LoRa gateway management method, system, device and product

Abstract

The application relates to the field of communication. The application provides a LoRa gateway management method, system, device and product. The method comprises the following steps: an instance creation module creates a first agent instance according to a first key, the first agent instance is used for isolation with other agent instances, the first agent instance sends a first natural language instruction input by a user to an inference model, the inference model selects one or more first target skills from a skill set according to the first natural language instruction, and generates according to the one or more first target skills and the first natural language instruction; and an application module verifies a first command line set according to the first key, and executes a first operation on a first gateway set based on the first command line set in the case of passing the verification.

Description

Technical Field

[0001] This application relates to the field of communications, and in particular to a LoRa gateway management method, system, device, and product. Background Technology

[0002] A LoRa gateway management platform is an IoT platform specifically designed for centralized and standardized management of LoRa gateways. It can manage LoRa gateways deployed by multiple customers in multiple locations. However, configuring LoRa gateways using a LoRa gateway management platform requires a relatively high level of expertise from the customer. With the increasing prevalence of artificial intelligence (AI), using AI to enable LoRa gateway configuration through management platforms, thereby reducing the skill requirements for customers, is a major trend. However, the use of AI can easily lead to data leaks, compromising user data security. Summary of the Invention

[0003] This application provides a LoRa gateway management method, system, device, and product that can reduce the difficulty of operating the gateway by utilizing artificial intelligence while ensuring the security of user data.

[0004] Firstly, a LoRa gateway management method is provided, including: The instance creation module creates a first intelligent agent instance based on the first key. The first intelligent agent instance includes a skill set. The first intelligent agent instance is used to isolate itself from other intelligent agent instances. The first key is used to indicate the first user's access rights to call the first API interface provided by the LoRa gateway management platform and access rights to the first set of gateways under the jurisdiction of the LoRa gateway management platform. The first intelligent agent instance sends a first natural language instruction input by the user to the inference model, wherein the first natural language instruction includes instruction information input by the user for instructing the execution of a first operation on the first gateway set; The inference model selects one or more first target skills from the skill set according to the first natural language instruction, and generates a first command line set according to the one or more first target skills and the first natural language instruction, wherein the first command line set is used to call a first API interface to perform a first operation on the first gateway set; The application module verifies the first command line set based on the first key, and if the verification is successful, it executes the first operation on the first gateway set based on the first command line set.

[0005] In some possible designs, before the instance creation module creates the first agent instance based on the first key, the method further includes: The key creation module obtains the user's input of the calling permissions of the first API interface and the access permissions of the first gateway set, creates a first key based on the calling permissions of the first API interface and the access permissions of the first gateway set, and displays it to the user; The instance creation module creates a first agent instance based on the first key, including: The instance creation module obtains the first key input by the user and creates a first intelligent agent instance based on the first key.

[0006] In some possible designs, the method further includes: The instance creation module creates a second intelligent agent instance based on the second key. The second intelligent agent instance includes the skill set. The second intelligent agent instance is used to isolate itself from the first intelligent agent instance. The second key is used to indicate the second user's access rights to call the second API interface provided by the LoRa gateway management platform and access rights to the second gateway set under the jurisdiction of the LoRa gateway management platform. The second intelligent agent instance sends a second natural language instruction input by the user to the inference model, wherein the second natural language instruction includes instruction information input by the user to instruct the second gateway set to perform a second operation; The inference model selects one or more second target skills from the skill set according to the second natural language instruction, and generates a second command line set according to the one or more second target skills and the second natural language instruction. The second command line set is used to call the second API interface to perform a second operation on the second gateway set. The application module verifies the second command line set according to the second key. If the verification is successful, the application module performs the second operation on the second gateway set based on the second command line set.

[0007] In some possible designs, the first agent instance sends the first natural language instruction input by the user to the inference model, including: the first agent instance sending the description text of each skill in the skill set and the first natural language instruction input by the user to the inference model; The inference model selects one or more first target skills from the skill set according to the first natural language instruction, including: the inference model selects one or more first target skills from the skill set according to the first natural language instruction and the description text of each skill in the skill set; The inference model generates a first command line set based on one or more first target skills and the first natural language instruction, including: the inference model receiving the full text of one or more first target skills sent by the first agent instance, and generating the first command line set based on the full text of one or more first target skills and the first natural language instruction.

[0008] In some possible designs, the inference model generates a first command line set based on one or more first target skills and the first natural language instruction, including: The inference model selects a gateway serial number query skill from the skill set according to the first natural language instruction, wherein the gateway serial number query skill is used to query the serial number of the gateway; The inference model generates a gateway sequence query instruction based on the first natural language instruction and the gateway sequence number query skill, wherein the gateway sequence query instruction is used to query the sequence number of the gateway in the first gateway set; The inference model sends the gateway sequence query instruction to the first agent instance; The inference model receives the sequence number of the gateway in the first gateway set sent by the first agent instance; The inference model generates the first command line set based on one or more first target skills, the first natural language instructions, and the serial numbers of the gateways in the first gateway set.

[0009] In some possible designs, the application module includes an application client and an application server, and the application module verifies the first command line set based on the first key, including: The application client obtains the first command line set and the first key, encapsulates the first command line set into the request body field of the network request, and encapsulates the first key into the header field of the network request; The application server obtains the first key from the header field of the network request, obtains the first command line set from the request body field of the network request, and verifies the first command line set based on the first key.

[0010] In some possible designs, the first key is encapsulated in the authorization field of the network request.

