tasks, such as product quality inspection based on image recognition in production parks, traffic violation identification, and autonomous driving, etc., to reduce the latency of reasoning response and the bandwidth required to upload data to the cloud. Edge reasoning technology improves performance and reduces dependence on network connections by shortening the time from input data to reasoning decisions, ultimately increasing business profits. Since AI model training is costly and requires a lot of computing resources, some small and medium-sized enterprises tend to adopt AI model algorithms that have been trained by third parties, and the third-party AI model algorithm provider is responsible for the maintenance and update of the AI ​​model algorithm on the MEC side.

[0052] However, MEC is deployed at the edge of the network, the environment is complex, and the infrastructure and applications belong to different owners. There is a risk of privacy leakage for businesses involving sensitive data such as face recognition, location data, and production data. In addition, it is necessary to consider protecting the inference algorithm of third-party AI models deployed in the MEC environment.

[0053] In response to the above problems, the embodiment of the present application provides an edge processing method. Before sending the information to be processed to the edge computing device, the business party device can first obtain the remote certification information of the trusted execution environment instance where the artificial intelligence model is deployed from the edge computing device, and verify the remote certification information through interaction with the remote certification server to achieve edge trusted verification of the artificial intelligence model. Furthermore, if the remote certification information is verified, the business party device can send the information to be processed to the edge computing device for processing, so as to implement the business information processing process safely and with low latency, and avoid the risk of business data leakage.

[0054] In addition, confidential computing based on hardware TEE is believed to be able to achieve a scalable solution where data is available but not accessible. By deploying AI models in TEE instances of edge computing infrastructure, sensitive data is input into TEE in ciphertext form to complete AI reasoning, and the reasoning results are output, thus effectively protecting artificial intelligence model applications.

[0055] In this way, edge inference based on confidential computing can protect the privacy of business information on the one hand, and protect artificial intelligence model applications on the other hand, including tenants of edge computing devices, system administrators, and infrastructure providers cannot obtain or change artificial intelligence model applications.

[0056] This edge processing method is applicable to business side devices in edge processing systems. Figure 1, which is a schematic diagram of the structure of an edge processing system provided in an embodiment of the present application. The edge processing system may include: a business party device 101, an edge computing device 102, and a remote certification server 103. The business party device 101 may be connected to the edge computing device 102 and the remote certification server 103, respectively.

[0057] In practical applications, there may be multiple business party devices 101, edge computing devices 102, and remote attestation servers 103 in the edge processing system. For ease of description, this application takes the case where there are only one business party device 101, edge computing device 102, and remote attestation server 103 as an example.

[0058] The business side device 101 may be a device for managing services such as product quality inspection, traffic violation identification or autonomous driving. The business side device 101 may send the image waiting to be processed information related to these services to the edge computing device 102 to obtain the information processing result of the edge computing device 102 for the information to be processed, thereby realizing services such as product quality inspection, traffic violation identification or autonomous driving.

[0059] Optionally, the service party device 101 may be a terminal or a server.

[0060] The edge computing device 102 is an edge infrastructure belonging to the operator, which can be used to provide virtual resources or physical host resources, and can have confidential computing capabilities based on hardware TEE, and can support the deployment of artificial intelligence models in TEE instances. Hardware TEE can be implemented based on CPU and GPU.

[0061] Optionally, the edge computing device 102 may be a physical device (such as a server) or a virtual device deployed on a physical device.

[0062] The remote attestation server 103 is used to provide remote attestation services. The remote attestation service is used to verify whether the trusted execution environment provided by the MEC operator is authentic and trustworthy. Optionally, the remote attestation service can be a remote attestation and verification service for TEE instances provided by a CPU or GPU manufacturer, or a local remote attestation and verification service agent for a data center provided by an infrastructure provider, or a local remote attestation and verification service agent provided by an MEC operator.

[0063] Optionally, the server involved in the present application may be a single server, or may be a server cluster composed of multiple servers. In some implementations, the server cluster may also be a distributed cluster. The embodiments of the present application do not impose any limitation on this.

