A device networking method and apparatus, an electronic device, and a storage medium

Abstract

The application discloses a device networking method and device, electronic equipment and storage medium. The first control strategy focusing on the communication efficiency of the bottom layer, the second control strategy focusing on the service execution efficiency and the third control strategy focusing on the user behavior preference are deduced in parallel in a virtual network. Before the physical networking is issued, the collaborative optimal networking strategy considering the network quality, scene automatic stability and user life preference is accurately generated, so that the device networking efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of device networking technology, and in particular to a device networking method, a device networking apparatus, an electronic device, and a readable storage medium. Background Technology

[0002] Currently, when dealing with the networking and scenario configuration of smart devices in complex home networks, related technologies mostly focus on the scheduling and post-maintenance of physical devices after they are put into operation. Typically, after all physical devices are actually installed, powered on and connected to the network, the home host directly performs on-site communication debugging, priority allocation and localized distribution of linkage rules on the real physical channel.

[0003] However, since large-scale smart home scenarios often involve hundreds of cross-protocol sub-devices and complex house layouts and wall spaces, this post-event optimization method that relies entirely on the physical environment results in the occurrence of resonance interference, sudden electromagnetic background noise, and channel congestion faults under multiple sources of sudden traffic concurrency when the devices are running on site. These issues only erupt after the hardware deployment is completed. The on-site debugging cycle is long, the manual troubleshooting cost is high, and the network configuration is easily affected by dynamic interference. Summary of the Invention

[0004] The present invention provides a device networking method, apparatus, electronic device, and readable storage medium to overcome or at least partially solve the above-mentioned problems.

[0005] To solve the above-mentioned technical problems, this application is implemented as follows: In a first aspect, embodiments of this application provide a device networking method, including: Identify the virtual devices corresponding to the target device and construct a virtual network to enable data communication between multiple virtual devices; Obtain the protocol stack status parameters of the virtual device in the virtual network, and generate a tentative configuration change command based on the protocol stack status parameters; In the virtual network, the tentative configuration change command is applied to the virtual device to generate a first control strategy that is correlated with the communication efficiency of the virtual network; Read the simulated traffic generated for a preset test scenario, and when executing the simulated traffic, apply the tentative configuration change command to the virtual device in the virtual network to generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario; Obtain user behavior preference settings for the virtual device, and generate a third control strategy that is related to the user behavior preference settings for the virtual device; A target control strategy is determined from the first control strategy, the second control strategy, and the third control strategy, and the target device is controlled to perform networking operations based on the target control strategy.

[0006] Optionally, it also includes: Generate target household raw data, which includes environmental radio frequency characteristics, floor plan information, and target device parameters for the target device; The equipment brand combination is determined based on the target equipment parameters, and the room structure characteristics are determined based on the floor plan information; A similar scenario search is performed in the anonymized experience knowledge base to match historical configuration information that corresponds to the device brand combination and the room structure features.

[0007] Optionally, it also includes: The historical configuration information is determined as the initial seed configuration; The original data of the target family and the initial seed configuration are subjected to multi-dimensional feature fusion processing to construct a virtual digital twin model, wherein the virtual digital twin model includes, A device virtual model used to store the protocol stack state parameters of the virtual device in the virtual network; This is used to take the environmental radio frequency characteristics as initial electromagnetic background parameters, and use the initial electromagnetic background parameters in combination with the floor plan information to generate a link signal model for simulating the physical attenuation and co-channel interference scenarios when signals penetrate physical walls and spatial obstacles. A user behavior model used to output user behavior preference settings.

[0008] Optionally, the step of obtaining the protocol stack status parameters of the virtual device in the virtual network and generating a tentative configuration change instruction based on the protocol stack status parameters includes: The pre-trained network protocol agent is invoked to read the protocol stack state parameters; The communication protocol type and current configuration parameters currently used by the virtual device are determined by the protocol stack status parameters. Based on the communication protocol type, retrieve the legal change boundary range corresponding to the communication protocol type from the preset protocol stack attribute database, and generate legal configuration parameter boundary range information through the legal change boundary range; Based on the current configuration parameters, a set of tentative adjustment increment values ​​for changing the current configuration parameters are calculated within the legal configuration parameter boundary range information, and a tentative configuration adjustment strategy including channel switching step size and transmit power adjustment magnitude is generated based on the tentative adjustment increment values. The tentative configuration adjustment strategy is encapsulated into a tentative configuration change instruction that conforms to the communication protocol type.

[0009] Optionally, the step of applying the tentative configuration change instruction to the virtual device in the virtual network and generating a first control strategy correlated with the communication efficiency of the virtual network includes: The network protocol agent is invoked to inject the tentative configuration change command into the virtual media access control layer and virtual physical layer of the virtual network, thereby adjusting the underlying communication state of the virtual device; After the virtual device completes the adjustment of the underlying communication state, real-time data of the underlying network state is generated by capturing the physical layer transmit power, received signal strength indication and channel bit error rate of the virtual network under the action of the exploratory configuration change command. Based on the pre-acquired real radio frequency coexistence interference curve, the average packet loss rate and round-trip delay of the real-time data of the network's underlying state under the influence of the simulated signal attenuation interference scenario are calculated, and link performance evaluation data reflecting the changes in communication link quality indicators are generated; the real radio frequency coexistence interference curve is an electromagnetic interference characteristic curve that expresses the quantitative correspondence between the degree of signal-to-noise ratio degradation and the degree of communication link impairment when heterogeneous wireless signals are concurrently emitted at the same frequency. The average packet loss rate of the virtual network is determined by comparing the link performance evaluation data. When the decrease in the average packet loss rate exceeds the preset packet loss rate threshold, the current exploratory configuration change instruction is marked as a high-value action, and communication efficiency correlation data containing the target channel allocation table and device power bias value is output. The communication efficiency-related data is structured and encapsulated to generate a first control strategy.

[0010] Optionally, the step of reading the simulated traffic generated for a preset test scenario, and applying the tentative configuration change command to the virtual device in the virtual network when executing the simulated traffic, to generate a second control strategy related to the execution efficiency of the virtual device in the test scenario, includes: The pre-trained scenario-linked intelligent agent is invoked to read the simulated traffic generated by the user behavior model for the test scenario, and the target concurrent business data packets corresponding to the target user behavior are obtained. The target concurrent service data packet is input into the link signal model, and the link signal model is controlled to replay the target concurrent service data packet in order to generate a virtual radio electromagnetic signal corresponding to the test scenario in the virtual network; When replaying the target concurrent service data packets, the tentative configuration change command is applied to the virtual device in the virtual network to control the transmission of the virtual radio electromagnetic signal under the analog signal attenuation interference scenario, and to generate service scenario robustness assessment data; the service scenario robustness assessment data is used to characterize the physical resilience of the virtual device in terms of anti-interference, anti-packet loss and tolerance to latency fluctuations when the virtual device performs data transmission in the virtual network. The robustness assessment data of the business scenario is used to calculate business layer automation performance data to express the success rate and stability of scenario linkage for the test scenario; A second control strategy is generated based on the automation performance data of the business layer, which is correlated with the execution efficiency of the virtual device in the test scenario.

[0011] Optionally, the tentative configuration change instruction includes physical layer transmit power, and the step of obtaining user behavior preference settings for the virtual device and generating a third control policy related to the user behavior preference settings for the virtual device includes: The device wake-up frequency of the virtual device is determined by the second control strategy, and a pre-trained user experience agent is invoked to read the user behavior preference settings through the user behavior model; the user behavior preference settings include the device standby power consumption limit and daily routine time periods; The increase in standby power consumption and the change in electromagnetic radiation of the virtual device due to the second control strategy are calculated by using the physical layer transmission power, the device wake-up frequency, the device standby power consumption limit, and the daily work and rest time period. Based on the user behavior preference settings, the user's silent mode preference during a specific period is determined, and the silent mode preference is used to generate quantitative data on life disturbances that are used to evaluate the potential disturbances caused to the user by the tentative configuration change command during the specific period. A weighted scoring operation is performed on the increase in standby power consumption, the change in electromagnetic radiation, and the quantitative data of daily life disturbance to generate user sensory comfort and low-carbon energy-saving satisfaction indicators for quantifying the potential impact of the tentative configuration change instruction and the second control strategy on daily life. Based on the user's sensory comfort and low-carbon energy-saving satisfaction indicators, a third control strategy that is correlated with the user's behavioral preference settings is output.

[0012] Optionally, the step of determining the target control strategy from the first control strategy, the second control strategy, and the third control strategy includes: The first control strategy, the second control strategy, and the third control strategy are aggregated into a decision set to be resolved. Feature cross-matching is performed on the decision set to be resolved. When it is determined that the transmit power adjustment action in the first control strategy, the redundancy backup action in the second control strategy, and the energy consumption upper limit constraint in the third control strategy are mutually exclusive, a multi-attribute decision conflict state is triggered. When the multi-attribute decision conflict state is detected to be triggered, the weighted negotiation algorithm based on multi-attribute decision is invoked and the preset global target weight is read. The benefits of the first control strategy, the second control strategy and the third control strategy are evaluated respectively, the comprehensive score for each control strategy is calculated, and the target control strategy is selected from the first control strategy, the second control strategy and the third control strategy based on the comprehensive score.

[0013] Optionally, the step of controlling the target device to perform networking operations based on the target control strategy further includes: The target control strategy is transformed into a set of physical configuration parameters that include the target channel number, node routing table, local degradation linkage logic, and maximum power limit threshold. The physical configuration parameter set is pushed into the physical register of the target device to control the target device to perform networking operations based on the physical configuration parameter set.

[0014] Optionally, it also includes: The link signal model outputs a wall attenuation coefficient to reflect the physical attenuation status, and an environmental noise floor to reflect the co-frequency interference status. Determine the target communication link in the virtual network; The received signal strength indication, average round-trip time delay, and link layer retransmission count of the target communication link are collected at preset time intervals to generate real-time link measurement data reflecting the current state of the target communication link. Read the theoretical prediction data output by the link signal model and the user behavior model at the same timestamp, and determine the absolute value of the deviation between the real-time measured data and the theoretical prediction data; When the absolute value of the deviation exceeds a preset threshold for a continuous preset period, the wall attenuation coefficient and the environmental noise floor are fitted and calculated to update the link signal model. After updating the link signal model, the network protocol agent, the scenario linkage agent, and the user experience agent are triggered to perform incremental virtual testing on the deviation area between the real-time link measured data and the theoretical prediction data, so as to output a fine-tuning configuration scheme for the target device.

[0015] Secondly, embodiments of this application provide a device networking apparatus, characterized in that it includes: The virtual network construction module is used to identify the virtual devices corresponding to the target device and construct a virtual network for enabling data communication between multiple virtual devices. The tentative configuration change instruction reading module is used to obtain the protocol stack status parameters of the virtual device in the virtual network, and generate tentative configuration change instructions based on the protocol stack status parameters; The first control strategy generation module is used to apply the tentative configuration change command to the virtual device in the virtual network and generate a first control strategy that is related to the communication efficiency of the virtual network. The second control strategy generation module is used to read the simulated traffic generated for a preset test scenario, and when the simulated traffic is executed, to apply the tentative configuration change command to the virtual device in the virtual network, and generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario. The third control strategy generation module is used to obtain user behavior preference settings for the virtual device and generate a third control strategy that is related to the user behavior preference settings for the virtual device. The device networking control module is used to determine a target control strategy from the first control strategy, the second control strategy and the third control strategy, and control the target device to perform networking operations based on the target control strategy.

[0016] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0017] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0018] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0019] The embodiments of the present invention have the following advantages: In this embodiment of the invention, by parallel deducing a first control strategy focusing on underlying communication efficiency, a second control strategy focusing on service execution efficiency, and a third control strategy focusing on user behavior preferences in a virtual network, a collaborative optimal networking strategy that takes into account network quality, scenario automation stability, and user lifestyle preferences is accurately generated before the physical networking is deployed, thereby improving the device networking efficiency. Attached Figure Description

[0020] Figure 1 This is a flowchart of the steps of a device networking method provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a device networking method provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of a device networking apparatus provided in an embodiment of the present invention; Figure 4 This is a hardware structure block diagram of an electronic device provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0023] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Reference Figure 1 The diagram illustrates a flowchart of a device networking method provided in an embodiment of the present invention, which may specifically include the following steps: Step 101: Determine the virtual device corresponding to the target device and construct a virtual network for enabling data communication between multiple virtual devices; Step 102: Obtain the protocol stack status parameters of the virtual device in the virtual network, and generate a tentative configuration change command based on the protocol stack status parameters; Step 103: Apply the tentative configuration change command to the virtual device in the virtual network to generate a first control strategy that is related to the communication efficiency of the virtual network; Step 104: Read the simulated traffic generated for the preset test scenario, and when executing the simulated traffic, apply the tentative configuration change command to the virtual device in the virtual network to generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario; Step 105: Obtain user behavior preference settings for the virtual device, and generate a third control strategy that is related to the user behavior preference settings for the virtual device; Step 106: Determine the target control strategy from the first control strategy, the second control strategy, and the third control strategy, and control the target device to perform networking operations based on the target control strategy.

[0025] In specific implementations, embodiments of the present invention can be applied to cloud computing platforms that run digital twin engines and various AI agents. The cloud computing platform can provide the computing power and storage space required to support high-concurrency virtual stress testing, large-scale data fitting, and conflict resolution algorithms.

[0026] According to the embodiments of the present invention, virtual devices corresponding to target devices can be identified, and a virtual network for enabling data communication between multiple virtual devices can be constructed. This allows for the creation of a digital mirror in the virtual digital twin space that fully corresponds to the real physical hardware and network topology, providing a secure and controllable foundational software environment for all subsequent virtual simulation tests and strategy deductions.

[0027] Target devices refer to various smart home hardware entities that actually exist in the physical home environment and are to be configured for networking.