[0011] In some possible designs, the first gateway set is one or more gateways under the jurisdiction of an organization within the LoRa gateway management platform, or one or more gateways under the jurisdiction of a subgroup within the LoRa gateway management platform, wherein the organization includes multiple subgroups.

[0012] Secondly, a LoRa gateway management system is provided, including: The instance creation module is used to create a first intelligent agent instance based on the first key. The first intelligent agent instance includes a skill set. The first intelligent agent instance is used to isolate itself from other intelligent agent instances. The first key is used to indicate the first user's access rights to call the first API interface provided by the LoRa gateway management platform and access rights to the first set of gateways under the jurisdiction of the LoRa gateway management platform. The first intelligent agent instance is used to send a first natural language instruction input by the user to the inference model, wherein the first natural language instruction includes instruction information input by the user for instructing to perform a first operation on the first gateway set; The inference model is used to select one or more first target skills from the skill set according to the first natural language instruction, and to generate a first command line set according to the one or more first target skills and the first natural language instruction, wherein the first command line set is used to call a first API interface to perform a first operation on the first gateway set; The application module is used to verify the first command line set based on the first key, and if the verification is successful, to perform the first operation on the first gateway set based on the first command line set.

[0013] Thirdly, a computing device is provided, comprising: a processor and a memory, wherein the processor is connected to the memory, the memory is used to store instructions, and the processor is used to read the instructions from the memory to perform the method as described in any of the first aspects.

[0014] Fourthly, a program product comprising instructions is provided, the instructions which, when executed by a computing device, cause the computing device to perform the method as described in any of the first aspects.

[0015] Fifthly, a readable storage medium is provided, including instructions that, when executed by a computing device, cause the computing device to perform the method as described in any of the first aspects.

[0016] In the above method, the target skill is automatically selected based on the user's input natural language commands through an inference model. The natural language commands are then translated into a set of command lines, which are used to execute corresponding operations on the gateway set. Users only need to express their intentions in natural language and do not need to master the underlying command line writing methods, thus reducing the professional skill requirements for users. In addition, when creating the first intelligent agent instance, the first key is strongly bound to the first intelligent agent instance, and different intelligent agent instances are completely isolated from each other. This ensures that each user's intelligent agent instance exists independently. User A's intelligent agent instance cannot access the data in User B's intelligent agent instance, preventing unauthorized access at the resource level. The first key also specifies the first user's access permissions to the first API interface provided by the LoRa gateway management platform and the access permissions to the first gateway set under the jurisdiction of the LoRa gateway management platform. If the subsequent access permissions for command lines exceed the permissions of the first key, the execution of the command line will be refused, further ensuring user security and preventing security issues caused by users issuing commands beyond their authorized scope due to insufficient professional knowledge. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the structure of an AI gateway management system provided in this application; Figure 2 This is an example diagram of a LoRa gateway management system deployed on a cloud platform according to this application; Figure 3 This is a flowchart illustrating a LoRa gateway configuration method provided in this application; Figure 4 This is a schematic diagram of the structure of a computing device provided in this application; Figure 5 This is a schematic diagram of the structure of a computing device cluster provided in this application; Figure 6 This is a schematic diagram of another computing device cluster provided in this application. Detailed Implementation

[0018] See Figure 1 , Figure 1 This is a structural diagram of an AI gateway management system provided in this application. Figure 1 As shown, the AI ​​gateway management system of this application includes: a client 110 and a LoRa gateway management system 120.

[0019] Client 110 is deployed on terminal devices, computing devices, or edge computing devices. Terminal devices include personal computers, smartphones, wearable devices, handheld processing devices, tablets, mobile laptops, augmented reality (AR) devices, virtual reality (VR) devices, smart conferencing devices, etc., without specific limitations. Computing devices can be bare metal servers (BMS), virtual machines, or containers. BMS refers to a general-purpose physical server, such as an ARM server or an x86 server; a virtual machine refers to a complete computer system simulated by software, possessing full hardware system functionality and running in a completely isolated environment. Any task that can be performed on a physical computer can also be performed in a virtual machine. When creating a virtual machine on a computing device, a portion of the physical machine's hard drive and memory capacity needs to be used as the virtual machine's hard drive and memory capacity. Each virtual machine has its own independent basic input / output system (BIOS), hard disk, and operating system, allowing it to be operated like a physical machine. A container is a portable software unit that combines an application and all its dependencies into a single software package. This package is not limited by the underlying host operating system, eliminating the need to build complex environments and simplifying the application development and deployment process. Edge computing devices are hardware devices located close to the data source or data user.

[0020] The LoRa gateway management system 120 may include: an instance creation module, an agent instance, an inference model, a target skill, and an application module. The LoRa gateway management system 120 can be deployed on a computing device or a cluster of computing devices, or it can be a chip or a chip system composed of multiple chips. The description of the computing device is as described above, and this application does not impose specific limitations. The chip can be implemented using a central processing unit (CPU), an application-specific integrated circuit (ASIC), or a programmable logic device (PLD). The PLD can be implemented using a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), a system-on-chip (SoC), or any combination thereof, and this application does not impose specific limitations. It may also include processors or accelerator cards that have good performance when performing machine learning tasks and can improve the speed and efficiency of neural network training, such as graphics processing units (GPUs), neural network processing units (NPUs), data processing units (DPUs), tensor processing units (TPUs), etc., which are not specifically limited in this application.

