Dynamic distribution method and device for network resources

Abstract

The invention discloses a dynamic distribution method and device for network resources. The method comprises the following steps: acquiring device data, user behavior data and environment data of each intelligent device in a home network; processing the equipment data, the user behavior data and the environment data through a scene recognition model to obtain a home collaborative scene type of the home network; the scene priority of the family cooperation scene type is calculated according to the family cooperation scene type, the user identity and the time sensitivity; generating a network resource allocation strategy according to the scene priority and the equipment demand information; and distributing network resources to each intelligent device according to the network resource distribution strategy. According to the method and the device, the technical problem that resource allocation and actual requirements are disjointed due to the fact that a resource allocation mode based on static QoS configuration or simple flow classification cannot adapt to the requirements of multi-device collaboration and dynamic change in a home network in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of network resource allocation technology, and more specifically, to a method and apparatus for dynamic allocation of network resources. Background Technology

[0002] In the smart home field, with the widespread adoption of IoT devices, home networks are no longer limited to traditional data transmission but now support diverse service needs, such as high-definition video streaming, health monitoring, and intelligent security systems. To meet the varying requirements of these services, intelligent allocation of network resources has become crucial. Currently, QoS allocation methods in home networks primarily rely on static configuration or simple traffic classification.

[0003] Existing technologies often lack the ability to perceive the types of home collaboration scenarios when allocating resources, such as the resource requirements of specific scenarios like home movie watching, multi-person video conferencing, or health monitoring, making it impossible to achieve accurate resource allocation.

[0004] Traditional QoS allocation methods often use fixed priorities, which cannot dynamically adjust resource allocation according to changes in network status or priorities in different scenarios. This leads to a decline in service quality and user experience when sudden network congestion or high-priority scenarios occur.

[0005] Some smart home QoS management solutions rely too heavily on cloud-based decision-making. When the local network is unstable or cloud services are delayed, it may lead to reduced device collaboration efficiency.

[0006] For example, existing technologies have proposed a QoS intelligent prediction and adjustment method based on traffic characteristics. This method collects traffic data in real time, constructs a multi-dimensional feature vector, combines an LSTM (Long Short-Term Memory) network for short-term prediction and a periodic trend analysis module for long-term prediction, and then optimizes resource allocation strategies through reinforcement learning and NPU (Network Processing Unit) hardware collaboration. While this method demonstrates some innovation in traffic prediction and resource allocation, it still has shortcomings in specific scenario perception and device collaboration within home networks.

[0007] For example, existing technologies have proposed a multi-terminal dynamic bandwidth allocation control method based on real-time traffic monitoring. This method dynamically adjusts bandwidth allocation priorities by continuously monitoring network congestion and terminal service demands. It emphasizes the classification of terminal service importance and user identity settings, which can optimize bandwidth utilization to a certain extent and ensure the transmission quality of critical services. However, this technology still has room for improvement in handling inter-device interaction and adaptive resource allocation in home collaboration scenarios.

[0008] Therefore, the current resource allocation method cannot dynamically adjust QoS allocation according to the type of home collaboration scenario: when handling multi-device collaboration, the current solution struggles to accurately identify the collaboration scenario, leading to a disconnect between resource allocation and actual needs, thus impacting user experience. Furthermore, there is insufficient quality of service (QoS) assurance for critical devices during network congestion: when home networks face sudden congestion, existing technologies struggle to prioritize the stability and QoS of high-priority services such as health monitoring and video calls.

[0009] There is currently no effective solution to the above problems. Summary of the Invention

[0010] This invention provides a method and apparatus for dynamic allocation of network resources, which at least solves the technical problem that the resource allocation methods based on static QoS configuration or simple traffic classification in related technologies cannot adapt to the needs of multi-device collaboration and dynamic changes in home networks, resulting in a disconnect between resource allocation and actual needs.

[0011] According to one aspect of the present invention, a method for dynamically allocating network resources is provided, comprising: acquiring device data, user behavior data, and environmental data of each smart device in a home network; processing the device data, user behavior data, and environmental data through a scene recognition model to obtain a home collaboration scene type of the home network, wherein the scene recognition model is a model trained by machine learning using multiple sets of training data, each set of training data including: sample device data, sample user behavior data, sample environmental data, and sample home collaboration scene type corresponding to the sample device data, the user behavior data, and the sample environmental data; calculating a scene priority of the home collaboration scene type based on the home collaboration scene type, user identity, and time sensitivity; generating a network resource allocation strategy based on the scene priority and device demand information; and allocating network resources to each smart device according to the network resource allocation strategy.

[0012] Optionally, acquiring device data, user behavior data, and environmental data of each smart device in the home network includes: collecting bandwidth demand data, latency sensitivity data, and network status data from each smart device through a network resource sensing component in the home network to obtain the device data; analyzing user interaction logs to obtain the user behavior data; and collecting the environmental data through an environmental sensing component in the home network.