[0028] A virtual device refers to a digital software image created in the cloud or a digital twin system to simulate the communication behavior and functional characteristics of the target device.

[0029] A virtual network refers to a virtual communication topology built in a digital twin space, used to simulate the wireless signal transmission and data interaction of multiple virtual devices under a specific apartment layout.

[0030] In this embodiment of the invention, the protocol stack status parameters of the virtual device in the virtual network can be obtained, and a tentative configuration change instruction can be generated through the protocol stack status parameters. By sensing the current communication configuration status of the virtual device and combining it with its protocol boundaries, a reasonable configuration fine-tuning action with probe characteristics can be designed, providing an input source for evaluating changes in the underlying communication performance.

[0031] Protocol stack state parameters refer to the set of static and dynamic parameters that a virtual device is currently operating at each layer of the network protocol (such as the network layer and the physical layer), including the current channel number, the current transmit power level, etc.

[0032] Exploratory configuration change commands are test commands generated by the system to change the current communication state of virtual devices, serving as exploratory inputs to probe network performance responses.

[0033] In this embodiment of the invention, the tentative configuration change command can be applied to the virtual device in the virtual network to generate a first control strategy that is related to the communication efficiency of the virtual network. By applying changes at the pure network level, the changes in underlying communication indicators such as wireless channel quality and bit error rate can be quantitatively analyzed, thereby deriving an optimal network layer configuration scheme aimed at maximizing physical transmission quality.

[0034] Communication efficiency refers to the physical quality indicators of data transmission in a virtual network, which is usually measured by a combination of latency, packet loss rate, signal strength and channel throughput.

[0035] The first control strategy refers to the configuration decision scheme derived by the network protocol agent based on changes in communication efficiency, which focuses on optimizing the underlying wireless link and eliminating channel interference.

[0036] In this embodiment of the invention, simulated traffic generated for a preset test scenario can be read. When the simulated traffic is executed, the tentative configuration change command is applied to the virtual device in the virtual network. A second control strategy is generated that is related to the execution efficiency of the virtual device in the test scenario. This allows for the superposition of concurrent data load under specific life scenarios with underlying configuration changes for testing. The robustness of business execution under multi-source traffic and network fluctuations is evaluated, thereby deriving a business layer configuration scheme that ensures a high success rate in automated scenarios.

[0037] The preset test scenarios refer to typical user life scenarios that the system pre-sets or simulates, such as returning home late at night or setting up defenses when away from home.

[0038] Simulated traffic refers to a stream of virtual service data packets generated by multiple virtual devices concurrently sending and receiving within a specific time period, based on a test scenario.

[0039] Execution efficiency refers to the probability and response speed at which automated linkage rules are executed completely, stably, and without lag under a specific business load.

[0040] The second control strategy refers to a business layer configuration decision scheme derived by the scenario-linked intelligent agent under concurrent traffic stress testing, which focuses on ensuring the success rate of scenario execution and system robustness.

[0041] In this embodiment of the invention, user behavior preference settings for the virtual device can be obtained, and a third control strategy with a correlation to the user behavior preference settings for the virtual device can be generated. This introduces the individual behavioral habits and subjective preferences of human users, quantitatively evaluates the potential human-computer interaction impact of the aforementioned technical adjustment scheme on the daily living environment, and thus derives an experience layer configuration scheme with "humanized" constraints.

[0042] User behavior preference settings refer to configuration data that records users' subjective lifestyle habits and requirements, such as green energy-saving preferences (limiting energy consumption), health preferences (limiting radiation), or sleep quiet preferences.

[0043] The third control strategy refers to the user experience intelligent agent's derivation of experience layer configuration decision-making schemes based on user preferences, which focus on constraining device power consumption, radiation, and limiting frequent device wake-ups during specific periods.

[0044] In this embodiment of the invention, a target control strategy can be determined from the first control strategy, the second control strategy, and the third control strategy, and the target device can be controlled to perform networking operations based on the target control strategy. This is to identify and resolve the mutual exclusion of control logic caused by different optimization dimensions among the three control strategies, and to select the optimal balance scheme for collaboration among all parties through global weight calculation, and to map and deploy it to the real world, so as to realize the immediate use of physical devices.

[0045] The target control strategy refers to the optimal collaborative configuration scheme that ultimately achieves the highest comprehensive score across the three dimensions of communication, business, and experience after resolving conflicts through a multi-attribute weighted negotiation algorithm.

[0046] Networking operation refers to the unified networking process in which physical devices automatically perform channel binding, route establishment, and automated scenario rule configuration in the real physical space after receiving the configuration.

[0047] This invention improves device networking efficiency by simultaneously extrapolating a first control strategy focusing on underlying communication efficiency, a second control strategy focusing on service execution efficiency, and a third control strategy focusing on user behavior preferences in a virtual network.

[0048] For example, Suppose a user purchases a set of smart home devices (including a smart door lock, a living room light, and a smart speaker, i.e., the target devices) and plans to deploy them in their newly renovated home.

[0049] Based on the floor plan and equipment list uploaded by the user, the cloud system creates corresponding virtual door locks, virtual headlights, and virtual speakers (virtual devices) in the digital twin space and connects them into a virtual network.

[0050] Read the current parameters of the virtual door lock and find that it is working on channel 11 by default with a power level of 3 (protocol stack status parameter). Then generate a test command (exploratory configuration change command) to switch it to channel 15 and adjust the power level to 5.

[0051] The network protocol agent executes this switching command in the virtual network and finds that after switching to channel 15, the Wi-Fi interference from the neighboring house disappears and the packet loss rate drops significantly (communication efficiency improves). Therefore, it generates a first control strategy focused on "forcing the whole house to switch to channel 15 and increasing some power".

[0052] The scene-linked intelligent agent simulated a user's "coming home late at night" scenario (a preset test scenario). In this scenario, the virtual door lock sends an opening signal, the audio system requests to load a large volume of music, and the headlights report their status (generating simulated traffic). When playing back this traffic, the system also applied the change command from step three. It was found that due to the large volume of audio traffic, channel 15 experienced sudden congestion, causing the headlights not to turn on (impaired execution efficiency). To ensure a high success rate for turning on the lights, the scene-linked intelligent agent designed a second control strategy: "When channel 15 is congested, the door lock directly wakes up the headlights via a local Bluetooth backup channel."

[0053] The user experience agent retrieved the user's settings and discovered that the user highly valued "nighttime green energy saving" and "low radiation at the bedside" (user behavior preference settings). After evaluation, the user experience agent found that the first control strategy, which sets the transmission power to level 5, would severely exceed energy consumption and increase nighttime radiation. Therefore, based on energy saving and comfort indicators, it generated a third control strategy that "strictly limits nighttime standby power to no more than level 3, and allows sacrificing 5ms latency at night in exchange for low power consumption."

[0054] The central decision-making system detected a conflict among the three strategies (high network power versus low user experience). The system then invoked a weighted negotiation algorithm, combining user-defined global weights (e.g., high weights for user experience and stability, low weights for absolute latency), to comprehensively score the three strategies and ultimately select a balanced solution (target control strategy): "Switch the entire house to channel 15, allowing the use of power level 4 during the day, but forcibly reducing power to level 2 after 11 PM, while simultaneously activating the Bluetooth local redundancy channel for speakers to ensure low latency." The system translated this strategy into physical parameters and sent them to the newly powered-on physical door locks, lights, and speakers. The devices then automatically completed the optimal unified network and scene configuration according to this plan (executed network operations).

[0055] Of course, the above examples are merely illustrative. Those skilled in the art can determine the target control strategy using any other method. The embodiments of the present invention do not limit this, for example, by not directly selecting the best option when determining a conflict.

[0056] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0057] In an optional embodiment of the present invention, it further includes: Generate target household raw data, which includes environmental radio frequency characteristics, floor plan information, and target device parameters for the target device; The equipment brand combination is determined based on the target equipment parameters, and the room structure characteristics are determined based on the floor plan information; A similar scenario search is performed in the anonymized experience knowledge base to match historical configuration information that corresponds to the device brand combination and the room structure features.

[0058] This invention provides a precise cold-start networking baseline for newly connected or yet-to-be-configured smart home environments by leveraging historical experience data from a digital twin cloud. By collecting raw data such as the target family's floor plan, device parameters, and radio frequency background, it extracts the combination of device brands and room structure characteristics that represent the family's technical features. Furthermore, it performs multi-dimensional similarity searches in a massive, anonymized database of successful historical cases, thereby matching and reusing the most suitable historical configuration scheme across families. This avoids the technical overhead of debugging from scratch in a new environment and significantly improves the initialization efficiency and baseline accuracy of network pre-configuration.

[0059] The target family raw data refers to the initial multimodal dataset collected from the real physical home environment to be configured, which has not yet undergone feature extraction and abstraction processing.

[0060] Environmental radio frequency characteristics refer to the radio electromagnetic environment parameters that objectively exist in the physical space, including the background noise floor of the current frequency band and the intensity of co-channel / adjacent-channel interference from neighboring Wi-Fi networks.

[0061] Floor plan information refers to digital data of the internal spatial layout, wall locations, room areas, and relative geometric positions of a house.

[0062] The target device parameters refer to the factory and operational attributes of the physical hardware to be networked, including device type, communication protocol type, hardware version number, and brand and model.

[0063] Device brand portfolio refers to the cross-brand heterogeneous hardware integration characteristics (such as "A brand device + B brand gateway" or "a single brand in the whole house") that are abstracted after a unified statistical analysis of the brands and ecosystems of all target devices in the whole house. It is used to characterize the ecosystem compatibility barriers between devices.

[0064] Room structural features refer to key feature parameters extracted after spatial geometry and physical material analysis of the original floor plan, such as the total number of rooms, the thickness and distribution topology of load-bearing walls (high attenuation medium), and the physical isolation barrier degree from the gateway to each room.

[0065] Anonymized experience knowledge base refers to a massive database of successful historical smart home deployment cases built in the cloud, which strictly erases users' personal privacy and sensitive identity information.

[0066] Historical configuration information refers to the set of optimal network and service parameters stored in an anonymized experience knowledge base that has been verified through actual operation in similar spaces and brand ecosystems, such as historical best channel allocation tables, static routing topologies, and local linkage degradation rules.

[0067] In practical applications, the implementation process of this invention is completed collaboratively by a cloud-based digital twin engine and a local edge gateway. For example, firstly, when a user powers on smart devices and starts the configuration system in a new home, the local edge gateway automatically scans the wireless channels in the physical space to capture environmental radio frequency characteristics. Simultaneously, it reads the brand and protocol parameters of each hardware device and, together with the digital floor plan uploaded by the user, encapsulates them into the target family's original data and uploads it to the cloud. Next, the feature extraction module in the cloud server immediately performs structured parsing on these data, abstracting the complex device list into a "device brand combination" representing ecological compatibility, and transforming the floor plan into "room structural features" representing electromagnetic interference strength. Finally, the cloud engine uses these two abstract features as a joint search key to perform high-dimensional vector similarity matching in an anonymized experience knowledge base containing tens of thousands of mature home networking cases. It instantly identifies a historically successful home that is closest in terms of spatial layout and brand mix, and extracts its complete set of historical configuration information, such as stable channels, power, and service scheduling, as the initial baseline scheme for the current target home to perform virtual twin simulation and physical deployment.

[0068] This invention, by abstracting the complex "physical space" and "heterogeneous hardware ecosystem" into standardized room structure features and device brand combinations, achieves intelligent migration and data reuse of cross-home networking experience. It completely solves the technical pain point of relying on blind on-site debugging during the cold start phase of multi-device, cross-brand smart home networking due to the lack of environmental benchmarks. At the same time, by introducing an anonymized knowledge base, it provides a highly reliable initial configuration foothold for subsequent intelligent agent virtual stress testing while ensuring absolute security of user data privacy. This significantly reduces the system's exploratory iteration convergence time in the virtual space, ensuring the speed and high initial stability of the networking solution.

[0069] In an optional embodiment of the present invention, it further includes: The historical configuration information is determined as the initial seed configuration; The original data of the target family and the initial seed configuration are subjected to multi-dimensional feature fusion processing to construct a virtual digital twin model, wherein the virtual digital twin model includes, A device virtual model used to store the protocol stack state parameters of the virtual device in the virtual network; This is used to take the environmental radio frequency characteristics as initial electromagnetic background parameters, and use the initial electromagnetic background parameters in combination with the floor plan information to generate a link signal model for simulating the physical attenuation and co-channel interference scenarios when signals penetrate physical walls and spatial obstacles. A user behavior model used to output user behavior preference settings.

[0070] This invention, based on historical experience-based cold-start matching, constructs a high-fidelity, full-dimensional virtual digital twin model in the cloud, providing a closed-loop simulation sandbox for subsequent agent strategy deduction. By transforming static historical configuration information into dynamic initial seed configurations and fusing them with the target household's original data, including space, hardware, and radio frequency, multi-dimensional features are integrated. This accurately maps and visualizes real-world physical entities, electromagnetic spaces, and human activity habits into virtual device models, link signal models, and user behavior models, thereby constructing a digital twin testbed capable of accurately simulating signal attenuation, co-channel interference, and service load.

[0071] Historical configuration information refers to a set of mature networking and communication parameters that have been verified in practice under similar hardware ecosystems and spatial topologies, retrieved and matched from an anonymized experience knowledge base.

[0072] Initial seed configuration refers to using the matched historical configuration information as the initial input baseline for the current households to be networked, serving as the logical starting point for the evolution and strategy fine-tuning of the digital twin model.

[0073] The target family raw data refers to the panoramic initial multimodal dataset collected directly from the physical home environment to be deployed, including radio frequency, house type, equipment parameters, etc.

[0074] Multidimensional feature fusion processing refers to the mathematical processing process by which the cloud engine cross-correlates and aligns heterogeneous physical space features, hardware brand boundaries, radio frequency background noise, and historical seed configurations across dimensions.