[0021] In one possible implementation, the client 110 and the LoRa gateway management system 120 can be deployed on the same computing device, or the client 110 and the LoRa gateway management system 120 can be deployed on different computing devices in the same computing device cluster, or the client 110 and the LoRa gateway management system 120 can be deployed on different computing devices in different computing device clusters. This application does not make any specific limitations.

[0022] In one possible implementation, the client 110 can be software or an application running on a terminal device or computing device controlled by the user, such as a personal computer (PC) client, a World Wide Web (web) client accessed through a browser, an application (APP) client running on a mobile terminal, or a console of a cloud platform. This application does not make any specific limitations.

[0023] It needs to be explained that, Figure 1 In addition to the above-described implementation, the architecture of the AI ​​gateway management system shown can also be deployed through data centers or cloud platforms. Specifically, client 110 can also be a client of a cloud platform, used for users to purchase and rent various cloud services. The LoRa gateway management method provided in this application can be one of these cloud services, which will not be listed here.

[0024] The preceding text has described in detail the possible deployment methods for client 110 and LoRa gateway management system 120. In actual deployment, any of the above methods can be flexibly adopted based on specific application scenarios and business requirements. The following section provides an illustrative example of the actual deployment methods for client 110 and LoRa gateway management system 120 in a specific application scenario.

[0025] As an example of an application scenario where an AI gateway management system is deployed on a local device, the client 110 and the LoRa gateway management system 120 can be deployed on office equipment within an enterprise. For example, the LoRa gateway management system 120 can be deployed on a service or server cluster purchased by the enterprise, and the client 110 can be deployed on the enterprise's office computer. The user is the enterprise's administrative staff, who can use the office computer to run the client 100. The client 100 sends the user's natural language commands to the LoRa gateway management system 120, and the LoRa gateway management system 120 can execute the LoRa gateway management method provided in this application to operate the gateway.

[0026] As an example of an application scenario where an AI gateway management system is deployed on a cloud platform, the LoRa gateway management system 120 can be deployed on a cloud platform. For example... Figure 2 This is an example diagram of a LoRa gateway management system 120 deployed on a cloud platform, as provided in this application. Figure 2As shown, a user can initiate a cloud service purchase request through client 110. The cloud service includes cloud services for operating the gateway. After client 100 sends the purchase request to the cloud platform, the cloud platform can provide the client 110 with cloud service access rights to the LoRa gateway management system 120. This allows the user to send natural language commands to the LoRa gateway management system 120 through client 110. The LoRa gateway management system 120 can then execute the data processing method provided in this application to operate the LoRa gateway.

[0027] The cloud platform also maintains various basic resources, including computing resources, storage resources, network resources, and security resources, to meet the computing needs of the LoRa Gateway Management System 120 under different scales and loads. Furthermore, these computing resources can be dynamically scaled according to the usage requirements of the LoRa Gateway Management System 120 to ensure the stable operation of the LoRa Gateway Management System 120 and to answer professional questions for users.

[0028] It is worth noting that, Figure 2 In the example shown, the LoRa gateway management system 120 is deployed and implemented in a cloud platform. Figure 2 This is merely an exemplary description of the LoRa gateway management system 120 involved in this application, and this application does not make any specific limitations.

[0029] In one possible implementation, the LoRa gateway management system 120 can also be deployed across multiple cloud platforms, for example, Figure 1 The instance creation module, the first intelligent agent instance, the first target skill, and the application module shown are deployed on the first cloud platform, while the inference model is deployed on the second cloud platform. The inference step, which involves reasoning according to each step and the knowledge, can be implemented by the second cloud platform. The above examples are for illustration only and are not intended to be specific.

[0030] In another possible implementation, the LoRa gateway management system 120 can also be deployed across multiple non-cloud platforms, for example... Figure 1 The instance creation module, the first intelligent agent instance, the first target skill, and the application module shown are deployed on the internal office server A1 of Company A, while the inference model is deployed on the internal office server of Company B. The above examples are for illustrative purposes only and are not intended to limit the scope of the application.

[0031] In another possible implementation, the LoRa gateway management system 120 can also be deployed across multiple cloud platforms and non-cloud platforms. For example, Figure 1 The instance creation module, the first intelligent agent instance, the first target skill, and the application module shown are deployed on the enterprise's internal office server, while the inference model is deployed on a cloud platform. The above examples are for illustration only and are not intended to be specific.

[0032] It should be understood that the above application scenarios are for illustrative purposes only. The LoRa gateway management system 120 and client 110 can be flexibly deployed according to actual business needs, and will not be listed in detail here. For ease of description, the following embodiments use... Figure 2 The system structure shown is used as an example for explanation.

[0033] See Figure 3 , Figure 3 This is a flowchart illustrating a LoRa gateway configuration method provided in this application. Figure 3 As shown, the LoRa gateway configuration method of this application includes the following steps: S101: The instance creation module creates the first intelligent agent instance based on the first key.

[0034] The first key is used to indicate the first user's access rights to call the first API interface provided by the LoRa gateway management platform and access rights to the first set of gateways under the jurisdiction of the LoRa gateway management platform.

[0035] The first user can be any user who needs to operate the gateways under the jurisdiction of the LoRa gateway management platform through the LoRa gateway management platform, such as maintenance personnel, system administrators, platform administrators, application developers, device users, tenants, etc.

[0036] The LoRa Gateway Management Platform is a centralized management system for LoRa gateway devices. It is responsible for unified access, status monitoring, configuration management, access control, data pass-through, and operation and maintenance management of all LoRa gateways within its jurisdiction, and provides API interfaces for users or upper-layer systems to call.