[0064] The terminal involved in the present application may be a user equipment (UE), access terminal, terminal unit, user terminal equipment (TE), mobile device, wireless communication device, terminal agent, tablet computer (pad), handheld device with wireless communication function, computing device or other processing device connected to a wireless modem, wearable device, or terminal device, other processing device connected to a wireless modem in a 5G network or a public land mobile network (PLMN) evolved after 5G. In addition, the terminal may be mobile or fixed. The embodiments of the present application are not limited to this.

[0065] like Figure 2 , which is a schematic diagram of the hardware structure of a business party device provided in an embodiment of the present application. The business party device includes a processor 21, a memory 22, a communication interface 23, and a bus 24. The processor 21, the memory 22, and the communication interface 23 can be connected through a bus 24.

[0066] The processor 21 is the control center of the service party device, which can be a processor or a general term for multiple processing elements. For example, the processor 21 can be a CPU or other general-purpose processors. Among them, the general-purpose processor can be a microprocessor or any conventional processor.

[0067] As an embodiment, the processor 21 may include one or more CPUs, such as Figure 2 CPU0 and CPU1 are shown in the figure.

[0068] The memory 22 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0069] In one possible implementation, the memory 22 may exist independently of the processor 21, and the memory 22 may be connected to the processor 21 via a bus 24 to store instructions or program codes. When the processor 21 calls and executes the instructions or program codes stored in the memory 22, the edge processing method provided in the following embodiments of the present application can be implemented.

[0070] In another possible implementation, the memory 22 may also be integrated with the processor 21 .

[0071] The communication interface 23 is used for connecting the service party device to other devices through a communication network, and the communication network may be Ethernet, wireless access network, wireless local area network (WLAN), etc. The communication interface 23 may include a receiving unit for receiving data and a sending unit for sending data.

[0072] The bus 24 may be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0073] It should be pointed out that Figure 2 The structure shown in the figure does not constitute a limitation on the business party equipment, except Figure 2 In addition to the components shown, the business party device may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0074] like Figure 3 FIG. 1 is a flow chart of an edge processing method provided in an embodiment of the present application. The edge processing method can be applied to Figure 1 or Figure 2 The business side device shown. The edge processing method includes: S301-S304.

[0075] S301. The business device sends a service request to the edge computing device.

[0076] The service request is used to request an information processing service of an artificial intelligence model on an edge computing device. The service request may include identification information for identifying the artificial intelligence model.

[0077] Optionally, the artificial intelligence model can be a pre-trained machine learning model for processing data of types such as video data, image data, audio data, or text data, or it can be a non-machine learning model.

[0078] Optionally, the information processing service may be a service for identifying workshop product image data, or a service for identifying traffic intersection image data, etc. The embodiment of the present application does not limit this.

[0079] When the business party device needs to process the information to be processed, it can send a service request to the edge computing device. Accordingly, the edge computing device can receive the service request from the business party device and read the certification information of the trusted execution environment instance where the artificial intelligence model is deployed. Furthermore, the edge computing device can send the read certification information to the business party device so that the business party device can verify whether the deployment environment of the artificial intelligence model is trustworthy, thereby improving the security of the information processed by the artificial intelligence model.

[0080] It should be noted that the artificial intelligence model is deployed on the edge computing device when the operator equipment determines that the remote attestation server has passed the verification of the remote attestation information.

[0081] Among them, in addition to edge computing devices, operator equipment also includes operation support systems (OSS), multi-access edge computing orchestrators (MEO), multiple mobile edge platform managers (MEplatform managers, MEPM), and virtualized infrastructure management modules.

[0082] The operation support system is used to send a second model instantiation request to the multi-access edge computing orchestrator in response to a first model instantiation request from a service-side device.

[0083] The multi-access edge computing orchestrator is used to respond to the second model instantiation request, determine an edge computing device that meets the preset conditions, and send a third model instantiation request to a target mobile edge platform manager corresponding to the edge computing device among the multiple mobile edge platform managers. The preset conditions include having instantiation resources corresponding to the artificial intelligence model and having trusted execution environment capabilities.