[0075] A virtual digital twin model refers to a dynamic digital flow map and mirror system built in the cloud or on a computing platform that is highly consistent with the real physical home environment in terms of physical characteristics, communication links, and user behavior.

[0076] Protocol stack state parameters refer to the set of static parameters and dynamic variables of a virtual device operating at various layers of the network protocol (such as the network layer, physical layer, and MAC layer).

[0077] A device virtual model refers to a digital software image component in a digital twin model that is responsible for storing and maintaining the protocol stack state parameters and simulating the communication behavior of physical hardware.

[0078] Environmental radio frequency characteristics refer to radio electromagnetic background parameters captured from the real physical space, such as channel noise floor and co-channel interference intensity of heterogeneous networks.

[0079] Initial electromagnetic background parameters refer to the reference parameters used as the initial static background noise input in the link signal model after mathematically modeling the radio frequency characteristics of the physical environment.

[0080] Floor plan information refers to the digital data of the planar structure of a house, which describes the geometric topology of the interior space, the location of wall barriers, and the building materials.

[0081] Physical walls and spatial obstacles refer to physical media that objectively exist in the physical layout of a house and can attenuate, reflect, or cause multipath effects on radio electromagnetic waves, such as load-bearing walls, glass or metal partitions.

[0082] Physical attenuation refers to the logarithmic distance path loss variation in received signal strength (RSSI) when a radio electromagnetic signal penetrates spatial obstacles of different materials and thicknesses.

[0083] Co-channel interference refers to the deterioration of channel SNR (signal-to-noise ratio) and decoding conflicts caused by the superposition and overlap of multiple electromagnetic signals (such as neighboring Wi-Fi, microwave ovens, and radiation from concurrent devices) in the same or adjacent frequency bands.

[0084] The simulated signal attenuation interference scenario refers to a virtual radio frequency simulation environment that is dynamically rendered from a link signal model and highly replicates the physical obstacles and electromagnetic background in real space.

[0085] The link signal model refers to the physical layer simulation component in the digital twin model, which is built based on the electromagnetic wave propagation algorithm and is specifically used to simulate the attenuation and collision fluctuations of wireless signals in spatial topology.

[0086] User behavior preference settings refer to a set of digital rules that record users' subjective lifestyle habits, daily routines, energy consumption limits, and sensory comfort requirements.

[0087] User behavior model refers to a high-level behavioral simulation component in a digital twin model that is specifically responsible for reading, maintaining, and dynamically outputting the user behavior preference settings, and can generate simulated traffic for test scenarios.

[0088] In practical applications, the implementation of this invention relies entirely on the modeling service layer of the cloud-based digital twin engine. For example, the cloud server system first activates the historical configuration information retrieved in the previous step and defines it as the initial seed configuration. Then, the distributed data alignment module of the cloud engine receives the original target family data reported by the user family, transforms the floor plan into a spatial geometric grid, converts the device parameters into protocol stack initialization variables, and performs multi-dimensional feature fusion processing with the initial seed configuration. Next, the engine allocates three concurrently running software threads in memory, instantiating them into three core sub-models, thus constructing a "device virtual..." The system uses a "model" to load and track the current protocol stack state parameters of virtual devices, constructs a "link signal model" to set the environmental radio frequency characteristics as the initial electromagnetic background and renders a simulated signal attenuation interference scenario that can simulate the decibel attenuation and co-frequency conflict when the signal penetrates the physical wall, and constructs a "user behavior model" to carry and output the user's behavior preference settings at any time. Finally, these three sub-models work together in the cloud to weave a high-fidelity virtual digital twin network, so that the subsequent AI intelligent agent does not need to struggle on the physical site, but can directly execute more than a hundred kinds of extreme network change security simulations concurrently in this virtual sandbox.

[0089] This invention, through an innovative "virtual-real fusion" modeling mechanism, organically combines static historical experience (initial seed configuration) with three-dimensional real-world multimodal data (RF, apartment layout, hardware). This not only avoids the inefficiency of traditional digital twin modeling, which relies entirely on manual modeling, but also builds a three-in-one high-fidelity simulation network in the cloud, encompassing device status (virtual device model), physical electromagnetic environment (link signal model), and human habit constraints (user behavior model). The direct technical advantage of this embodiment is that, before the physical networking scheme is deployed, the system can "clear mines" by partially reproducing real-world signal attenuation and co-channel interference without interfering with real homes. This provides an extremely realistic, secure, and mathematically interpretable digital space foundation for subsequent multi-agent high-risk exploratory configuration changes and high-concurrency business scenario stress testing.

[0090] In an optional embodiment of the present invention, the step of obtaining the protocol stack state parameters of the virtual device in the virtual network and generating a tentative configuration change instruction based on the protocol stack state parameters includes: The pre-trained network protocol agent is invoked to read the protocol stack state parameters; The communication protocol type and current configuration parameters currently used by the virtual device are determined by the protocol stack status parameters. Based on the communication protocol type, retrieve the legal change boundary range corresponding to the communication protocol type from the preset protocol stack attribute database, and generate legal configuration parameter boundary range information through the legal change boundary range; Based on the current configuration parameters, a set of tentative adjustment increment values ​​for changing the current configuration parameters are calculated within the legal configuration parameter boundary range information, and a tentative configuration adjustment strategy including channel switching step size and transmit power adjustment magnitude is generated based on the tentative adjustment increment values. The tentative configuration adjustment strategy is encapsulated into a tentative configuration change instruction that conforms to the communication protocol type.

[0091] This invention employs an intelligent agent to explore the state of virtual devices under compliance constraints, providing a secure and precise action trigger source for policy deduction of the network infrastructure. By invoking a pre-trained network protocol intelligent agent, it automatically identifies the current protocol type and operating parameters of the virtual device, performs mathematical calculations within the hard compliance boundaries of a preset database, derives an exploratory configuration adjustment strategy including channel and power fine-tuning, and encapsulates it into change instructions in a standard protocol format. This provides a set of dynamic input parameters that balance "security and compliance" with "active probe" characteristics for subsequent virtual network efficiency testing, without violating protocol standards or hardware physical limits.

[0092] Pre-trained network protocol agents refer to AI core control components that run in the cloud or within a system and are pre-trained based on deep reinforcement learning or expert systems. They are specifically responsible for deep perception of the state of the underlying network communication protocol stack, action reasoning, and policy generalization.

[0093] Protocol stack status parameters refer to the real-time configuration indicators that the virtual device is currently loading and running in the network protocols at each layer, such as the current working channel number and the current transmit power in decibels.

[0094] The communication protocol type refers to the specific communication standards and technical specifications followed by the virtual device in its current network interaction, such as Wi-Fi 6, ZigBee 3.0, Bluetooth Mesh, or Matter protocols.

[0095] The current configuration parameters refer to the specific physical and logical control values ​​that are currently in effect for the virtual device under a specific communication protocol type.

[0096] The pre-built protocol stack attribute database refers to an authoritative technical database that is pre-built by the system and stores the legal operating limits, frequency band allocations, channel spacing, and hardware safe power consumption ranges of various mainstream IoT communication protocol standards.

[0097] The legal change boundary range refers to the absolutely safe range and legal frequency band limits that are allowed for dynamic adjustment by equipment according to a specific protocol standard and retrieved from the database.

[0098] The information on the boundary range of legal configuration parameters refers to the set of hard boundary rules that are generated by encapsulating the legal change boundary range in a structured manner, and are used to restrict the agent from blindly exploring (such as limiting "ZigBee channels can only be adjusted between 11 and 26, and the maximum power cannot exceed 19dBm").

[0099] The tentative adjustment increment value refers to the amount of fine-tuning change (i.e., step size change Delta value, such as channel +2, power -1dBm) calculated by the agent within the boundary range for the current configuration parameters.

[0100] The channel switching step size refers to the discrete integer intervals in the tentative adjustment increment value, which are specifically manifested as the wireless channel number jumping up or down.

[0101] The transmit power adjustment range refers to the incremental value of the tentative adjustment, specifically manifested as a continuous or stepped decibel (dBm) value in which the wireless signal transmission strength is increased or decreased.

[0102] The tentative configuration adjustment strategy refers to an unencapsulated set of action schemes generated by the agent through reasoning, which is composed of the channel switching step size and the transmit power adjustment magnitude.

[0103] The tentative configuration change command refers to the low-level control message command that can be directly parsed and executed by virtual devices or real hardware protocol stacks after the tentative configuration adjustment strategy is encapsulated according to the frame structure and control syntax of the target communication protocol.

[0104] In practical applications, the implementation of this invention runs entirely on the dynamic decision-making and control layer of the cloud-based digital twin engine. For example, firstly, the system triggers the intervention of a pre-trained network protocol agent, which directly reads the protocol stack state parameters stored in the device's virtual model through a memory data interface. It then deeply analyzes and determines that the virtual device is currently communicating using the ZigBee protocol and is currently operating on channel 11 with a power of 10dBm (current configuration parameters). Next, using this as an index, the network protocol agent retrieves the legally defined boundary range of the protocol from a pre-set protocol stack attribute database, generating a range containing "channel 11-26, power 0- The agent first obtains the legal configuration parameter boundary range information of "19dBm". Then, using its built-in exploration algorithm, combined with the current 11 channels and 10dBm parameters, the agent calculates a tentative adjustment increment of +4 for channel and +3dBm for power within the boundary range. Based on this, it generates a tentative configuration adjustment strategy containing a channel switching step size of 4 and a transmit power adjustment amplitude of 3dBm. Finally, the agent calls the corresponding ZigBee protocol packet assembly engine to translate and structure the strategy into a tentative configuration change instruction conforming to the ZigBee standard control frame format, preparing to inject it into the underlying protocol stack of the virtual network.

[0105] In an optional embodiment of the present invention, the step of applying the tentative configuration change instruction to the virtual device in the virtual network to generate a first control strategy correlated with the communication efficiency of the virtual network includes: The network protocol agent is invoked to inject the tentative configuration change command into the virtual media access control layer and virtual physical layer of the virtual network, thereby adjusting the underlying communication state of the virtual device; After the virtual device completes the adjustment of the underlying communication state, real-time data of the underlying network state is generated by capturing the physical layer transmit power, received signal strength indication and channel bit error rate of the virtual network under the action of the exploratory configuration change command. Based on the pre-acquired real radio frequency coexistence interference curve, the average packet loss rate and round-trip delay of the real-time data of the network's underlying state under the influence of the simulated signal attenuation interference scenario are calculated, and link performance evaluation data reflecting the changes in communication link quality indicators are generated; the real radio frequency coexistence interference curve is an electromagnetic interference characteristic curve that expresses the quantitative correspondence between the degree of signal-to-noise ratio degradation and the degree of communication link impairment when heterogeneous wireless signals are concurrently emitted at the same frequency. The average packet loss rate of the virtual network is determined by comparing the link performance evaluation data. When the decrease in the average packet loss rate exceeds the preset packet loss rate threshold, the current exploratory configuration change instruction is marked as a high-value action, and communication efficiency correlation data containing the target channel allocation table and device power bias value is output. The communication efficiency-related data is structured and encapsulated to generate a first control strategy.

[0106] This invention, through injection-based probing and closed-loop performance evaluation of the underlying network protocol stack, derives an optimal network layer strategy that can significantly optimize the physical transmission quality of the entire network. The system calls a network protocol agent to directly apply previously generated probing commands to the virtual media access control layer (MAC) and the virtual physical layer (PHY) of the virtual network. By capturing real-time changes in underlying signal indicators and combining them with real radio frequency coexistence interference curves for cross-calculation under adverse scenarios, the convergence performance of the average packet loss rate across the entire network is used as the evaluation benchmark to screen and extract high-value channel and power configuration combinations. This achieves the technical goal of eliminating co-channel interference and improving transmission link quality at the pure network communication level.

[0107] Network protocol intelligent agents refer to pre-trained AI components that run in the cloud and are specifically responsible for performing network communication protocol stack-level state control, data capture, and underlying policy evolution in the digital twin space.

[0108] A tentative configuration change command refers to a low-level control message command that conforms to a specific communication protocol frame structure and contains specific channel switching step size and transmit power adjustment range.

[0109] The Virtual Media Access Control (Virtual MAC) layer refers to the software layer in a digital twin network that simulates the characteristics of the physical hardware MAC protocol. It is responsible for managing virtual channel allocation, channel collision avoidance, and frame transmission timing.

[0110] The virtual physical layer (virtual PHY layer) refers to the software layer in a digital twin network that simulates the physical hardware physical layer characteristics. It is responsible for simulating the modulation and demodulation of radio frequency signals, transmission power control, and spatial mapping of electromagnetic waves.

[0111] The underlying communication state refers to the specific combination of operating parameters exhibited by the virtual device at the MAC layer and PHY layer, such as the current physical channel number and the current absolute value of the transmit power.

[0112] Physical layer transmit power refers to the energy intensity of a virtual device when it actually radiates radio electromagnetic signals into the virtual space at the physical layer, usually measured in decibels and milliwatts (dBm).

[0113] Received Signal Strength Indication (RSSI) refers to the absolute value of the virtual wireless signal target strength measured at the receiving device after spatial attenuation.

[0114] The channel bit error rate (BER) refers to the percentage of bits that are decoded incorrectly at the receiving end due to noise or interference during data transmission, out of the total number of transmitted bits. It is used to measure the original decoding quality of the link.

[0115] Real-time network underlying status data refers to a multi-dimensional raw dataset that dynamically reflects the health of the physical layer and media access layer of a virtual network, consisting of captured transmit power, RSSI, and channel bit error rate.

[0116] The real radio frequency coexistence interference curve refers to a mathematical matrix or empirical statistical characteristic curve that describes the relationship between the signal-to-noise ratio (SNR) and the degree of communication impairment when different heterogeneous wireless signals (such as Wi-Fi and ZigBee) are transmitted concurrently at the same frequency in the same space, based on tests conducted in the real electromagnetic environment of the physical world.