[0037] The first API interface can be a standardized call interface provided by the LoRa gateway management platform to implement operations related to the LoRa gateway. It serves as the unified entry point for the first user to interact with the LoRa gateway management platform and manage the LoRa gateway. In other words, the first API interface can be one or more of the following: a call entry point, function, URL address, etc., provided by the LoRa gateway management platform to the outside world. For example, the first API interface may include interfaces for querying gateway status, reading data, modifying configuration, restarting the gateway, and deleting the gateway, etc.

[0038] The first gateway set refers to one or more gateways under the jurisdiction of an organization within the LoRa gateway management platform, or one or more gateways under the jurisdiction of a subgroup within the LoRa gateway management platform. An organization can be a top-level permission unit within the LoRa gateway management platform, categorized by user unit, enterprise, department, or project, representing an independent user. For example, an organization could be a company, a school, a factory, an industrial park, etc. A subgroup can be a smaller group within an organization, used to classify and manage gateways by area, floor, workshop, or purpose within the same organization. For example, assuming the organization is a "company," the subgroups could be the company's workshops, office areas, north gate industrial park, warehouse area, etc. For instance, the first gateway set might include gateways from three subgroups under one organization, or gateways from three organizations, or subgroup A under organization A, subgroup B under organization B, etc.

[0039] The first key can include a first secret key, a first permission, and a second permission. The first secret key can be an encrypted string. As an identity credential, the first secret key is used to prevent forgery, theft, and interception, ensuring access security. By verifying this key, the legitimacy of the current access can be determined.

[0040] The first permission record specifies which API interfaces the first user can access in the LoRa gateway management platform and in what ways. For example, the first key specifies that the gateway information query interface can only be accessed via read mode, not write mode, and prohibits access to the delete gateway interface, modify gateway configuration interface, etc.

[0041] The second permission records which gateways under the LoRa gateway management platform the first user can access, and prevents them from viewing or operating other gateway devices beyond their authorized scope. For example, the second permission is used to indicate that operations can only be performed on gateways under the first gateway set.

[0042] The first agent instance is an independent execution entity deployed and running in a virtualized environment such as containers or virtual machines. It is generated by instantiating an agent template and possesses independent resource space, runtime state, and permission system. It is used to independently perform inference computation, task processing, interface calls, and data interaction operations. The first agent instance is used to isolate itself from other agent instances. That is, each agent instance is independent of each other in terms of permission scope, data access, and runtime resources. Each agent instance can only use its own bound key and operate within the authorized set of gateways and API interfaces, and cannot access the data and resources of other agent instances beyond its authorized scope, ensuring system security and resource isolation in a multi-tenant environment.

[0043] The specific implementation process of the instance creation module in creating the first agent instance based on the first key includes: Step 1: The instance creation module receives the creation request submitted by the first user. The creation request includes the user-inputted first key, first permission, and second permission. Optionally, the creation request may also include one or more of the following: agent instance name, agent identifier, user identifier, user name, validity period, expiration time, runtime environment type (e.g., running in a container or virtual machine), resource quotas (e.g., CPU size, memory size, storage size, etc.), instance triggering conditions, data storage conditions, etc.

[0044] Step 2: The instance creation module verifies the first key. For example, the instance creation module can use a preset RSA public key to decrypt the first key of the first key and restore the original key information; then, it uses the SHA256 hash algorithm to recalculate the restored original key information, compares the calculation result with the signature value in the first key, and if the two do not match, it is determined that the first key has been tampered with, the verification fails, and "key signature invalid" is returned.

[0045] Step 3: The instance creation module binds the first permission, the second permission, and the first agent instance. For example, the instance creation module can convert the first permission and the second permission into two whitelists for agent instances. Then, it can solidify the two whitelists into a permission configuration file in JSON / YAML format and mount the permission configuration file to a specified directory of the container or virtual machine, thereby ensuring that the permission configuration file is deeply bound to the instance runtime environment. This configuration file will be loaded when the instance is migrated or restarted, and it cannot be directly modified externally.

[0046] Before creating the first agent instance based on the first key, the first key must first be created. For example, the key creation module obtains the user's input of the first API interface's calling permissions and the first gateway set's access permissions, creates the first key based on the first API interface's calling permissions and the first gateway set's access permissions, and displays it to the first user.

[0047] S102: The first agent instance sends the first natural language instruction input by the first user to the inference model.

[0048] The first natural language instruction includes user-inputted instructions that direct the execution of a first operation on the first set of gateways. The first natural language instruction is an operation command input by the first user in the form of natural language used in everyday human communication (rather than code / API call instructions) and directed to the LoRa gateway management platform. Its core purpose is to allow the first user to instruct the first intelligent agent instance to perform a preset operation on a specified first set of gateways without needing to understand technical syntax, simply by using plain language.

[0049] The first natural language instruction may include an operation object and an operation action. The operation object explicitly refers to the first set of gateways. The operation action specifies the first operation to be performed. Optionally, the first natural language instruction may also include operation constraints. These constraints indicate supplementary information such as the conditions, scope, and time of the operation. For example, the first natural language instruction could be something like, "Check the online status of all gateways in Building 1 of the XX Park."

[0050] S103: The inference model selects one or more first target skills from the skill set based on the first natural language instruction, and generates a first command line set based on one or more first target skills and the first natural language instruction.