[0084] The target mobile edge platform manager is used to send a fourth model instantiation request to the virtualization infrastructure management module in response to the third model instantiation request.

[0085] The virtualization infrastructure management module is used to respond to the fourth model instantiation request, determine the instantiation resources of the artificial intelligence model on the edge computing device, and the remote attestation information of the trusted execution environment instance corresponding to the instantiation resources, and send a second attestation request to the remote attestation server. The second attestation request is used to request verification of whether the trusted execution environment instance is trustworthy. The instantiation resources include computing resources, storage resources, network resources, and trusted execution environment resources.

[0086] The virtualization infrastructure management module is also used to instantiate the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model when a second pass response is received from the remote attestation server.

[0087] Or further, the operator equipment may also include a mobile edge platform (ME platform, MEP). The mobile edge platform manager is also used to send a service configuration request to the edge computing platform. The edge computing platform is used to respond to the service configuration request, obtain remote attestation information, and send a third attestation request to the remote attestation server. The third attestation request is used to verify whether the trusted execution environment instance is trustworthy. The edge computing platform is also used to determine the configuration information of the artificial intelligence model when a third pass response is received from the remote attestation server. The configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

[0088] Or further, the edge computing platform is further used to send a service configuration response to the target mobile edge platform manager. The target mobile edge platform manager is further used to send a first model instantiation response to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is further used to send a second model instantiation response to the service side device through the operation support system.

[0089] It should be understood that the above-mentioned operator equipment includes edge computing devices, operation support systems, multi-access edge computing orchestrators, mobile edge platform managers, virtualized infrastructure management modules, and mobile edge platforms that can be deployed on the same physical device or on different physical devices, and the embodiments of the present application do not limit this.

[0090] like Figure 4 As shown, it is a deployment flow chart of an artificial intelligence model provided in an embodiment of the present application. Figure 4 The process shown for pre-deploying an artificial intelligence model on an edge computing device includes: S1-S10.

[0091] S1. The service-side device sends a first model instantiation request to the operation support system.

[0092] The first model instantiation request is used to request to instantiate a virtual container or VM on an edge computing device, and to instantiate an application of an artificial intelligence model in the virtual container or VM.

[0093] S2. The operation support system sends a second model instantiation request to the multi-access edge computing orchestrator.

[0094] S3. The multi-access edge computing orchestrator processes the second model instantiation request and sends a third model instantiation request to the target mobile edge platform manager.

[0095] Among them, the preset conditions include having instantiated resources corresponding to the artificial intelligence model and having trusted execution environment capabilities.

[0096] The multi-access edge computing orchestrator can determine, in response to the second model instantiation request, an edge computing device that meets the preset conditions, and a target mobile edge platform manager corresponding to the edge computing device from among the multiple mobile edge platform managers. Specifically, the multi-access edge computing orchestrator can identify the configuration of the application package of the artificial intelligence model, and select an edge computing device and an associated MEPM that meet the preset conditions based on the resource requirements and TEE capability requirements in its description file.

[0097] S4. The target mobile edge platform manager sends a fourth model instantiation request to the virtualization infrastructure management module.

[0098] The virtualized infrastructure management module is used to manage edge computing devices with TEE capabilities.

[0099] The fourth model instantiation request is used to request the allocation of instantiation resources and instantiate the artificial intelligence model, and may carry the image information (such as the image address) of the artificial intelligence model application.

[0100] S5. The virtualization infrastructure management module determines the instantiated resources of the artificial intelligence model on the edge computing device, as well as the remote attestation information of the trusted execution environment instance corresponding to the instantiated resources, and sends a second attestation request to the remote attestation server.

[0101] The second certification request is used to request verification of whether the trusted execution environment instance is trustworthy. Instantiated resources include computing resources, storage resources, network resources, and trusted execution environment resources.

[0102] S6. Upon receiving the second pass response from the remote attestation server, the virtualization infrastructure management module instantiates the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model, and sends an instantiation response to the target mobile edge platform manager.