[0117] The simulated signal attenuation interference scenario refers to a virtual radio frequency simulation environment that is dynamically rendered by a link signal model and highly replicates the physical wall obstruction and external electromagnetic background noise in real space.

[0118] Average packet loss rate refers to the proportion of data packets that fail to reach the receiver in a virtual network within a specific statistical period, out of the total number of data packets sent by the sender.

[0119] Round-trip time (RTT) refers to the total end-to-end time consumed for a virtual data packet to travel from the sender to the receiver, where it is acknowledged and a response message is returned to the sender.

[0120] Link performance evaluation data refers to a core service performance dataset consisting of calculated average packet loss rate and round-trip delay, used to intuitively evaluate changes in network layer communication quality.

[0121] The average packet loss rate of the entire network refers to the macroscopic indicator that characterizes the overall communication quality of the entire home network after calculating the packet loss rate of all active communication links in the virtual network using global topology weighting.

[0122] The packet loss rate threshold refers to the percentage improvement limit that the system pre-sets to determine whether a certain trial configuration can bring about a "qualitative improvement" in performance.

[0123] High-value actions refer to specific exploratory configuration change commands that are determined in reinforcement learning decision-making mechanisms to significantly reduce the average packet loss rate of the entire network and have a significant positive optimization effect on the underlying communication quality of the network.

[0124] The target channel allocation table refers to the optimal wireless communication channel planning scheme that has been verified by high-value actions and enables conflict-free coexistence of heterogeneous hardware throughout the house.

[0125] The device power bias value refers to the dynamic transmit power correction Delta value set for the default transmit power of each node's hardware, which takes into account both communication distance and interference suppression.

[0126] Communication efficiency-related data refers to a dataset that maximizes the physical transmission quality of the network, consisting of a target channel allocation table and device power bias values.

[0127] The first control strategy refers to a network layer control scheme that focuses on optimizing the quality of the underlying network links and eliminating co-channel interference, generated by structuring and encapsulating communication efficiency-related data.

[0128] In practical applications, the implementation of this invention runs entirely on the business linkage simulation layer of the cloud-based digital twin engine. For example, firstly, the system calls a network protocol agent to directly inject previously encapsulated tentative configuration change instructions into the virtual MAC and PHY layers of the virtual network, forcing the virtual device to perform channel switching and power biasing, thereby adjusting its underlying communication state. Next, once the virtual device state stabilizes, the agent uses a link signal model to capture in real-time the virtual network's physical layer transmit power, received signal strength indication (RSSI), and channel bit error rate after the change, generating real-time data on the network's underlying state. Subsequently, the agent introduces pre-acquisition... The system takes real radio frequency coexistence interference curves and calculates the average packet loss rate and round-trip time of these real-time data under the influence of the current simulated signal attenuation interference scenario (such as including load-bearing wall attenuation and neighbor Wi-Fi interference) to generate link performance evaluation data. Finally, the agent calculates the average packet loss rate of the entire network and compares it with the initial benchmark. When it determines that the decrease exceeds the preset packet loss rate threshold (e.g., the packet loss rate has decreased by more than 30%), it immediately marks the current instruction as a high-value action, outputs communication efficiency correlation data containing the target channel allocation table and device power bias value, and the encapsulation engine packages it into the first control strategy, locking it as one of the optimal solutions for the underlying communication.

[0129] This invention, through the introduction of "network command injection" and "multi-dimensional link performance closed-loop evaluation" into the digital twin network, achieves in-depth security reshuffling and quality optimization of the underlying network physical configuration scheme without disturbing the user's physical network. The direct technical advantages are: by combining "real radio frequency coexistence interference curves" with "simulated signal attenuation interference scenarios," the average packet loss rate and round-trip delay calculated in virtual space have extremely high physical reliability, enabling accurate identification of truly anti-interference "high-value actions"; the final generated first control strategy can directly provide the optimal target channel allocation table and device power bias values ​​for the entire system, cutting off co-frequency conflicts and signal blind spots from the network layer, creating a stable, high-bandwidth, and ultra-low packet loss rate physical communication foundation for the entire intelligent networking system.

[0130] In an optional embodiment of the present invention, the step of reading the simulated traffic generated for a preset test scenario, and applying the tentative configuration change command to the virtual device in the virtual network when executing the simulated traffic, and generating a second control strategy that is related to the execution efficiency of the virtual device in the test scenario, includes: The pre-trained scenario-linked intelligent agent is invoked to read the simulated traffic generated by the user behavior model for the test scenario, and the target concurrent business data packets corresponding to the target user behavior are obtained. The target concurrent service data packet is input into the link signal model, and the link signal model is controlled to replay the target concurrent service data packet in order to generate a virtual radio electromagnetic signal corresponding to the test scenario in the virtual network; When replaying the target concurrent service data packets, the tentative configuration change command is applied to the virtual device in the virtual network to control the transmission of the virtual radio electromagnetic signal under the analog signal attenuation interference scenario, and to generate service scenario robustness assessment data; the service scenario robustness assessment data is used to characterize the physical resilience of the virtual device in terms of anti-interference, anti-packet loss and tolerance to latency fluctuations when the virtual device performs data transmission in the virtual network. The robustness assessment data of the business scenario is used to calculate business layer automation performance data to express the success rate and stability of scenario linkage for the test scenario; A second control strategy is generated based on the automation performance data of the business layer, which is correlated with the execution efficiency of the virtual device in the test scenario.

[0131] This invention, through simulating the dynamic superposition of high-concurrency business traffic and network changes in a digital twin space, derives a business-layer control strategy capable of ensuring highly stable operation in complex intelligent scenarios. The system invokes a pre-trained scenario-linking intelligent agent, transforming typical daily user behaviors into simulated traffic and mapping them to virtual radio electromagnetic signals. While replaying the traffic, it forcibly applies tentative configuration changes to the underlying network layer, simulating the extreme stress environment of simultaneous multi-source traffic concurrency and network parameter adjustments in reality. Through closed-loop quantitative evaluation, the system assesses the scenario's execution efficiency under the influence of physical attenuation and sudden congestion, ultimately generating a second control strategy that effectively resists network fluctuations and maximizes the success rate and robustness of automated scenario linkage.

[0132] Pre-trained scenario-linked intelligent agents refer to AI core control components that run in the cloud and are pre-trained based on deep learning or multi-agent collaborative technology. They are specifically responsible for traffic scheduling, robustness assessment, and multi-device linkage strategy inference in the upper-layer business automation scenarios of smart homes.

[0133] User behavior model refers to the behavioral simulation component in a digital twin model that is responsible for simulating, maintaining, and deriving high-level business data flow and dynamic traffic characteristics based on specific life scenarios.

[0134] Test scenarios refer to typical user life automation linkage scenarios that are pre-set by the system for stress testing of the home network, such as the "late-night home scenario" (involving the concurrent linkage of door locks, headlights, and speakers).

[0135] Simulated traffic refers to high-concurrency business data streams generated by multiple virtual devices sending and receiving data intensively at the same timestamp, derived from user behavior models for test scenarios.

[0136] Target user behavior refers to the specific actions a user takes when triggering an automated scenario in real life, such as opening a door, walking around, or pressing a trigger switch.

[0137] The target concurrent service data packet refers to the set of virtual network data packets generated by instantiating simulated traffic at the network layer, used to simulate multiple sub-devices simultaneously sending control commands or status reports to the gateway.

[0138] The link signal model refers to the physical layer simulation component in the digital twin model, which is built based on the electromagnetic wave propagation algorithm and is responsible for simulating the attenuation, multipath effect and collision fluctuation of wireless signals in the topology of a house.

[0139] A virtual network refers to a virtual communication topology built in a digital twin space to carry the simulated transmission of electromagnetic waves and the flow of data packets between various virtual devices.

[0140] Virtual radio electromagnetic signals refer to digital radio waves that are generated by inputting data packets from the service layer into the physical layer and then rendered using a link signal model to simulate the radiation of real wireless hardware into space.

[0141] A tentative configuration change command refers to a low-level control message command that conforms to a specific communication protocol frame structure and contains specific channel switching step size and transmit power adjustment range.

[0142] Virtual devices refer to digital software images created in a digital twin system to simulate the communication and business behavior of physical hardware (such as smart gateways and smart switches) in the real world.

[0143] The simulated signal attenuation interference scenario refers to a virtual radio frequency simulation environment that is dynamically rendered by a link signal model and highly replicates the physical wall obstruction and external electromagnetic background noise in real space.

[0144] Business scenario robustness assessment data refers to a dataset of macro-level stress indicators recorded and generated by the system under severe stress conditions, including sudden surges in concurrent traffic and changes in underlying configurations. This dataset reflects whether each device experiences execution delays, packet loss, or command disconnection.

[0145] Scene linkage success rate refers to the percentage of times that all preset participating sub-devices are executed completely and correctly in a specific test scenario, out of the total number of triggered tests.

[0146] Scene linkage stability refers to the dispersion (jitter range) of system response time and resilience against sudden failures and network congestion when an automated scene is triggered multiple times in a row.

[0147] Business layer automation performance data refers to a set of core indicators, composed of scenario linkage success rate and scenario linkage stability, used to intuitively and quantitatively evaluate the operational quality of upper-layer business scenarios.

[0148] Execution efficiency refers to the overall performance of the scene automation linkage rules in a complex technical environment, including response speed, throughput, and smooth operation without lag or data loss.

[0149] The second control strategy refers to the business layer configuration decision scheme derived by the scenario-linked intelligent agent under concurrent traffic stress testing, which focuses on ensuring the success rate of scenario execution, business continuity and system robustness (such as including localized degradation logic, cross-protocol redundant channel scheduling, etc.).

[0150] For example, firstly, the system invokes a pre-trained scenario-linking agent, which directly reads the simulated traffic generated by the user behavior model for a specific "late-night home test scenario" through a data interface, and converts it into a set of target concurrent business data packets simulating door locks, headlights, and speakers communicating concurrently in the same second. Next, the agent inputs this set of business data packets into a link signal model for playback, controlling the link signal model to generate corresponding virtual radio electromagnetic signals in the virtual network. Subsequently, during the tens of milliseconds of dense playback of the data packets, the agent again applies tentative configuration change commands (such as forced signal switching) generated by the network protocol agent to the virtual headlights and speakers in the virtual network. The agent controls the virtual radio electromagnetic signal to perform complex signal superposition and transmission in simulated signal attenuation and interference scenarios (such as those involving load-bearing walls), thereby capturing link changes under multi-source traffic and command collisions and generating service scenario robustness assessment data. Finally, the agent performs statistical analysis on the assessment data to calculate the scenario linkage success rate and scenario linkage stability under the dual pressure of specific channel changes and concurrent traffic, obtaining service layer automation performance data. Based on this performance data, a second control strategy is inferred and encapsulated, which includes "starting dual-transmission control frames of the Bluetooth local backup channel when channel switching is congested," ensuring peaceful coexistence between scenario operation and network changes.

[0151] This invention, through the construction of a multi-layered stress testing mechanism for "concurrent service traffic replay" and "underlying configuration changes" at microsecond-level timestamps, completely breaks the control limitation of traditional network optimization that isolates and separates "network signal quality" and "upper-layer service linkage." Its direct technical advantage lies in: by realistically reproducing the potential channel congestion and collisions caused by applying network switching commands during high-volume concurrency in virtual space, the system can accurately assess the linkage success rate and stability limits of automated scenarios under extreme dynamic environments; the resulting second control strategy can specifically supplement service layer optimization logic such as cross-protocol dual-channel redundant backup or edge degradation scheduling flows, ensuring that the final physically deployed configuration scheme not only has high underlying communication quality but also possesses strong resilience and anti-lag robustness in upper-layer scenario execution.

[0152] In an optional embodiment of the present invention, the tentative configuration change instruction includes physical layer transmit power, and the step of obtaining user behavior preference settings for the virtual device and generating a third control policy associated with the user behavior preference settings for the virtual device includes: The device wake-up frequency of the virtual device is determined by the second control strategy, and a pre-trained user experience agent is invoked to read the user behavior preference settings through the user behavior model; the user behavior preference settings include the device standby power consumption limit and daily routine time periods; The increase in standby power consumption and the change in electromagnetic radiation of the virtual device due to the second control strategy are calculated by using the physical layer transmission power, the device wake-up frequency, the device standby power consumption limit, and the daily work and rest time period. Based on the user behavior preference settings, the user's silent mode preference during a specific period is determined, and the silent mode preference is used to generate quantitative data on life disturbances that are used to evaluate the potential disturbances caused to the user by the tentative configuration change command during the specific period. A weighted scoring operation is performed on the increase in standby power consumption, the change in electromagnetic radiation, and the quantitative data of daily life disturbance to generate user sensory comfort and low-carbon energy-saving satisfaction indicators for quantifying the potential impact of the tentative configuration change instruction and the second control strategy on daily life. Based on the user's sensory comfort and low-carbon energy-saving satisfaction indicators, a third control strategy that is correlated with the user's behavioral preference settings is output.

[0153] This invention, by introducing subjective constraints and lifestyle habits from a human perspective, conducts a human-centered, multi-dimensional sensory evaluation of the aforementioned underlying communication optimization and upper-layer business scenario strategies, thereby generating a third control strategy that balances technical performance and user experience. The system invokes a pre-trained user experience agent to extract user behavior settings, including daily routines, energy consumption limits, and nighttime quiet preferences, and uses these as weighted conditions in time and space. It quantifies the increase in total standby power consumption and cumulative electromagnetic radiation changes caused by the technical adjustment scheme in real-world time and space, comprehensively measuring the potential interference of frequent network probes on users' daily lives. This establishes a hard safety and comfort red line for the networking scheme in the dimension of "human-computer interaction experience."

[0154] Pre-trained user experience intelligent agents refer to AI core control components that run in the cloud and are pre-trained based on human behavior and sensory comfort measurement algorithms. They are specifically responsible for assessing the subjective impact of technology configuration changes on people's daily life, energy consumption costs, and health radiation.