[0051] The inference model receives the first natural language instruction from the first user and, through a series of processes such as word segmentation, semantic understanding, intent recognition, and entity extraction, transforms the unstructured natural language requirement into structured task information that is understandable and executable by the machine, while simultaneously matching the target skill. The inference model can be a large language model, a specialized NLP model, a deep learning sequence model, a hybrid inference model, and so on.

[0052] A skill set can include one or more skills. A skill is a standardized unit of capability that can independently complete a specific type of technical task, or a packaged unit of smallest executable action. It is a bridge connecting high-level AI instructions and low-level hardware execution, or a composable, minimal-granularity tool invocation unit driven by a large language model (LLM).

[0053] Each primary skill can include configuration information. This configuration information may include one or more of the following: skill ID, skill name, supported user intents, supported entity types, and corresponding command-line template. The skill ID is the skill's identifier, used to distinguish different skills. The skill name is the skill's functional name, such as: gateway status query skill, gateway restart skill, gateway parameter setting skill. Supported user intents indicate which types of user intents the skill can handle, such as: supporting querying gateway status, supporting restarting the gateway, and supporting modifying gateway parameters. Supported entity types indicate which key entities the skill can handle, such as the supported gateway set, supported gateway IDs, supported operation objects, supported parameter types, required parameters, etc. The corresponding command-line template is the command format to be invoked or executed after the skill is selected.

[0054] The first command line set includes one or more first command lines. These command lines are used to invoke the first API interface to perform a first operation on the first gateway set. A command line is a standardized instruction string that can be directly recognized and executed by the LoRa gateway management system. Unlike the ambiguity of natural language, it has a fixed syntax, explicit parameters, and can be directly parsed and executed; it is machine language instruction that the LoRa gateway management system can understand.

[0055] The specific implementation process of the inference model selecting one or more first target skills from the skill set based on the first natural language instruction, and generating a first command line set based on one or more first target skills and the first natural language instruction includes: Step 1: The inference model cleans, preprocesses, and segments the first natural language instruction input by the user to obtain a segmented sequence.

[0056] The inference model first cleans and normalizes the first natural language instruction, removing redundant spaces, special symbols (such as @, #, and spaces), useless interjections (such as "oh," "ne," "please," and "help me"), and redundant modifiers (such as "trouble," "quickly," and "help me a little"). Simultaneously, the statement is uniformly converted to lowercase to avoid case differences affecting the parsing results. For example, if a user inputs "Please help me check the signal strength of gateway GW-A01 in the first gateway set, every 5 minutes," after cleaning and preprocessing, it is converted to "Check the signal strength of gateway GW-A01 in the first gateway set every 5 minutes." Then, the inference model performs word segmentation, employing a domain-adaptive segmentation algorithm (such as a dictionary-based forward maximum matching method combined with a Hidden Markov Model) to segment the preprocessed statement, dividing the continuous text into independent semantic units, resulting in a segmented word sequence.

[0057] Step 2: The inference model identifies the user's intent based on the word segmentation sequence and extracts key entities from the word segmentation set.

[0058] First, the inference model can filter verbs and verb-object phrases with clear action meanings from the segmented sequence as core verbs. For example, core verbs include query verbs (search, view, obtain, statistics), control verbs (restart, start, stop, reset), configuration verbs (modify, set, adjust, configure), log verbs (export, view, delete), and monitoring verbs (timed, every, real-time), etc.

[0059] Secondly, the inference model uses the object, modifier, and sentence structure in the tokenized sequence as context information. For example, "check" combined with the object "signal strength" indicates the intention of "querying the gateway signal strength"; if the tokenization contains "restart" and "gateway", the intention is "restarting the gateway"; if it contains "modify" and "reporting interval", the intention is "modifying the gateway reporting interval"; at the same time, combined with the modifier "every 5 minutes", it supplements the judgment that the intention includes the additional requirement of "timed query".

[0060] Then, the inference model inputs the extracted core verb and context information into a pre-trained intention classification model for classification. Among them, the intention classification model is used to classify user intentions and calculate the confidence score (ranging from 0 to 1) for each user intention; the higher the confidence score, the higher the matching degree of the user intention with the user input instruction, and the user intention with the highest confidence score is selected as the target user intention.

[0061] Finally, the inference model extracts key entities from the token set. Among them, the key entities can be operation objects, core parameters, entities related to constraint conditions, etc., which are used to clarify "who to operate on, what parameters to operate with, and under what conditions to operate".

[0062] Step 3: The inference model selects one or more first target skills from the skill set according to the user intention and the key entity, and generates a first command line set according to one or more first target skills and the first natural language instruction. Optionally, the first agent instance can also send information about all the skills in the skill set that the inference model instance has, or only transmit the names and description texts of all the skills in the skill set to help it quickly match the user intention; after the inference model selects the first target skill and notifies the first agent instance, the first agent instance then passes the full text of the skill selected by the inference model (including parameter format, usage examples, execution constraints, etc.) to the inference model, and the inference model generates a first command line set according to the user intention, the key entity, and the full text of the first target skill.

[0063] For example, a first user issues a first natural language command. The inference model performs semantic understanding and task breakdown on the command, determines which skills need to be invoked sequentially, and informs the first intelligent agent instance. Based on the inference model's selection, the first intelligent agent instance transmits the complete documentation of the relevant first target skill to the model. The model then converts the user's raw natural language into a set of command lines that the platform, device, or program can directly execute, and returns this set to the first instance. This achieves automatic conversion from user natural language to machine-executable instructions, lowering the user's learning curve and improving the convenience and intelligence of system interaction. Multiple skills in the skill set can be invoked by the first intelligent agent instance, allowing the first user to achieve the goal of converting different natural language commands into different command lines without needing to master related programming or skill invocation knowledge. For example, the inference model can convert natural language commands into query commands based on gateway status query skills; the inference model can convert natural language commands into restart commands based on gateway restart skills; the inference model can convert natural language commands into configuration commands based on gateway configuration modification skills; the inference model can convert natural language commands into signal detection commands based on gateway signal detection skills; and the inference model can convert natural language commands into log commands based on log acquisition skills.