[0103] The virtualization infrastructure management module can allocate corresponding resources on the edge computing device to create a container or VM instance. In addition, the virtualization infrastructure management module can obtain remote attestation information from the TEE instance environment allocated on the edge computing device and initiate remote attestation verification to the remote attestation service. If the remote attestation verification passes, the virtualization infrastructure management module can download the artificial intelligence model software image and instantiate the artificial intelligence model in the container or VM.

[0104] S7. The mobile edge platform manager sends a service configuration request to the edge computing platform.

[0105] S8. The edge computing platform obtains remote certification information from the edge computing device and sends a third certification request to the remote certification server.

[0106] The third attestation request is used to verify whether the trusted execution environment instance is trustworthy.

[0107] S9. Upon receiving the third pass response from the remote attestation server, the edge computing platform determines the configuration information of the artificial intelligence model and sends a service configuration response to the target mobile edge platform manager.

[0108] Among them, the configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

[0109] S10. The target mobile edge platform manager sends a model instantiation response to the service-side device through the multi-access edge computing orchestrator and the operation support system.

[0110] Based on the above process, this application can realize the trusted deployment of artificial intelligence models in edge computing devices through the interaction between the operator's equipment and the remote attestation server, and can effectively determine the trusted deployment environment. In this way, the artificial intelligence model can be instantiated in the trusted execution environment instance of the edge computing device to effectively protect the service party's artificial intelligence model application.

[0111] S302: The business party device receives remote certification information from the edge computing device.

[0112] Among them, the remote attestation information is the attestation information of the trusted execution environment instance deployed with the artificial intelligence model, which is pre-registered in the remote attestation server.

[0113] Remote attestation information includes the identifier of the trusted execution environment instance where the artificial intelligence model is deployed, the activation certificate of the trusted execution environment instance, the private key signature corresponding to the activation certificate, the trusted execution environment software version number, the processor identifier, and the trusted computing base information. The processor identifier can be a CPU identifier or a GPU identifier.

[0114] S303: The business party device sends a first certification request to the remote certification server.

[0115] The first attestation request includes remote attestation information of the trusted execution environment instance, and is used to request verification of the remote attestation information to determine whether the trusted execution environment instance is trustworthy.

[0116] S304: When the business party device receives the first passing response from the remote attestation server, it sends an information processing request carrying the information to be processed to the edge computing device.

[0117] Among them, the information processing request is used to request the artificial intelligence model to process the information to be processed.

[0118] The first passing response can be used to indicate that the remote attestation information verification has passed, and the trusted execution environment instance where the artificial intelligence model is deployed is trustworthy.

[0119] Optionally, the information to be processed may be video data, image data, audio data, text data, etc.

[0120] If the first passing response is received from the remote proof server, it can indicate that the information has a high level of security during the processing of the artificial intelligence model. Then the business party device can send an information processing request carrying the information to be processed to the edge computing device. Furthermore, the edge computing device can process the information to be processed through the artificial intelligence model, obtain the information processing result, and return the information processing result to the business party device.

[0121] Based on the above embodiments, it can be seen that the present application can realize trusted verification of the artificial intelligence model operating environment through the interaction between the business party equipment and the edge computing equipment and the remote certification server, so as to realize the business information processing process safely and with low latency and avoid the risk of business information leakage.

[0122] In a possible embodiment, Figure 5 FIG. 1 is a flow chart of another edge processing method provided in an embodiment of the present application. Figure 3 In the above S304, that is, when the business party device sends an information processing request carrying information to be processed to the edge computing device, an embodiment of the present application provides an optional implementation method, including: S3041-S3042.

[0123] S3041. The business party device establishes a transport layer security protocol connection with the edge computing device based on the activation certificate of the trusted execution environment instance.

[0124] That is to say, the business party device establishes a transport layer security protocol connection with the artificial intelligence model on the edge computing device based on the activation certificate of the trusted execution environment instance.

[0125] S3042. The business party device sends the information to be processed to the edge computing device through the transport layer security protocol connection.