[0155] User behavior model refers to the high-level behavioral simulation component in a digital twin model that is responsible for storing, maintaining, and dynamically outputting users' subjective lifestyle habits, static preference settings, and daily routines.

[0156] User behavior preference settings refer to a set of digital rules that are actively set by users or learned and accumulated by the system, reflecting users' subjective requirements for their home environment.

[0157] The standby power consumption limit of a device refers to the maximum power or electrical energy threshold (usually measured in watts or kilowatt-hours per day) that a user or system sets for physical hardware in a non-working standby state.

[0158] Daily routine time periods refer to the time intervals on the timeline of different behavioral patterns of users and their family members, such as sleep, going out to work, and home activities, within a 24-hour period of a day.

[0159] Physical layer transmit power refers to the underlying electromagnetic energy intensity index when a virtual device radiates wireless signals into virtual space, originating from the first control strategy (network protocol layer).

[0160] Device wake-up frequency refers to the number of times a virtual device is activated from its dormant state and performs handshake communication per unit time in order to maintain a response time of at least one second, according to the second control strategy (business scenario layer).

[0161] The second control strategy refers to a business layer configuration decision scheme derived by the scenario-linked intelligent agent under concurrent traffic stress testing, which focuses on ensuring the success rate of scenario execution and system robustness.

[0162] The increase in standby power consumption refers to the total amount of additional power consumed by a virtual device over a day or a specific period of time compared to its original default configuration due to the implementation of new transmit power and wake-up frequency.

[0163] Electromagnetic radiation variation refers to the increase in electromagnetic radiation dose generated by virtual devices in a specific space and time period due to increased transmission power and increased wake-up frequency.

[0164] Special time periods refer to specific intervals within daily routines that have special environmental requirements, such as the late-night sleep period or the lunch break period.

[0165] Silent mode preference refers to the subjective tendency of users to avoid disturbances by setting a specific time period during which all hardware in the house should not produce buzzing sounds, high-frequency bright indicator lights, or frequent high-frequency radio electromagnetic interference.

[0166] A tentative configuration change command refers to a low-level control message command that conforms to a specific communication protocol frame structure and contains specific channel switching step size and transmit power adjustment range.

[0167] The quantitative data on life disturbance refers to the digital indicators calculated by the system to assess the intensity of potential mental and physiological disturbances to users' peaceful sleep at night caused by network protocols during special periods (such as sleep periods) due to frequent sending of status broadcast packets or high-frequency wake-ups.

[0168] The weighted scoring operation refers to the system assigning different mathematical weight matrices to the three dimensions of power consumption, radiation, and life interference based on the user's different subjective emphasis on "energy saving", "health" and "network speed", and then performing a comprehensive benefit evaluation process of product summation.

[0169] The user sensory comfort index is a percentage score that quantifies whether a technical solution meets the requirements of human health and psychological comfort in terms of radiation control, nighttime quietness, and non-disturbing performance.

[0170] The low-carbon and energy-saving satisfaction index refers to a percentage score that quantifies whether the total energy consumption loss brought about by the technical solution is within the user's economic budget and green and low-carbon requirements.

[0171] The third control strategy refers to the user experience agent encapsulating and outputting an experience layer control decision scheme that focuses on the upper limit constraint of physical device standby power consumption, nighttime do-not-disturb restriction rules, and allows sacrificing some network speed during sleep to achieve energy saving and low noise.

[0172] In practical implementation, the device wake-up frequency of the virtual device can be determined based on the second control strategy through the built-in dynamic transition convergence logic. For example: The second control strategy establishes the virtual device wake-up frequency based on the joint and collaborative calculation of three physical dimensions: the baseline wake-up frequency, the environmental silence attenuation coefficient, and the cumulative duration of the environment maintaining absolute silence.

[0173] Specifically, when the system detects that the home is currently in a preset working mode (such as a late-night sleep mode) and the data reported by the physical sensors throughout the house shows a high level of silence, the second control strategy executes reverse suppression regulation. At this time, the system uses the reference wake-up frequency as an initial base, and as the cumulative duration of absolute silence in the environment increases, it exhibits an inverse exponential convergence and decay pattern; furthermore, the larger the environmental silence decay coefficient, or the longer the cumulative duration of absolute silence in the environment, the greater the suppression and reduction of the reference wake-up frequency.

[0174] Finally, the execution unit performs interval mapping and rounding up on the continuous values ​​calculated by the above exponential decay with the locally preset discrete step threshold. In this way, under the premise of ensuring that the physical host is not frequently woken up and the whole machine maintains low power consumption, it adaptively determines and outputs the device wake-up frequency with a standardized time period that best meets the timeliness requirements under the current working conditions (for example, adjusting from high-frequency polling under normal conditions to routine wake-up once every 30 minutes).

[0175] In practical applications, the implementation of this invention runs entirely on the user human-computer interaction experience evaluation layer of the cloud-based digital twin engine. For example, firstly, the system calls a pre-trained user experience agent to read user behavior preference settings from the user behavior model via a data interface, extracting feature data including "standby power consumption limit is 0.2 kWh per day" and "daily work hours are 08:00-18:00 during the day and 23:00-07:00 during the night." Next, the agent uses the physical layer transmit power (e.g., increased to 15 dBm) in the first control strategy and the device wake-up frequency (e.g., 5 wake-ups per second) in the second control strategy as technical inputs. Combining the aforementioned power consumption limit and work / rest periods, it accurately calculates the increase in standby power consumption and electromagnetic radiation variation of the virtual device due to the second control strategy using a time integration algorithm. (See the example below for details on the implementation process); Subsequently, the agent locked in the special time period of 23:00-07:00 based on the user's sleep schedule and extracted the user's "strong silence mode preference". By analyzing the behavior of the network protocol sending broadcast packets at high frequency at night to test the link, the agent calculated the potential impact of this behavior on the human sleep quality in highly sensitive spaces such as the bedside and generated quantitative data on life disturbance. Finally, the agent performed multi-attribute weighted scoring on the calculated power consumption increase, radiation change value and life disturbance data, and transformed them into intuitive user sensory comfort index and low-carbon energy saving satisfaction index. Based on these two indexes, the agent inferred and encapsulated a third control strategy that includes "forcibly limiting the power to within 5dBm during the nighttime sleep period and reducing the wake-up frequency to once per minute", thus completely domesticating the networking scheme into a "humanized" strategy that conforms to human living habits.

[0176] Furthermore, to help those skilled in the art better understand how the determination of "the increase in standby power consumption and the change in electromagnetic radiation" is necessarily related to and calculated with parameters such as "daily work and rest time", the following provides a detailed explanation of the mathematical logic and algorithm implementation process: 1. Specific calculation of the increase in standby power consumption: The system is known to use a second control strategy and a trial command to increase the device's transmit power from the initial value P. base Upgraded to P new And reduce the device wake-up frequency from F in standby mode. base (times / minute) increased to F new (times / minute).

[0177] Step A (Instantaneous Power Consumption Increment Calculation): The user experience agent first calculates the difference in instantaneous power consumption ΔP (watts) between the original high-power control frame transmitted during a single wake-up and the original power consumption characteristic curve of the chip hardware, as well as the duration T of a single wake-up, based on the chip hardware's factory power consumption characteristic curve. active .

[0178] Step B (Introducing "Daily Routine Time Segmentation"): The device is not in standby mode all day. The smart agent reads the daily routine time segment and identifies that the user is working outside all day (08:00 - 18:00, a total of 10 hours). During this period, no user triggers the whole-house automation scenario, and the device is in the "absolutely unattended pure standby segment". At the same time, it identifies the nighttime sleep segment (23:00 - 07:00, a total of 8 hours), during which the device is in the "low-frequency operation standby segment".

[0179] Step C (Time Integration and Total Determination): The agent calculates the time integration of the instantaneous power consumption increment Delta P over the corresponding pure standby time period.

[0180] Increase in standby power consumption = ΔP * T active * [ (F new - F base [Total standby minutes] The total standby minutes are determined by the work and rest period. 2. The specific calculation of electromagnetic radiation variation values: The effects of electromagnetic radiation on the human body in medicine and behavioral science depend not only on the instantaneous intensity of radiation (determined by the physical layer's emission power), but also on the "spatial coexistence exposure duration" of the human body in the radiation field (determined by daily routines).

[0181] Step A (Static Space Radiation Field Modeling): The agent, based on the increased physical layer emission power P... new Based on the floor plan information, a spatial distribution map of the static electromagnetic radiation field intensity formed by the virtual device in space is rendered.

[0182] Step B (Introducing "Daily Routine Time" to Calculate Human-Machine Coexistence Exposure Duration): If the routine time shows 08:00-18:00 during the day when the user is not at home, even if the device's radiation intensity spikes due to high bandwidth or frequent wake-ups, the "human-machine coexistence exposure duration" is 0 during this period because no one is in the space, and the effective value of radiation variation to the human body is counted as 0. If the routine time shows 23:00-07:00 at night as the sleep period, the user is lying still in the bedroom for a long time (8 hours), and the smart bedside lamp in the bedroom is driving under the second control strategy to ensure constant linkage, emitting F new Its high-frequency wake-up frequency frequently radiates electromagnetic waves outwards.

[0183] Step C (Calculation of dynamic cumulative radiation variation): The agent locks the 8 hours (480 minutes) of the nighttime sleep period as a high-sensitivity time weight (assigning a weight coefficient α = 1.5), integrates and accumulates the electromagnetic radiation intensity variation value under high-frequency wake-up over the total sleep duration, and calculates the "cumulative electromagnetic radiation variation value (increase in absorbed dose)" that the user's body is subjected to during sleep.

[0184] This invention, through its innovative approach to network optimization (network quality and service linkage) at the purely technical level, establishes an intelligent experience feedback and constraint framework centered on "human living habits and comfort." This completely resolves the technical drawbacks of traditional intelligent networking solutions, which often suffer from uncontrolled device power consumption, excessive electromagnetic radiation, and frequent nighttime wake-ups that disturb residents, in pursuit of extreme communication metrics. The direct technical advantages are as follows: by placing static "physical layer transmission power" and "device wake-up frequency" within a dynamic "daily routine," multi-dimensional spatiotemporal joint calculations ensure that the assessment of standby power consumption increases and electromagnetic radiation changes is no longer based on unrealistic theoretical instantaneous values, but rather on quantifiable values ​​that closely match the user's actual home space exposure and power consumption. The resulting third control strategy establishes rigid physical parameter constraint rules with human health and low-carbon energy conservation as the top priorities, achieving a deep and harmonious synergy between communication technology efficiency, scenario robustness, and a green and healthy living environment.

[0185] In an optional embodiment of the present invention, the step of determining the target control strategy from the first control strategy, the second control strategy, and the third control strategy includes: The first control strategy, the second control strategy, and the third control strategy are aggregated into a decision set to be resolved. Feature cross-matching is performed on the decision set to be resolved. When it is determined that the transmit power adjustment action in the first control strategy, the redundancy backup action in the second control strategy, and the energy consumption upper limit constraint in the third control strategy are mutually exclusive, a multi-attribute decision conflict state is triggered. When the multi-attribute decision conflict state is detected to be triggered, the weighted negotiation algorithm based on multi-attribute decision is invoked and the preset global target weight is read. The benefits of the first control strategy, the second control strategy and the third control strategy are evaluated respectively, the comprehensive score for each control strategy is calculated, and the target control strategy is selected from the first control strategy, the second control strategy and the third control strategy based on the comprehensive score.

[0186] This invention employs intelligent resolution technology to mitigate deep logical conflicts between control strategies across the network, service, and user experience dimensions, thereby selecting the globally optimal final control scheme. The system aggregates control strategies independently derived by the three agents into a decision set. Using feature cross-matching, it accurately identifies physical and logical conflicts between "high-power communication," "high-volume service redundancy," and "low-energy human preference." Based on this, it activates a multi-attribute decision-weighted negotiation mechanism, introducing global weights to comprehensively score and compromise on the multi-dimensional benefits of each strategy's execution. This resolves conflicting interests during multi-objective optimization, selecting a target control strategy that balances transmission efficiency, scenario robustness, and a green user experience.

[0187] The first control strategy refers to the network layer configuration decision scheme derived by the network protocol agent based on changes in the underlying communication quality, which focuses on maximizing the network physical transmission efficiency (such as eliminating interference and reducing packet loss rate).

[0188] The second control strategy refers to a business layer configuration decision scheme derived by the scenario-linked intelligent agent under concurrent traffic stress testing, which focuses on ensuring the success rate and continuity of automated scenarios under extreme multi-source loads.

[0189] The third control strategy refers to the user experience intelligent agent's decision-making scheme for the experience layer, which is derived from human routines and habits and focuses on controlling the upper limit of device power consumption, reducing electromagnetic radiation, and ensuring non-interference requirements during special periods.

[0190] The decision set to be resolved refers to the comprehensive data set that collects, accommodates, and awaits conflict review and fusion processing of the three control strategies that are generated independently from different optimization dimensions.

[0191] Feature cross-matching refers to the technical means by which the central control system performs cross-dimensional correlation verification and logical intersection calculation on all control actions, operating parameters and boundary constraints in the decision set to be resolved, in order to check whether there are mutually exclusive instructions.

[0192] The transmit power adjustment action refers to the specific physical operation in the first control strategy of increasing the transmit decibel value (dBm) of the wireless hardware in order to enhance signal penetration and reduce packet loss rate.

[0193] Redundancy backup refers to the business operation proposed in the second control strategy to combat channel congestion and prevent hardware linkage failures such as headlights, by repeatedly and concurrently sending control data packets through dual channels (such as Wi-Fi and Bluetooth collaboration).

[0194] Energy consumption upper limit constraint refers to the hard control red line set in the third control strategy to limit the total power consumption and maximum standby power of hardware throughout the day or a specific period, based on human low-carbon preferences and green energy-saving requirements.