[0064] Optionally, if the first natural language instruction sent by the first user does not include the serial numbers of the gateways in the first gateway set, then the inference model selects a gateway serial number query skill from the skill set based on the first natural language instruction. The gateway serial number query skill is used to query the serial numbers of gateways. The inference model generates a gateway sequence query instruction based on the first natural language instruction and the gateway serial number query skill. This gateway sequence query instruction is used to query the serial numbers of gateways in the first gateway set. The inference model sends the gateway sequence query instruction to the first agent instance. The first agent instance queries the database to obtain the serial numbers of the gateways in the first gateway set based on the gateway sequence query instruction. Then, the first agent instance sends the serial numbers of the gateways in the first gateway set to the inference model. The inference model receives the serial numbers of the gateways in the first gateway set sent by the first agent instance. The inference model generates a first command line set based on one or more first target skills, the first natural language instruction, and the serial numbers of the gateways in the first gateway set.

[0065] S104: The application module verifies the first command line set based on the first key, and if the verification is successful, performs the first operation on the first gateway set based on the first command line set.

[0066] The application module loads the first key and the first command line set, performs basic validity checks on the first key to ensure its validity, such as verifying its expiration date, whether its RSA signature matches the public key of the first agent instance, and whether the first key is complete. Then, the application module performs permission matching verification on the first key and the first command line set, such as verifying whether the operation objects in the first command line set belong to the first gateway set, and whether the operation type of the first command line has the calling permission for the first API interface of the first key. Finally, the application module executes the first operation on the first gateway set according to the execution order of the first command line set.

[0067] In the above method, the target skill is automatically selected based on the user's input natural language commands through an inference model. The natural language commands are then translated into a set of command lines, which are used to execute corresponding operations on the gateway set. Users only need to express their intentions in natural language and do not need to master the underlying command line writing methods, thus reducing the professional skill requirements for users. In addition, when creating the first intelligent agent instance, the first key is strongly bound to the first intelligent agent instance, and different first intelligent agent instances are completely isolated from each other. This ensures that each user's intelligent agent instance exists independently. User A's intelligent agent instance cannot access the data in User B's intelligent agent instance, preventing unauthorized access at the resource level. The first key also specifies the first user's access permissions to the first API interface provided by the LoRa gateway management platform and the access permissions to the first gateway set under the jurisdiction of the LoRa gateway management platform. If the subsequent access permissions for command lines exceed the permissions of the first key, the execution of the command line will be refused, further ensuring user security and preventing security issues caused by users issuing commands beyond their authorized scope due to insufficient professional knowledge.

[0068] Figure 3 The example shown illustrates how a first user creates only a first agent instance and uses this first agent instance to operate the LoRa gateway. In practical applications, based on user input, the instance creation module can also create a second agent instance using a second key, and then use this second agent instance to operate the LoRa gateway. Specifically, this can include the following steps: Step 1: The instance creation module creates the first intelligent agent instance based on the first key.

[0069] The first key is used to indicate the first or second user's access rights to call the second API interface provided by the LoRa gateway management platform and access rights to the set of second gateways under the jurisdiction of the LoRa gateway management platform.

[0070] The second user can be another user who needs to operate the gateway under the jurisdiction of the LoRa gateway management platform through the LoRa gateway management platform, such as maintenance personnel, system administrators, platform administrators, application developers, device users, tenants, etc.

[0071] The second API interface can be a standardized call interface provided by the LoRa gateway management platform to implement operations related to the LoRa gateway. It serves as a unified entry point for the first or second user to interact with the LoRa gateway management platform and perform management operations on the LoRa gateway.

[0072] The second gateway set is one or more gateways under the jurisdiction of an organization within the LoRa gateway management platform, or one or more gateways under the jurisdiction of a subgroup within the LoRa gateway management platform.

[0073] The second agent instance is an independent execution entity deployed and running in a virtualized environment such as containers or virtual machines. It is generated by instantiating an agent template and has its own independent resource space, running state, and permission system. It is used to independently perform inference computation, task processing, interface calls, and data interaction operations. The second agent instance is isolated from the first agent instance.

[0074] The specific implementation process of the instance creation module creating a second agent instance based on the second key is similar to the specific implementation process of the instance creation module creating a first agent instance based on the first key, and will not be elaborated here.

[0075] Step 2: The second agent instance sends the second natural language command input by the second user to the inference model.

[0076] The second natural language instruction includes user-inputted instructions that direct the execution of a second operation on the second set of gateways. The second natural language instruction is an operation command input by either the first or second user in the form of natural language used in everyday human communication (rather than code / API call instructions) and directed to the LoRa gateway management platform. Its core purpose is to allow the second user to instruct a second intelligent agent instance to perform a preset operation on a specified set of second gateways using only plain language, without needing to master technical syntax.

[0077] The second natural language instruction may include an operation object and an operation action. The operation object explicitly refers to the second set of gateways. The operation action specifies the second operation to be performed. Optionally, the second natural language instruction may also include operation constraints. These constraints indicate supplementary information such as the conditions, scope, and time of the operation. For example, the second natural language instruction could be something like "Check the signal strength and data reporting frequency of the GW-A01 gateway."