[0126] Based on this, the activation certificate of the TEE instance can be used as a digital certificate authority (CA) certificate between the business party device and the edge computing device to establish a transport layer security protocol (TLS) connection to achieve the transmission of information to be processed.

[0127] In a possible embodiment, Figure 6 FIG. 1 is a flow chart of another edge processing method provided in an embodiment of the present application. Figure 3 , after the above S304, the edge processing method provided in the embodiment of the present application further includes: S305.

[0128] S305. The business party device receives the information processing result from the edge computing device.

[0129] The edge computing device can process the information to be processed based on the artificial intelligence model deployed in the trusted execution environment instance to obtain the information processing results. In addition, the edge computing device can send the information processing results to the business party device so that the business party device can implement the management business of the workshop products or traffic intersections.

[0130] In a possible embodiment, Figure 7 As shown, it is a flow chart of another edge processing method provided in an embodiment of the present application. Figure 7 The edge processing method shown is a remote attestation process, including: S401-S404.

[0131] S401. The business party device obtains remote certification information from the edge computing device.

[0132] S402: The business party device sends a remote attestation request to the remote attestation server.

[0133] S403: The remote certification server verifies whether the remote certification information is correct.

[0134] S404: The remote attestation server sends a remote attestation response to the business party device.

[0135] Based on this, the business party device can obtain remote attestation information from the edge computing device and generate a remote attestation report. Furthermore, the business party device sends a remote attestation request (such as a first attestation request) carrying a remote attestation report to the remote attestation server. The remote attestation server can verify whether the remote attestation information is correct based on the pre-registered information of the locally stored TEE instance to determine whether the current TEE instance environment is trustworthy, and further return a remote attestation response to the business party device to indicate whether the current TEE instance environment is trustworthy. If the remote attestation response indicates that the current TEE instance environment is untrustworthy, the business party device may terminate the interaction with the application (such as an artificial intelligence model) in the TEE instance. If the remote attestation response indicates that the current TEE instance environment is trustworthy (such as the first pass response), the business party device may send information to be inferred to the application (such as an artificial intelligence model) in the TEE instance to implement related business.

[0136] Before sending the information to be processed to the edge computing device, the business device can first obtain the remote attestation information of the trusted execution environment instance where the artificial intelligence model is deployed from the edge computing device, and verify the remote attestation information through interaction with the remote attestation server to achieve edge trusted verification of the artificial intelligence model. Furthermore, if the remote attestation information is verified, the business device can send the information to be processed to the edge computing device for processing, so as to implement the business information processing process safely and with low latency, and avoid the risk of business data leakage.

[0137] In addition, confidential computing based on hardware TEE is believed to be able to achieve a scalable solution where data is available but not accessible. By deploying AI models in TEE instances of edge computing infrastructure, sensitive data is input into TEE in ciphertext form to complete AI reasoning, and the reasoning results are output, thus effectively protecting artificial intelligence model applications.

[0138] In this way, edge inference based on confidential computing can protect the privacy of business information on the one hand, and protect artificial intelligence model applications on the other hand, including tenants of edge computing devices, system administrators, and infrastructure providers cannot obtain or change artificial intelligence model applications.

[0139] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to realize the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0140] The embodiment of the present application can divide the functional modules of the business party device according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules. Optionally, the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0141] like Figure 8 As shown, it is a structural schematic diagram of an edge processing device provided by an embodiment of the present application. The edge processing device can be applied to a business party device, including: a sending module 501 and a receiving module 502. The sending module 501 is used to send a service request to the edge computing device. The service request is used to request the information processing service of the artificial intelligence model on the edge computing device. The receiving module 502 is used to receive remote certification information from the edge computing device. The remote certification information is the certification information of the trusted execution environment instance deployed with the artificial intelligence model pre-registered in the remote certification server. The sending module 501 is also used to send a first certification request to the remote certification server. The first certification request is used to request verification of the remote certification information. The sending module 501 is also used to send an information processing request carrying the information to be processed to the edge computing device when the receiving module 502 receives the first passing response from the remote certification server. The information processing request is used to request the artificial intelligence model to process the information to be processed.