[0195] Mutual exclusion refers to a state of technological opposition where multiple opposing control actions or constraints cannot be fully satisfied simultaneously in the same space and at the same time stamp due to the limitations of objective physical laws or logical rules (e.g., high transmission power and multi-channel redundancy will inevitably lead to a surge in energy consumption, thereby directly destroying the low energy consumption constraint).

[0196] Multi-attribute decision conflict state refers to a logical alarm state that is automatically triggered and marked by the system when feature cross-matching determines that the actions and constraints of multiple strategies are incompatible and contradictory. This state indicates that "the system is stuck in a multi-objective optimization deadlock and needs to start a higher-level algorithm to compromise interests."

[0197] Weighted negotiation algorithms based on multi-attribute decision-making refer to a mathematical decision-making model built into the system that is specifically designed to solve multi-objective and multi-criteria conflict problems, such as the Analytic Hierarchy Process (AHP) or fuzzy comprehensive evaluation algorithm. It enables conflicting parties to make concessions and maximize their interests based on their weights.

[0198] The preset global target weights refer to the mathematical weight vector matrix that is preset by the system factory or assigned by the user according to their own family priorities (such as network speed, energy saving or stability) to balance the relative importance of the three major indicators of communication quality, service continuity and sensory comfort.

[0199] Benefit assessment refers to the mathematical modeling process that combines negotiation algorithms with global target weights to quantitatively calculate the overall positive technical value (benefit minus cost) that each control strategy can bring to the whole-house network system when it is fully or partially accepted.

[0200] The overall score refers to the standardized scalar score that represents the contribution of each control strategy to the overall system after benefit evaluation and weighted calculation.

[0201] The target control strategy refers to the optimal integration solution that achieves the highest overall score among the three conflicting strategies and achieves a synergistic balance and mutual compromise in the three dimensions of communication, business, and experience.

[0202] In practical applications, the implementation of this invention runs entirely on the central conflict resolution control layer of the cloud-based digital twin engine. For example, firstly, the system packages and mounts the first control strategy output by the network protocol agent, the second control strategy output by the scene linkage agent, and the third control strategy output by the user experience agent into a unified decision set to be resolved. Next, the central controller's feature cross-matching module is activated, performing matrix intersection operations on the specific parameters included in these three strategies. It determines that the transmission power adjustment action in the first strategy (aiming to increase the transmission power to 18dBm) and the redundant backup action in the second strategy (aiming to enable cross-protocol dual-transmission control frames) are directly conflicting with the energy consumption limit constraint in the third strategy (the maximum power during the nighttime sleep period must not exceed 5W) based on physical laws, and therefore cannot be satisfied simultaneously. This instantly triggers a multi-attribute decision conflict state. Subsequently, upon detecting this conflict state, the system immediately invokes a weighted negotiation algorithm based on multi-attribute decision-making and retrieves the preset global target weights set in user preferences: "50% for nighttime sleep experience, 30% for business stability, and 20% for underlying absolute network speed." Finally, the algorithm evaluates the effectiveness of the three strategies under the current weight matrix. It finds that fully adopting the first strategy will result in a significant deduction in experience due to excessive energy consumption, while fully adopting the third strategy will result in a deduction in underlying score due to slightly slower network speed. Ultimately, through a comprehensive benefit maximization game, the system calculates the comprehensive scores for each strategy and selects the target control strategy with the highest comprehensive score, which compromises and integrates the parameters into "moderate power adjustment to 10dBm, local short retransmission only for door locks and night lights, and silent speed reduction for high-bandwidth speakers at night."

[0203] This invention, through innovative "feature cross-conflict detection" and "multi-attribute weighted negotiation mechanism," completely solves the technical pain points of conflicting strategies and control failures caused by independent decision-making and a lack of overall perspective among various subsystems (network, services, user habits) in traditional smart home configuration or network optimization. The direct technical advantage lies in the fact that the system no longer blindly and uniformly favors a single extreme communication indicator. Instead, it elevates the conflict between "high power," "high redundancy," and "hard energy-saving red lines" to a "multi-attribute decision-making conflict state," using scientific mathematical algorithms and global target weights for benefit evaluation. The final selected target control strategy is a "greatest common divisor" solution that maximizes compromise among all parties and overall benefit. This avoids both uncontrolled energy consumption and excessive radiation due to blindly pursuing network quality, and the embarrassing situation of scene lag and connection failures caused by excessive energy saving, achieving deep synergy between technological energy efficiency and human livability.

[0204] In an optional embodiment of the present invention, the step of controlling the target device to perform networking operations based on the target control strategy further includes: The target control strategy is transformed into a set of physical configuration parameters that include the target channel number, node routing table, local degradation linkage logic, and maximum power limit threshold. The physical configuration parameter set is pushed into the physical register of the target device to control the target device to perform networking operations based on the physical configuration parameter set.

[0205] This invention enables smart home networks to be used immediately upon installation by thoroughly integrating and securely deploying high-level abstract decision-making schemes derived in the cloud or digital twin space onto real-world hardware entities. The system performs deep compilation and low-level protocol translation on the target control strategy obtained after conflict resolution, transforming it into a digital parameter matrix (i.e., a physical configuration parameter set) that can be directly recognized and executed by physical hardware chips. This parameter matrix is ​​then silently pushed to the underlying physical registers of the target device via an efficient data transmission channel. This seamlessly maps the optimal derivation results from the virtual world into network operations performed by the physical device, including channel binding, routing establishment, and automated operation, ensuring the complete implementation of the virtual-physical control loop.

[0206] The target control strategy refers to the collaborative optimal networking configuration decision scheme with the highest global benefit after resolving conflicts between communication, services and experience through a multi-attribute weighted negotiation algorithm.

[0207] The target channel number refers to the anti-interference and conflict-free wireless operating frequency band number that wireless communication hardware (such as wireless network cards and RF transceiver chips) should be bound to in the real physical space, specifically defined in the physical configuration parameter set.

[0208] A node routing table is a data structure mapping table specifically contained in the physical configuration parameter set, used to guide how hardware data packets of each child node in a heterogeneous whole-house network should be transmitted, avoid congestion, and establish the best forwarding path.

[0209] Local degradation linkage logic refers to the emergency survival control rules specifically defined in the physical configuration parameters. When extreme technical conditions such as sudden severe network disconnection or cloud unavailability occur, the physical hardware automatically switches to relying on local area networks such as local Bluetooth and single-point broadcast to maintain the operation of core automated scenarios.

[0210] The maximum power limit threshold refers to the upper limit of the maximum wireless signal transmission power (dBm) that is specifically defined in the physical configuration parameters and is forcibly added to the radio frequency chip to meet the user's low-carbon energy-saving preferences and health radiation requirements.

[0211] The physical configuration parameter set refers to the binary or hexadecimal underlying physical configuration dataset generated by uniformly compiling the target channel number, node routing table, local degradation linkage logic, and maximum power limit threshold, which can be directly parsed by the underlying firmware and microcontroller of the hardware device.

[0212] Target devices refer to various smart home hardware entities that actually exist in the real home physical environment and are awaiting final network and scenario deployment.

[0213] Physical registers refer to the lowest-level high-speed storage units in the target device's internal communication chips (such as SoC chips, RF control chips) or microcontrollers (MCUs) used to directly store core control instructions and determine the hardware's operating state.

[0214] Push refers to the process by which a system silently and losslessly transmits a set of physical configuration parameters to a hardware device using over-the-air (OTA) technology or an encrypted local area network communication protocol.

[0215] Networking operations refer to the unified network deployment process that automatically executes in the real physical space after physical devices change their physical register parameters, including channel switching, topology reconstruction, neighbor node binding, and the readiness of automated scenario rules.

[0216] For example, firstly, the system's control module receives the target control policy output after multi-attribute weighted negotiation, calls the preset driver translation engine, decomposes the high-level abstract policy layer by layer, and compiles it into a set of physical configuration parameters including "switch to the 15th wireless channel (target channel number)", "establish the shortest hop topology from the door lock to the gateway (node ​​routing table)", "when the channel suddenly becomes congested, the door lock directly wakes up the night light via Bluetooth (local degradation linkage logic)", and "the transmit power during the nighttime sleep period is absolutely prohibited from exceeding 8dBm (maximum power limit threshold)". Then, the system communicates with the target via the home host. A dedicated pre-configured wireless security channel is established between devices, and a distribution thread is initiated to silently push the physical configuration parameter set into the underlying physical register of the target device in the form of an encrypted data stream, directly overwriting the original factory default value. Finally, the control firmware of the target device (such as a physical smart door lock or smart gateway) detects that the register value has been rewritten and immediately triggers the underlying hardware RF phase-locked loop and state machine reset. Based on this new parameter set, it automatically performs networking operations such as channel switching, routing table loading, and local backup link establishment in the physical world, completing the network setup process of the new home network without manual blind debugging and is optimal as soon as it is powered on.

[0217] This invention addresses the most crucial step in the "virtual-real integration" control loop of digital twins, enabling the efficient and secure implementation of a perfectly simulated virtual solution in the physical hardware world. It completely resolves the technical pain points of traditional smart home network configurations, which rely entirely on manual on-site hardware adjustments or where the issued solutions are incompatible with the hardware. The direct technical advantages are as follows: by finely transforming complex global optimization strategies into a set of physical configuration parameters including "target channel number," "node routing table," "local degradation linkage logic," and "maximum power limit threshold," the final control commands simultaneously consider the underlying wireless spectrum planning, network topology stability, service self-healing capabilities during sudden network outages, and power consumption radiation limits from a human perspective. Furthermore, through a silent push mechanism that directly rewrites "physical registers," hardware devices can reconstruct the underlying network operations without being aware of the changes, resulting in extremely high deployment efficiency and physical-level stability during operation.

[0218] In an optional embodiment of the present invention, it further includes: The link signal model outputs a wall attenuation coefficient to reflect the physical attenuation status, and an environmental noise floor to reflect the co-frequency interference status. Determine the target communication link in the virtual network; The received signal strength indication, average round-trip time delay, and link layer retransmission count of the target communication link are collected at preset time intervals to generate real-time link measurement data reflecting the current state of the target communication link. Read the theoretical prediction data output by the link signal model and the user behavior model at the same timestamp, and determine the absolute value of the deviation between the real-time measured data and the theoretical prediction data; When the absolute value of the deviation exceeds a preset threshold for a continuous preset period, the wall attenuation coefficient and the environmental noise floor are fitted and calculated to update the link signal model. After updating the link signal model, the network protocol agent, the scenario linkage agent, and the user experience agent are triggered to perform incremental virtual testing on the deviation area between the real-time link measured data and the theoretical prediction data, so as to output a fine-tuning configuration scheme for the target device.

[0219] This invention, through the construction of a data-driven dynamic closed-loop feedback and local incremental calibration mechanism, ensures that the digital twin system maintains a high degree of "virtual-real synchronization" with the complex and ever-changing physical world during the operation of IoT systems such as vehicles or homes. The system collects real-time, multi-dimensional communication data from physical links and performs rolling difference comparisons with theoretical predictions from the cloud-based twin engine at the same timestamp. This allows for accurate detection of control failure risks caused by drastic environmental changes (such as sudden electromagnetic interference or changes in physical obstacles). By triggering an online fitting algorithm to dynamically update underlying environmental parameters and driving multiple agents to perform low-computational-power incremental virtual tests only on local "deviation areas," the system quickly outputs accurate online fine-tuning configuration schemes with minimal computational cost, completely eliminating the technical limitations of static network configurations in adapting to the dynamic evolution of the physical environment.

[0220] The link signal model refers to the physical layer simulation component in the digital twin model, which is built based on electromagnetic propagation and radio frequency impairment algorithms and is specifically used to simulate the attenuation and collision fluctuations of wireless signals in spatial topology.

[0221] Physical attenuation refers to the path logarithmic loss of the received signal strength when a radio electromagnetic signal penetrates spatial obstacles (such as load-bearing walls and metal structures) of different materials and thicknesses.

[0222] The wall attenuation coefficient refers to a quantitative continuous mathematical characteristic value (decibels, dB) in the link signal model used to characterize the degree of obstruction and absorption of radio electromagnetic wave energy by a specific physical obstacle.

[0223] Co-channel interference refers to the phenomenon of channel signal-to-noise ratio degradation and decoding conflict caused by the superposition and overlap of heterogeneous wireless signals and multi-source burst traffic in the same or adjacent frequency bands.

[0224] The ambient noise floor refers to the basic thermal noise power spectral density value within the wireless communication frequency band, which is composed of background stray electromagnetic waves, heterogeneous network interference, and thermal noise from electronic components.

[0225] A virtual network refers to a virtual communication topology constructed in a digital twin space, used to simulate electromagnetic signal transmission and data interaction between nodes with high fidelity.

[0226] A target communication link refers to a specific end-to-end communication channel (such as from a gateway to a specific terminal device) that is specified by the system, operates synchronously in the real physical world and virtual network, and is a key object of observation and optimization.

[0227] The preset time interval refers to the rolling time period set in advance by the system for periodically triggering physical layer data acquisition and performance reporting (e.g., once every 10 seconds).

[0228] The Received Signal Strength Indicator (RSSI) refers to the actual physical wireless signal strength value measured at the receiving device after spatial attenuation.

[0229] Average Round-Trip Time (RTT) refers to the expected end-to-end time from when a data packet is physically sent by the sender to when it is received by the receiver and a response acknowledgment message is finally delivered to the sender.

[0230] Link layer retransmission count refers to the cumulative frequency at which the sender is forced to retransmit the data frame to ensure data delivery in order to ensure that the data is delivered, due to channel noise, packet collisions, or bit errors causing the receiver's verification to fail.

[0231] Real-time link test data refers to a multi-dimensional raw operational status dataset that is captured and encapsulated directly from real target communication links in the physical world at preset time intervals. It consists of received signal strength indication, average round-trip time delay, and link layer retransmission count.