[0078] Step 3: The inference model selects one or more second target skills from the skill set based on the second natural language instructions, and generates a second command line set based on one or more second target skills and the second natural language instructions.

[0079] The second command line set includes one or more second command lines. These second command lines are used to invoke a second API interface to perform a second operation on the second gateway set.

[0080] The specific implementation process of the inference model selecting one or more second target skills from the skill set based on the second natural language instruction, and generating a second command line set based on one or more second target skills and the second natural language instruction is similar to the specific implementation process of the inference model selecting one or more first target skills from the skill set based on the first natural language instruction, and generating a first command line set based on one or more first target skills and the first natural language instruction. Therefore, it will not be elaborated here.

[0081] Step 4: The application module verifies the second command line set based on the second key. If the verification is successful, it performs the second operation on the second gateway set based on the second command line set.

[0082] The application module loads the second key and the second command line set, performs basic validity checks on the second key to ensure its validity, such as verifying its expiration date, whether its RSA signature matches the public key of the second agent instance, and whether the second key is complete. Then, the application module performs permission matching verification on the second key and the second command line set, such as verifying whether the operation objects in the second command line set belong to the second gateway set, and whether the operation type of the second command line has the calling permission for the second API interface of the second key. Finally, the application module executes the second operation on the second gateway set according to the execution order of the second command line sets.

[0083] This application also provides a computing device 200. For example... Figure 4 As shown, the computing device 200 includes a bus 202, a processor 204, a memory 206, and a communication interface 208. The processor 204, the memory 206, and the communication interface 208 communicate with each other via the bus 202. The computing device 200 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in the computing device 200.

[0084] Bus 202 can be a Peripheral Component Interconnect Express (PCIe) bus, an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. The Unified Bus is also known as the Lingqu Bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus 204 may be represented by a single line, but this does not mean that there is only one bus or one type of bus. The bus 204 may include a path for transmitting information between various components of the computing device 200 (e.g., memory 206, processor 204, communication interface 208).

[0085] The processor 204 may include any one or more computing devices such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP) or a digital signal processor (DSP), an ASIC, an FPGA, a CPLD, an NPU, a SoC, an offload card, or an accelerator card.

[0086] Memory 206 may include volatile memory, such as random access memory (RAM). Processor 104 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD). Furthermore, memory 106 may also be implemented using storage class memory (SCM), phase change memory (PCM), or other types of storage media.

[0087] It is worth noting that the same type of storage medium can be configured in the same computing device to realize the function of memory 206, or two or more types of storage media can be configured to realize the function of memory 206. This application does not limit this.

[0088] The memory 206 stores executable program code, which the processor 204 executes to implement the XXX method. In other words, the memory 106 stores instructions for executing the LoRa gateway configuration method.

[0089] Alternatively, the memory 206 may store executable code, which the processor 204 executes to implement the LoRa gateway configuration method. That is, the memory 206 may contain instructions for executing the LoRa gateway configuration method.

[0090] The communication interface 203 uses transceiver modules, such as, but not limited to, network interface cards and transceivers, to enable communication between the computing device 200 and other devices or communication networks.

[0091] This application also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0092] like Figure 5 As shown, the computing device cluster includes at least one computing device 200. The memory 206 in one or more computing devices 200 in the computing device cluster may store the same instructions for executing LoRa gateway configuration methods.

[0093] In some possible implementations, the memory 206 of one or more computing devices 200 in the computing device cluster may also store partial instructions for executing the LoRa gateway configuration method. In other words, a combination of one or more computing devices 200 can jointly execute instructions for executing the LoRa gateway configuration method.

[0094] In some possible implementations, one or more computing devices in a computing device cluster can be connected via a network. This network can be a wide area network (WAN) or a local area network (LAN), etc. Figure 6 One possible implementation is shown. For example... Figure 6 As shown, the two computing devices 200A and 200B are connected via a network. Specifically, they are connected to the network through the communication interfaces in each computing device.

[0095] It should be understood that Figure 6The functions of the computing device 200A shown can also be performed by multiple computing devices 200. Similarly, the functions of the computing device 200B can also be performed by multiple computing devices 200.

[0096] This application also provides another computing device cluster. The connection relationships between the computing devices in this computing device cluster can be similarly referred to... Figure 5 and Figure 6 The connection method of the computing device cluster is different in that the memory 206 of one or more computing devices 200 in the computing device cluster can store the same instructions for executing the LoRa gateway configuration method.

[0097] In some possible implementations, the memory 206 of one or more computing devices 200 in the computing device cluster may also store partial instructions for executing the LoRa gateway configuration method. In other words, a combination of one or more computing devices 200 can jointly execute instructions for executing the LoRa gateway configuration method.

[0098] This application also provides a computer program product containing instructions. The computer program product may be a software or program product containing instructions, capable of running on a computing device or stored on any usable medium. When the computer program product is run on at least one computing device, it causes the at least one computing device to execute a LoRa gateway configuration method.