[0142] Optionally, the remote attestation information includes the identifier of the trusted execution environment instance where the artificial intelligence model is deployed, the activation certificate of the trusted execution environment instance, and the private key signature corresponding to the activation certificate, the trusted execution environment software version number, the processor identifier, and the trusted computing base information.

[0143] Optionally, the sending module 501 is specifically configured to: establish a transport layer security protocol connection with the edge computing device based on the activation certificate of the trusted execution environment instance, and send the information to be processed to the edge computing device through the transport layer security protocol connection.

[0144] Optionally, the receiving module 502 is also used to receive information processing results from the edge computing device.

[0145] Optionally, the artificial intelligence model is deployed on the edge computing device when the operator device determines that the remote attestation server has passed the verification of the remote attestation information. The operator device includes an operation support system, a multi-access edge computing orchestrator, multiple mobile edge platform managers, and a virtualized infrastructure management module. The operation support system is used to respond to the first model instantiation request from the service-side device and send a second model instantiation request to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is used to respond to the second model instantiation request, determine the edge computing device that meets the preset conditions, and send a third model instantiation request to the target mobile edge platform manager corresponding to the edge computing device in the multiple mobile edge platform managers. The preset conditions include having instantiation resources corresponding to the artificial intelligence model and having trusted execution environment capabilities. The target mobile edge platform manager is used to respond to the third model instantiation request and send a fourth model instantiation request to the virtualized infrastructure management module. The virtualized infrastructure management module is used to respond to the fourth model instantiation request, determine the instantiation resources of the artificial intelligence model on the edge computing device, and the remote attestation information of the trusted execution environment instance corresponding to the instantiation resources, and send a second attestation request to the remote attestation server. The second attestation request is used to request verification of whether the trusted execution environment instance is trustworthy. The instantiated resources include computing resources, storage resources, network resources, and trusted execution environment resources. The virtualization infrastructure management module is also used to instantiate the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model when receiving the second pass response from the remote attestation server.

[0146] Optionally, the operator device also includes a mobile edge platform. The mobile edge platform manager is also used to send a service configuration request to the edge computing platform. The edge computing platform is used to respond to the service configuration request, obtain remote attestation information, and send a third attestation request to the remote attestation server. The third attestation request is used to verify whether the trusted execution environment instance is trustworthy. The edge computing platform is also used to determine the configuration information of the artificial intelligence model when receiving a third pass response from the remote attestation server. The configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

[0147] Optionally, the edge computing platform is further used to send a service configuration response to the target mobile edge platform manager. The target mobile edge platform manager is further used to send a first model instantiation response to the multi-access edge computing orchestrator. The multi-access edge computing orchestrator is further used to send a second model instantiation response to the service party device through the operation support system.

[0148] Those skilled in the art should be aware that in one or more of the above examples, the functions described in this application can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer-readable storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0149] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0150] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may be a physical unit or multiple physical units, that is, it may be located in one place, or it may be distributed in multiple different places. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0151] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An edge processing method, characterized in that: Applied to business equipment, including: Send a service request to an edge computing device; the service request is used to request an information processing service of an artificial intelligence model on the edge computing device; Receiving remote attestation information from the edge computing device; the remote attestation information is attestation information of a trusted execution environment instance in which the artificial intelligence model is deployed and pre-registered in a remote attestation server; Sending a first certification request to the remote certification server; the first certification request is used to request verification of the remote certification information; Upon receiving a first passing response from the remote attestation server, an information processing request carrying information to be processed is sent to the edge computing device; the information processing request is used to request the artificial intelligence model to perform information processing results on the information to be processed.

2. The edge processing method according to claim 1, characterized in that: The remote attestation information includes the identifier of the trusted execution environment instance where the artificial intelligence model is deployed, the activation certificate of the trusted execution environment instance, and the private key signature corresponding to the activation certificate, the trusted execution environment software version number, the processor identifier, and the trusted computing base information.