[0232] User behavior model refers to the high-level behavioral simulation component in a digital twin model that is responsible for simulating, maintaining, and dynamically outputting human daily habits and business control loads.

[0233] Same timestamps refer to absolute or relative time points (or time periods) that are perfectly aligned on the timeline. This is used to ensure that the environment background of cloud simulation (such as concurrent user behavior at a certain second or specific time noise) is completely consistent with the real background when the physical world collects measured data.

[0234] Theoretical prediction data refers to the theoretical network indicators (theoretical received signal strength index RSSI, theoretical round-trip time, and theoretical retransmission count) simulated by the cloud-based digital twin engine using the current link signal model and user behavior model at the same timestamp for the target communication link.

[0235] The absolute value of deviation refers to the mathematical difference between the various indicators in the real-time link measured data and the corresponding theoretical prediction data, obtained by taking the absolute value of the difference, and is used to characterize the "degree of deviation between virtual and real".

[0236] The continuous preset period refers to the observation window set by the system to filter sudden and occasional jitter. It requires that the absolute value of the deviation must exceed the threshold for multiple consecutive collection periods to be considered a valid anomaly (e.g., 5 consecutive periods).

[0237] The preset threshold refers to the hard deviation threshold set in advance by the system to distinguish between normal network fluctuations and severe electromagnetic / space environment changes (such as the RSSI deviation exceeding 6dBm, or the retransmission number deviation exceeding 20%).

[0238] Fitting calculation refers to the process by which a mathematical optimization algorithm, run by an online regression estimator, uses the collected deviation data as input to deduce and approximate the physical continuous parameter values ​​that best match the actual electromagnetic variations.

[0239] The update refers to replacing the old dynamic environmental parameters in the link signal model with the latest wall attenuation coefficient and environmental noise baseline obtained from fitting calculations (usually using an exponential moving average method), so that the underlying physical parameters of the digital twin model are "aligned" with the latest real-world state.

[0240] Network protocol intelligent agent, scene linkage intelligent agent, and user experience intelligent agent refer to the three pre-trained AI agent control components that run concurrently within the system, focusing on underlying network efficiency optimization, upper-layer scene linkage robustness assurance, and human sensory comfort and energy-saving preference constraints, respectively.

[0241] Deviation areas refer to the localized affected software and hardware domains that first manifest as excessive differences at the data layer and are further mapped by the cloud to specific physical topology locations (such as the area around the partition wall between the living room and bedroom) and specific scenario-linked automated business flows.

[0242] Incremental virtual testing refers to a system that, after discovering a discrepancy between virtual and real systems, refuses to perform a full simulation in order to save computing resources. Instead, it commands each agent to conduct local virtual stress tests and performance re-simulations for the affected "deviation area" by introducing updated environmental parameters.

[0243] The target device refers to an IoT hardware terminal entity in the real physical environment that is affected by deviations and needs to receive the latest instructions to adjust its own state.

[0244] The fine-tuning configuration scheme refers to the on-demand configuration scheme that, after the three intelligent agents have undergone incremental virtual testing in the deviation area, performs local fine-tuning and parameter offsetting on the original target control strategy (such as local channel fine-tuning and nighttime instantaneous power consumption bias correction), and outputs a scheme that can quickly eliminate virtual and real deviations and restore the system to its optimal operating state.

[0245] In practical applications, the implementation of this embodiment relies entirely on the dynamic data flow closed-loop calibration layer of the cloud-based digital twin engine during operation. For example, firstly, during system operation, the control link signal model outputs a wall attenuation coefficient (initially set to 12dB) reflecting physical attenuation and an environmental noise floor (initially set to -95dBm) reflecting co-channel interference, and locks the "living room gateway to bedroom smart curtains" as the target communication link in the virtual network. Then, the local edge gateway collects data on the real-world conditions of this link every 10 seconds (a preset time interval). The Received Signal Strength Indicator (RSSI), average round-trip time (RTT), and link layer retransmission count are packaged into real-time link measurement data and reported to the cloud. Subsequently, the cloud-based digital twin engine retrieves the link signal model and user behavior model at the same timestamp (at this point, the model is simulating a traffic environment with users sleeping at night and smart bedside lamps operating concurrently), calculates the theoretical prediction data corresponding to that moment, and calculates the difference between the measured data and the theoretical prediction data to determine the absolute value of the deviation. Once it is found that the absolute value of the deviation exceeds a preset threshold (e.g., measured RSSI) for five consecutive collection cycles (a consecutive preset cycle), the system will take action. (The signal was consistently 10dBm lower than the theoretical value, and the number of retransmissions increased threefold). The system then determined that a drastic change in the physical environment had occurred (such as a user placing a new metal wardrobe in the wall or a neighbor turning on high-power Wi-Fi). It immediately triggered an online regression estimator to perform inverse fitting calculations on the bedroom wall attenuation coefficient and the ambient noise floor. By using an exponential moving average, the wall attenuation coefficient was updated from 12dB to 20dB, thereby updating the link signal model. Finally, after aligning the underlying parameters of the model, in order to save computing power, the system refused to recalculate the entire house. Instead, it locked the affected bedroom space and related late-night sleep services as the "deviation area." It triggered the pre-trained network protocol agent, scene linkage agent, and user experience agent to conduct incremental virtual stress and comfort tests only on this deviation area. Within a few hundred milliseconds, it quickly inferred and output a local fine-tuning configuration scheme that included "switching the link to the backup Bluetooth channel at night and allowing the maximum power consumption limit to be slightly increased by 1dBm" and silently sent it to the target device, realizing the network's online accurate self-healing in complex dynamic environments.

[0246] The embodiments of the present invention endow the entire digital twin networking system with the ability of "active correction" and "precise self-healing" when facing complex dynamic environments during operation, and completely solve the major technical pain point of traditional pre-configuration technology, which is prone to the failure of the predetermined configuration and network slowdown due to dynamic and unpredictable changes in the physical environment (such as furniture moving or external electromagnetic surges).

[0247] To enable those skilled in the art to better understand the embodiments of the present invention, an example is used below to illustrate the embodiments of the present invention.

[0248] refer to Figure 2 , Figure 2 This is a flowchart illustrating a device networking method provided in an embodiment of the present invention; 1. Data Acquisition and Construction of Self-Evolving Digital Twin Models: Home Host Initialization: When a user first starts up their home host after purchasing it, the host scans the surrounding environment through its built-in multi-protocol communication module (Wi-Fi, Bluetooth, Zigbee, etc.) or guides the user to upload a list of existing devices and floor plan information in their home via a mobile app.

[0249] Twin Model Generation: The home host uploads the collected data to a cloud-based digital twin engine. This engine uses this data to construct a virtual digital twin model that corresponds 1:1 to the user's home. This model comprises three core sub-models: - Virtual device model: Create a virtual copy for each device in the list (such as lights, air conditioners, sensors), accurately recording its model, supported communication protocols, functional characteristics, etc.

[0250] - Link signal model: Based on the collected radio frequency data (such as signal strength and signal-to-noise ratio), a signal propagation model is established for the communication link between devices to accurately simulate the attenuation and interference of signals in physical environments such as walls and furniture.

[0251] - User behavior model: By analyzing users' usage habits (such as daily routines and scenario preferences), we can construct the activity patterns of family members and provide real user behavior data for subsequent traffic replay.

[0252] Self-evolution initiation and knowledge transfer: After the model is built, it is not statically stored, but enters a "self-evolution" state. The system retrieves the most similar deployed cases to the current household from the cloud knowledge base, uses their successful configuration as the initial seed, and pre-corrects the link signal model and user behavior model, so that the model has higher prediction accuracy before the first virtual test.

[0253] 2. Multi-agent cooperative testing and optimization: Multi-specialized Agent Deployment: After the digital twin model is built, the system automatically deploys three types of AI agents: network protocol agent, scenario linkage agent, and user experience agent. Each agent has its own independent optimization goals and shares intermediate results through a "blackboard".

[0254] Protocol stack-level virtualization test: The network protocol agent first performs protocol stack-level virtualization on the devices in the model to accurately simulate the behavior of protocols such as Zigbee, Wi-Fi, and Matter at the physical layer, link layer, network layer, transport layer, and application layer, and reproduce the interaction and conflict between different protocols (such as channel contention between Zigbee and Wi-Fi).

[0255] The specific implementation process is described as follows: After the digital twin model is built, the system first initializes the AI ​​Agent as a policy network with a clear optimization objective. This Agent does not generate the final configuration all at once, but rather adopts a cyclical working mode of "trial and error—evaluation—iterative improvement." The specific operation steps are as follows: First, the Agent reads the protocol stack state parameters of each virtual device in the digital twin model, including the currently used channel number, physical layer transmit power level, MAC layer retransmission counter threshold, and application layer message transmission interval. Second, the Agent actively applies a set of tentative configuration changes in the virtual environment, such as switching the Zigbee coordinator's channel from the default channel 11 to channel 15, while simultaneously reducing the transmit power of the Wi-Fi access point by one level, and recording the changes in communication quality indicators of all links in the virtual network after the changes. Third, the system calls the pre-built protocol stack interference analysis module, which can calculate the actual increase in the bit error rate of the Zigbee signal under the influence of Wi-Fi signal sideband leakage based on the actual radio frequency coexistence interference curve. Fourth, the Agent compares the performance change data obtained from this trial with historical trial records, using a built-in reinforcement learning evaluation function to determine whether the configuration adjustment direction is positive or negative. If a channel switching action reduces the average packet loss rate of the entire network by more than 2%, the action is marked as a high-value action, and the Agent will prioritize using similar action combinations in subsequent trials. Fifth, the above process is automatically executed in a loop in the digital twin environment at a speed tens of times faster than real time. Typically, after about two hundred to five hundred trial iterations, the Agent's optimization curve tends to converge. At this point, the system extracts the channel allocation table, device power bias values, and multicast group partitioning scheme in the converged state as the optimal pre-configuration scheme output. The entire process does not require manual intervention; the Agent gradually approaches the global optimum through hundreds of virtual trials, rather than statically configuring based on a preset rule base.

[0256] Collaborative exploration and conflict negotiation: - Network Protocol Agent: By simulating scenarios such as channel interference and signal attenuation, it proposes optimization suggestions such as channel selection and transmit power adjustment, and records the key decision-making basis.

[0257] - Scene Linkage Agent: Based on the suggestions of the network agent, the traffic generated by the user behavior model is replayed in a virtual environment, and faults such as device offline and network latency are injected to test the success rate and robustness of scene linkage, and the optimization logic is recorded.

[0258] - User Experience Agent: Based on user behavior models and preference settings (such as silent mode at night), assess the impact of optimization solutions on daily life and provide feedback to the former two.

[0259] - Conflict Negotiation: When conflicting suggestions exist between different agents (e.g., the network agent requests reduced power to save energy, while the scenario agent requests increased power to ensure successful linkage), the system initiates a negotiation mechanism to find the optimal balance point based on a preset global objective (e.g., "overall satisfaction first"). The negotiation process records all key decision-making basis for subsequent interpretation.

[0260] 3. Explainable decision-making and one-click deployment: Decision visualization: After optimization, the system generates an interpretable report, which includes: the final configuration scheme, the decision path of each agent (e.g., "The network agent chose channel 11 because it has the lowest overlap with the neighbor's Wi-Fi"), the trade-off explanation (e.g., "This scheme increases the linkage success rate from 92% to 98%, but increases energy consumption by 3%), and the alternative schemes that were not adopted and their advantages and disadvantages.

[0261] User interaction and confirmation: Users can view the report through the home console screen or mobile APP, and can manually adjust the optimization weights (such as prioritizing energy efficiency or performance). The system will then fine-tune the configuration based on user feedback.

[0262] One-click deployment: After confirmation, the configuration plan is automatically sent to the home host. When the user installs and powers on the physical devices, the home host automatically performs unified network and scene configuration for all devices according to the pre-configured plan. The entire process requires no manual intervention and is ready to use immediately.

[0263] 4. Virtual-Real Synchronous Closed Loop and Continuous Evolution: Real-time data feedback: During operation, the physical home host continuously collects real network performance data (such as device response time, packet loss rate, and power consumption fluctuations) and synchronizes it to the digital twin model in the cloud in real time through an encrypted channel.

[0264] Deviation-triggered incremental testing: The twin model compares expected performance with actual performance. When a deviation exceeds a preset threshold (e.g., a drop in linkage success rate exceeding 5%), the relevant AI Agent is automatically triggered to perform incremental virtual testing. Incremental testing only re-optimizes the problem area, rather than recalculating the entire dataset, thereby reducing computational overhead.

[0265] Detailed Implementation Process: The closed-loop correction mechanism of the self-evolving digital twin model operates throughout the entire lifecycle of the home host. Its implementation process is divided into three stages: "difference detection—local correction—smooth update." In the difference detection stage, the physical network management module of the home host reports a set of actual observations of key links to the cloud at preset intervals (e.g., ten minutes). These observations include the received signal strength indication from the selected reference device to the gateway, the average round-trip time, and the number of link-layer retransmissions. Simultaneously, the cloud-based digital twin engine runs a lightweight simulation thread to calculate the theoretical prediction value of the corresponding link at the same timestamp. The system compares the measured value with the predicted value of the same link. When the absolute value of the deviation exceeds a preset threshold for three consecutive reporting cycles (e.g., signal strength deviation exceeding five dB / mW, or packet loss rate deviation exceeding five percent), it is determined that the digital twin model of that local area has significantly drifted from the actual environment, and the correction process is triggered. In the local correction stage, the system does not retrain the global model but only adjusts the parameters of the drifting communication link and its adjacent spatial regions. Specifically, the system extracts recently accumulated deviation data sequences and inputs them into an online regression estimator. This estimator is specifically designed to fit the changes in two key environmental parameters: the wall attenuation coefficient and the ambient noise floor. For example, when the signal strength reported by the smart socket in the master bedroom is consistently weaker than expected, the system infers that the RF attenuation factor of the load-bearing wall between the master bedroom and the living room gateway has increased (possibly due to the addition of large metal furniture). It then multiplies the attenuation coefficient of this wall in the link signal model by a correction factor. During the smooth update phase, the corrected environmental parameters do not instantly overwrite the original parameters but are gradually replaced using an exponential moving average to avoid model oscillations caused by single abnormal measurements. After the update is complete, subsequent predictions from the digital twin model will automatically apply the new environmental parameters, thus re-aligning with the actual physical world. This process is fully automated and imperceptible to the user, achieving the effect of the digital twin model evolving synchronously with changes in the real home environment.