[0099] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct a computing device to execute a LoRa gateway configuration method.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A LoRa gateway management method, characterized in that, include: The instance creation module creates a first intelligent agent instance based on the first key. The first intelligent agent instance includes a skill set. The first intelligent agent instance is used to isolate itself from other intelligent agent instances. The first key is used to indicate the first user's access rights to call the first API interface provided by the LoRa gateway management platform and access rights to the first set of gateways under the jurisdiction of the LoRa gateway management platform. The first intelligent agent instance sends a first natural language instruction input by the user to the inference model, wherein the first natural language instruction includes instruction information input by the user for instructing the execution of a first operation on the first gateway set; The inference model selects one or more first target skills from the skill set according to the first natural language instruction, and generates a first command line set according to the one or more first target skills and the first natural language instruction, wherein the first command line set is used to call a first API interface to perform a first operation on the first gateway set; The application module verifies the first command line set based on the first key, and if the verification is successful, it executes the first operation on the first gateway set based on the first command line set.

2. The method according to claim 1, characterized in that, Before the instance creation module creates the first agent instance based on the first key, the method further includes: The key creation module obtains the user's input of the calling permissions of the first API interface and the access permissions of the first gateway set, creates a first key based on the calling permissions of the first API interface and the access permissions of the first gateway set, and displays it to the user; The instance creation module creates a first agent instance based on the first key, including: The instance creation module obtains the first key input by the user and creates a first intelligent agent instance based on the first key.

3. The method according to claim 1, characterized in that, The method further includes: The instance creation module creates a second intelligent agent instance based on the second key. The second intelligent agent instance includes the skill set. The second intelligent agent instance is used to isolate itself from the first intelligent agent instance. The second key is used to indicate the second user's access rights to call the second API interface provided by the LoRa gateway management platform and access rights to the second gateway set under the jurisdiction of the LoRa gateway management platform. The second intelligent agent instance sends a second natural language instruction input by the user to the inference model, wherein the second natural language instruction includes instruction information input by the user to instruct the second gateway set to perform a second operation; The inference model selects one or more second target skills from the skill set according to the second natural language instruction, and generates a second command line set according to the one or more second target skills and the second natural language instruction. The second command line set is used to call the second API interface to perform a second operation on the second gateway set. The application module verifies the second command line set according to the second key. If the verification is successful, the application module performs the second operation on the second gateway set based on the second command line set.

4. The method according to any one of claims 1 to 3, characterized in that, The first agent instance sends the first natural language instruction input by the user to the inference model, including: the first agent instance sending the description text of each skill in the skill set and the first natural language instruction input by the user to the inference model; The inference model selects one or more first target skills from the skill set according to the first natural language instruction, including: the inference model selects one or more first target skills from the skill set according to the first natural language instruction and the description text of each skill in the skill set; The inference model generates a first command line set based on one or more first target skills and the first natural language instruction, including: the inference model receiving the full text of one or more first target skills sent by the first agent instance, and generating the first command line set based on the full text of one or more first target skills and the first natural language instruction.

5. The method according to any one of claims 1 to 3, characterized in that, The inference model generates a first command line set based on one or more first target skills and the first natural language instructions, including: The inference model selects a gateway serial number query skill from the skill set according to the first natural language instruction, wherein the gateway serial number query skill is used to query the serial number of the gateway; The inference model generates a gateway sequence query instruction based on the first natural language instruction and the gateway sequence number query skill, wherein the gateway sequence query instruction is used to query the sequence number of the gateway in the first gateway set; The inference model sends the gateway sequence query instruction to the first agent instance; The inference model receives the sequence number of the gateway in the first gateway set sent by the first agent instance; The inference model generates the first command line set based on one or more first target skills, the first natural language instructions, and the serial numbers of the gateways in the first gateway set.

6. The method according to any one of claims 1 to 3, characterized in that, The application module includes an application client and an application server. The application module verifies the first command line set based on the first key, including: The application client obtains the first command line set and the first key, encapsulates the first command line set into the request body field of the network request, and encapsulates the first key into the header field of the network request; The application server obtains the first key from the header field of the network request, obtains the first command line set from the request body field of the network request, and verifies the first command line set based on the first key.

7. The method according to claim 6, characterized in that, The first key is encapsulated in the authorization field of the network request.

8. The method according to any one of claims 1 to 3, characterized in that, The first gateway set is one or more gateways under the jurisdiction of an organization in the LoRa gateway management platform, or one or more gateways under the jurisdiction of a subgroup in the LoRa gateway management platform, wherein the organization includes multiple subgroups.

9. A LoRa gateway management system, characterized in that, include: The instance creation module is used to create a first intelligent agent instance based on the first key. The first intelligent agent instance includes a skill set. The first intelligent agent instance is used to isolate itself from other intelligent agent instances. The first key is used to indicate the first user's access rights to call the first API interface provided by the LoRa gateway management platform and access rights to the first set of gateways under the jurisdiction of the LoRa gateway management platform. The first intelligent agent instance is used to send a first natural language instruction input by the user to the inference model, wherein the first natural language instruction includes instruction information input by the user for instructing to perform a first operation on the first gateway set; The inference model is used to select one or more first target skills from the skill set according to the first natural language instruction, and to generate a first command line set according to the one or more first target skills and the first natural language instruction, wherein the first command line set is used to call a first API interface to perform a first operation on the first gateway set; The application module is used to verify the first command line set based on the first key, and if the verification is successful, to perform the first operation on the first gateway set based on the first command line set.

10. A computing device, characterized in that, include: A processor and a memory, wherein the processor is connected to the memory, the memory is used to store instructions, and the processor is used to read the instructions from the memory to perform the method as claimed in any one of claims 1-8.

11. A program product containing instructions, characterized in that, Includes instructions that, when executed by a computing device, cause the computing device to perform the method as described in any one of claims 1-8.

Images