3. The edge processing method according to claim 2, characterized in that: The sending an information processing request carrying information to be processed to the edge computing device includes: Establishing a transport layer security protocol connection with the edge computing device based on the activation certificate of the trusted execution environment instance; The information to be processed is sent to the edge computing device through the transport layer security protocol connection.

4. The edge processing method according to claim 1, characterized in that: After sending the information processing request carrying the information to be processed to the edge computing device, the method further includes: Receive the information processing result from the edge computing device.

5. The edge processing method according to claim 1, characterized in that: The artificial intelligence model is deployed on the edge computing device when the operator device determines that the remote attestation server has passed the verification of the remote attestation information; the operator device includes an operation support system, a multi-access edge computing orchestrator, multiple mobile edge platform managers, and a virtualized infrastructure management module; The operation support system is used to send a second model instantiation request to the multi-access edge computing orchestrator in response to the first model instantiation request from the service-side device; The multi-access edge computing orchestrator is used to determine the edge computing device that meets the preset conditions in response to the second model instantiation request, and send a third model instantiation request to a target mobile edge platform manager corresponding to the edge computing device among the multiple mobile edge platform managers; the preset conditions include having instantiation resources corresponding to the artificial intelligence model and having trusted execution environment capabilities; The target mobile edge platform manager is used for sending a fourth model instantiation request to the virtualization infrastructure management module in response to the third model instantiation request; The virtualization infrastructure management module is used to determine, in response to the fourth model instantiation request, instantiated resources of the artificial intelligence model on the edge computing device, and remote attestation information of the trusted execution environment instance corresponding to the instantiated resources, and send a second attestation request to the remote attestation server; The second certification request is used to request verification of whether the trusted execution environment instance is trustworthy; the instantiated resources include computing resources, storage resources, network resources, and trusted execution environment resources; The virtualization infrastructure management module is also used to instantiate the artificial intelligence model in the trusted execution environment instance based on the image information of the artificial intelligence model when a second passing response is received from the remote attestation server.

6. The edge processing method according to claim 5, characterized in that: The operator equipment also includes a mobile edge platform; The mobile edge platform manager is further used to send a service configuration request to the edge computing platform; The edge computing platform is used to respond to the service configuration request, obtain the remote attestation information, and send a third attestation request to the remote attestation server; the third attestation request is used to verify whether the trusted execution environment instance is trustworthy; The edge computing platform is also used to determine the configuration information of the artificial intelligence model when a third pass response is received from the remote certification server; the configuration information includes domain name system rule information, service discovery configuration information, flow rule information, and trusted execution environment remote verification configuration information.

7. The edge processing method according to claim 6, characterized in that: The edge computing platform is also used to send a service configuration response to the target mobile edge platform manager; The target mobile edge platform manager is further configured to send a first model instantiation response to the multi-access edge computing orchestrator; The multi-access edge computing orchestrator is further configured to send a second model instantiation response to the service-side device through the operation support system.

8. An edge processing device, characterized in that: Applied to business equipment, including: sending module and receiving module; The sending module is used to send a service request to the edge computing device; the service request is used to request an information processing service of the artificial intelligence model on the edge computing device; The receiving module is used to receive remote attestation information from the edge computing device; the remote attestation information is attestation information of a trusted execution environment instance in which the artificial intelligence model is deployed and pre-registered in a remote attestation server; The sending module is further used to send a first certification request to the remote certification server; the first certification request is used to request verification of the remote certification information; The sending module is also used to send an information processing request carrying information to be processed to the edge computing device when the receiving module receives a first passing response from the remote proof server; the information processing request is used to request the artificial intelligence model to process the information results to be processed.

9. A computer device, characterized in that: It comprises a memory and a processor; the memory is used to store computer execution instructions, and the processor is connected to the memory through a bus; when the computer device is running, the processor executes the computer execution instructions stored in the memory, so that the computer device executes the edge processing method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes computer-executable instructions. When the computer-executable instructions are executed on a computer device, the computer device executes the edge processing method according to any one of claims 1 to 7.

Images