[0266] Silent optimization and configuration updates: Incremental testing generates fine-tuning solutions, which are then silently deployed to the home host via OTA (Over-The-Air) updates, requiring no user intervention. Simultaneously, optimization results are fed back to the self-evolving model, updating the link signal model or user behavior model to make the model more accurate.

[0267] 5. Knowledge Accumulation and Cross-Scenario Transfer: Experience knowledge base construction: An anonymized experience knowledge base is built in the cloud to store key characteristics of each family's optimization process (house type, number of rooms, equipment list, network topology, final configuration scheme, performance indicators) and optimization decision-making trajectory. All data has been anonymized to ensure user privacy.

[0268] Similar scenario retrieval: When a new user is deployed, the system retrieves the most similar historical cases from the knowledge base based on their family characteristics (such as equipment brand combination and room structure) and extracts their successful configurations as initial seeds.

[0269] Rapid seed validation: Seed configurations are directly loaded into the digital twin model of the new home. The AI ​​Agent only needs to perform a few validation tests and fine-tuning to generate the final configuration. As the knowledge base continues to grow, the deployment time for the new home is reduced from minutes to seconds, creating a data flywheel effect.

[0270] Continuous iteration: The optimization results of each family are also fed back into the knowledge base after anonymization, continuously enriching the knowledge reserves and making the system smarter with use.

[0271] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0272] Reference Figure 3 The diagram illustrates a structural block diagram of a device networking apparatus provided in an embodiment of the present invention, which may specifically include the following modules: The virtual network construction module 301 is used to determine the virtual device corresponding to the target device and construct a virtual network for enabling data communication between multiple virtual devices; The tentative configuration change instruction reading module 302 is used to obtain the protocol stack status parameters of the virtual device in the virtual network, and generate a tentative configuration change instruction through the protocol stack status parameters; The first control strategy generation module 303 is used to apply the tentative configuration change command to the virtual device in the virtual network and generate a first control strategy that is related to the communication efficiency of the virtual network. The second control strategy generation module 304 is used to read the simulated traffic generated for a preset test scenario, and when the simulated traffic is executed, to apply the tentative configuration change command to the virtual device in the virtual network, and generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario. The third control strategy generation module 305 is used to obtain user behavior preference settings for the virtual device and generate a third control strategy that is related to the user behavior preference settings for the virtual device. The device networking control module 306 is used to determine a target control strategy from the first control strategy, the second control strategy and the third control strategy, and control the target device to perform networking operations based on the target control strategy.

[0273] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0274] In addition, embodiments of the present invention also provide an electronic device, such as... Figure 4 As shown, it includes a processor 401, a communication interface 402, a memory 403, and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404. Memory 403 is used to store computer programs; When processor 401 executes the program stored in memory 403, it implements any of the device networking methods described in the above embodiments: The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0275] The communication interface is used for communication between the aforementioned terminal and other devices.

[0276] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0277] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0278] like Figure 5 As shown, in another embodiment of the present invention, a computer-readable storage medium 501 is also provided, which stores instructions that, when executed on a computer, cause the computer to perform the device networking method described in the above embodiment.

[0279] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described device networking method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0280] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0281] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0282] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0283] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A device networking method, characterized in that, include: Identify the virtual devices corresponding to the target device and construct a virtual network to enable data communication between multiple virtual devices; Obtain the protocol stack status parameters of the virtual device in the virtual network, and generate a tentative configuration change command based on the protocol stack status parameters; In the virtual network, the tentative configuration change command is applied to the virtual device to generate a first control strategy that is correlated with the communication efficiency of the virtual network; Read the simulated traffic generated for a preset test scenario, and when executing the simulated traffic, apply the tentative configuration change command to the virtual device in the virtual network to generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario; Obtain user behavior preference settings for the virtual device, and generate a third control strategy that is related to the user behavior preference settings for the virtual device; A target control strategy is determined from the first control strategy, the second control strategy, and the third control strategy, and the target device is controlled to perform networking operations based on the target control strategy.

2. The method according to claim 1, characterized in that, Before the step of determining the virtual device corresponding to the target device and constructing a virtual network for enabling data communication between multiple virtual devices, the method further includes: Generate target household raw data, which includes environmental radio frequency characteristics, floor plan information, and target device parameters for the target device; The equipment brand combination is determined based on the target equipment parameters, and the room structure characteristics are determined based on the floor plan information; A similar scenario search is performed in the experience knowledge base to match historical configuration information that corresponds to the device brand combination and the room structure features.

3. The method according to claim 2, characterized in that, Also includes: The historical configuration information is determined as the initial seed configuration; The target family's original data and the initial seed configuration are subjected to multi-dimensional feature fusion processing to construct a virtual digital twin model, wherein the virtual digital twin model includes: a device virtual model for storing the protocol stack state parameters of the virtual device in the virtual network; This is used to take the environmental radio frequency characteristics as initial electromagnetic background parameters, and use the initial electromagnetic background parameters in combination with the floor plan information to generate a link signal model for simulating the physical attenuation and co-channel interference scenarios when signals penetrate physical walls and spatial obstacles. A user behavior model used to output user behavior preference settings.

4. The method according to claim 3, characterized in that, The step of obtaining the protocol stack status parameters of the virtual device in the virtual network and generating a tentative configuration change command based on the protocol stack status parameters includes: The pre-trained network protocol agent is invoked to read the protocol stack state parameters; The communication protocol type and current configuration parameters currently used by the virtual device are determined by the protocol stack status parameters. Based on the communication protocol type, retrieve the legal change boundary range corresponding to the communication protocol type from the preset protocol stack attribute database, and generate legal configuration parameter boundary range information through the legal change boundary range; Based on the current configuration parameters, a set of tentative adjustment increment values ​​for changing the current configuration parameters are calculated within the legal configuration parameter boundary range information, and a tentative configuration adjustment strategy including channel switching step size and transmit power adjustment magnitude is generated based on the tentative adjustment increment values. The tentative configuration adjustment strategy is encapsulated into a tentative configuration change instruction that conforms to the communication protocol type.

5. The method according to claim 4, characterized in that, The step of applying the tentative configuration change command to the virtual device in the virtual network and generating a first control strategy correlated with the communication efficiency of the virtual network includes: The network protocol agent is invoked to inject the tentative configuration change command into the virtual media access control layer and virtual physical layer of the virtual network, thereby adjusting the underlying communication state of the virtual device; After the virtual device completes the adjustment of the underlying communication state, real-time data of the underlying network state is generated by capturing the physical layer transmit power, received signal strength indication and channel bit error rate of the virtual network under the action of the exploratory configuration change command. Based on the pre-acquired real radio frequency coexistence interference curve, the average packet loss rate and round-trip delay of the real-time data of the network's underlying state under the influence of the simulated signal attenuation interference scenario are calculated, and link performance evaluation data reflecting the changes in communication link quality indicators are generated; the real radio frequency coexistence interference curve is an electromagnetic interference characteristic curve that expresses the quantitative correspondence between the degree of signal-to-noise ratio degradation and the degree of communication link impairment when heterogeneous wireless signals are concurrently emitted at the same frequency. The average packet loss rate of the virtual network is determined by comparing the link performance evaluation data. When the decrease in the average packet loss rate exceeds the preset packet loss rate threshold, the current exploratory configuration change instruction is marked as a high-value action, and communication efficiency correlation data is output. The communication efficiency-related data is structured and encapsulated to generate a first control strategy.

6. The method according to claim 4, characterized in that, The steps of reading the simulated traffic generated for a preset test scenario, and applying the tentative configuration change command to the virtual device in the virtual network when executing the simulated traffic, to generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario, include: The pre-trained scenario-linked intelligent agent is invoked to read the simulated traffic generated by the user behavior model for the test scenario, and the target concurrent business data packets corresponding to the target user behavior are obtained. The target concurrent service data packet is input into the link signal model, and the link signal model is controlled to replay the target concurrent service data packet in order to generate a virtual radio electromagnetic signal corresponding to the test scenario in the virtual network; When replaying the target concurrent service data packets, the tentative configuration change command is applied to the virtual device in the virtual network to control the transmission of the virtual radio electromagnetic signal under the analog signal attenuation interference scenario, and to generate service scenario robustness assessment data; the service scenario robustness assessment data is used to characterize the physical resilience of the virtual device in terms of anti-interference, anti-packet loss and tolerance to latency fluctuations when the virtual device performs data transmission in the virtual network. The robustness assessment data of the business scenario is used to calculate business layer automation performance data to express the success rate and stability of scenario linkage for the test scenario; A second control strategy is generated based on the automation performance data of the business layer, which is correlated with the execution efficiency of the virtual device in the test scenario.

7. The method according to claim 6, characterized in that, The tentative configuration change instruction includes physical layer transmit power, and the step of obtaining user behavior preference settings for the virtual device and generating a third control policy related to the user behavior preference settings for the virtual device includes: The device wake-up frequency of the virtual device is determined by the second control strategy, and a pre-trained user experience agent is invoked to read the user behavior preference settings through the user behavior model; the user behavior preference settings include the device standby power consumption limit and daily routine time periods; The increase in standby power consumption and the change in electromagnetic radiation of the virtual device due to the second control strategy are calculated by using the physical layer transmission power, the device wake-up frequency, the device standby power consumption limit, and the daily work and rest time period. Based on the user behavior preference settings, the user's silent mode preference during a specific period is determined, and the silent mode preference is used to generate quantitative data on life disturbances that are used to evaluate the potential disturbances caused to the user by the tentative configuration change command during the specific period. A weighted scoring operation is performed on the increase in standby power consumption, the change in electromagnetic radiation, and the quantitative data of daily life disturbance to generate user sensory comfort and low-carbon energy-saving satisfaction indicators for quantifying the potential impact of the tentative configuration change instruction and the second control strategy on daily life. Based on the user's sensory comfort and low-carbon energy-saving satisfaction indicators, a third control strategy that is correlated with the user's behavioral preference settings is output.

8. The method according to claim 7, characterized in that, The step of determining the target control strategy from the first control strategy, the second control strategy, and the third control strategy includes: The first control strategy, the second control strategy, and the third control strategy are aggregated into a decision set to be resolved. Feature cross-matching is performed on the decision set to be resolved. When it is determined that the transmit power adjustment action in the first control strategy, the redundancy backup action in the second control strategy, and the energy consumption upper limit constraint in the third control strategy are mutually exclusive, a multi-attribute decision conflict state is triggered. When the multi-attribute decision conflict state is detected to be triggered, the weighted negotiation algorithm based on multi-attribute decision is invoked and the preset global target weight is read. The benefits of the first control strategy, the second control strategy and the third control strategy are evaluated respectively, the comprehensive score for each control strategy is calculated, and the target control strategy is selected from the first control strategy, the second control strategy and the third control strategy based on the comprehensive score.

9. The method according to claim 7, characterized in that, The step of controlling the target device to perform networking operations based on the target control strategy further includes: The target control strategy is transformed into a set of physical configuration parameters that include the target channel number, node routing table, local degradation linkage logic, and maximum power limit threshold. The physical configuration parameter set is pushed into the physical register of the target device to control the target device to perform networking operations based on the physical configuration parameter set.

10. The method according to claim 9, characterized in that, Also includes: The link signal model outputs a wall attenuation coefficient to reflect the physical attenuation status, and an environmental noise floor to reflect the co-frequency interference status. Determine the target communication link in the virtual network; The received signal strength indication, average round-trip time delay, and link layer retransmission count of the target communication link are collected at preset time intervals to generate real-time link measurement data reflecting the current state of the target communication link. Read the theoretical prediction data output by the link signal model and the user behavior model at the same timestamp, and determine the absolute value of the deviation between the real-time measured data and the theoretical prediction data; When the absolute value of the deviation exceeds a preset threshold for a continuous preset period, the wall attenuation coefficient and the environmental noise floor are fitted and calculated to update the link signal model. After updating the link signal model, the network protocol agent, the scenario linkage agent, and the user experience agent are triggered to perform incremental virtual testing on the deviation area between the real-time link measured data and the theoretical prediction data, so as to output a fine-tuning configuration scheme for the target device.

11. A device networking apparatus, characterized in that, include: The virtual network construction module is used to identify the virtual devices corresponding to the target device and construct a virtual network for enabling data communication between multiple virtual devices. The tentative configuration change instruction reading module is used to obtain the protocol stack status parameters of the virtual device in the virtual network, and generate tentative configuration change instructions based on the protocol stack status parameters; The first control strategy generation module is used to apply the tentative configuration change command to the virtual device in the virtual network and generate a first control strategy that is related to the communication efficiency of the virtual network. The second control strategy generation module is used to read the simulated traffic generated for a preset test scenario, and when the simulated traffic is executed, to apply the tentative configuration change command to the virtual device in the virtual network, and generate a second control strategy that is related to the execution efficiency of the virtual device in the test scenario. The third control strategy generation module is used to obtain user behavior preference settings for the virtual device and generate a third control strategy that is related to the user behavior preference settings for the virtual device. The device networking control module is used to determine a target control strategy from the first control strategy, the second control strategy and the third control strategy, and control the target device to perform networking operations based on the target control strategy.

12. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the method as claimed in claims 1-10.

13. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the method as described in claims 1-10.

Images