[0013] Optionally, the device data, user behavior data, and environmental data are processed by a scene recognition model to obtain the home collaboration scene type of the home network. This includes: preprocessing the device data, user behavior data, and environmental data to obtain preprocessed device data, user behavior data, and environmental data; and inputting the preprocessed device data, user behavior data, and environmental data into the scene recognition model to process the preprocessed device data, user behavior data, and environmental data using the scene recognition model to obtain the home collaboration scene type.

[0014] Optionally, the device data, user behavior data, and environmental data are processed by a scene recognition model to obtain the home collaboration scene type of the home network, including: converting the user behavior data into time series feature data; fusing the environmental data and the device data to generate a scene feature vector; and classifying the scene feature vector according to the time series feature data to obtain the home collaboration scene type.

[0015] Optionally, generating a network resource allocation strategy based on the scenario priority and device requirement information includes: extracting bandwidth requirement data and latency sensitivity data of each smart device from a scenario device mapping table to obtain the device requirement information, wherein the scenario device mapping table is pre-generated to record the bandwidth required by each smart device in different collaborative scenarios; calculating the device priority of each smart device based on the device type of each smart device and the scenario priority; and calculating the bandwidth allocation ratio of each smart device in the home collaborative scenario type based on the device priority to obtain the network resource allocation strategy.

[0016] Optionally, after allocating network resources to each of the smart devices according to the network resource allocation strategy, the dynamic allocation method further includes: monitoring the home network to obtain monitoring results of the home network; when the monitoring results indicate that the home network is in a congested state, adjusting the resource allocation strategy to obtain an adjusted resource allocation strategy; and allocating network resources to each of the smart devices according to the adjusted resource allocation strategy.

[0017] Optionally, the home network is monitored to obtain monitoring results, including monitoring the bandwidth utilization, packet loss rate, and latency jitter of the home network to obtain the monitoring results.

[0018] Optionally, the resource allocation strategy is adjusted to obtain an adjusted resource allocation strategy, including: calculating the congestion index of the home network based on the bandwidth utilization, the packet loss rate, and the latency jitter; and adjusting the resource allocation strategy based on the congestion index and the scenario priority to obtain the adjusted resource allocation strategy.

[0019] Optionally, the dynamic allocation method further includes: when the home network is detected to be in a weak network or out of network state, obtaining the resource allocation strategy from the ontology cache of the home network.

[0020] Optionally, after allocating network resources to each of the smart devices according to the network resource allocation strategy, the dynamic allocation method further includes: collecting feedback data from each of the smart devices, wherein the feedback data includes: actual bandwidth usage data and latency index data; and optimizing the scene recognition model and the resource allocation strategy based on the feedback data.

[0021] According to another aspect of the present invention, a dynamic allocation device for network resources is also provided, comprising: an acquisition unit, configured to acquire device data, user behavior data, and environmental data of each smart device in a home network; a processing unit, configured to process the device data, user behavior data, and environmental data through a scene recognition model to obtain a home collaboration scene type of the home network, wherein the scene recognition model is a model trained by machine learning using multiple sets of training data, each set of training data including: sample device data, sample user behavior data, sample environmental data, and a sample home collaboration scene type corresponding to the sample device data, the user behavior data, and the sample environmental data; a calculation unit, configured to calculate a scene priority of the home collaboration scene type based on the home collaboration scene type, user identity, and time sensitivity; a generation unit, configured to generate a network resource allocation strategy based on the scene priority and device requirement information; and an allocation unit, configured to allocate network resources to each smart device according to the network resource allocation strategy.

[0022] Optionally, the acquisition unit includes: a first acquisition module, used to acquire bandwidth demand data, latency sensitivity data, and network status data from each of the smart devices through a network resource sensing component in the home network to obtain the device data; an analysis module, used to analyze user interaction logs to obtain the user behavior data; and a second acquisition module, used to acquire the environmental data through an environmental sensing component in the home network.

[0023] Optionally, the processing unit includes: a preprocessing module for preprocessing the device data, user behavior data, and environmental data to obtain preprocessed device data, user behavior data, and environmental data; and a processing module for inputting the preprocessed device data, user behavior data, and environmental data into the scene recognition model to process the preprocessed device data, user behavior data, and environmental data using the scene recognition model to obtain the home collaboration scene type.

[0024] Optionally, the processing unit includes: a conversion module for converting the user behavior data into time-series feature data; a fusion module for fusing the environmental data and the device data to generate a scene feature vector; and a classification module for classifying the scene feature vector according to the time-series feature data to obtain the home collaboration scene type.

[0025] Optionally, the generation unit includes: an extraction module, configured to extract bandwidth requirement data and latency sensitivity data of each smart device from a scene device mapping table to obtain the device requirement information, wherein the scene device mapping table is pre-generated to record the bandwidth required by each smart device in different collaborative scenarios; a first calculation module, configured to calculate the device priority of each smart device according to the device type of each smart device and the scene priority; and a second calculation module, configured to calculate the bandwidth allocation ratio of each smart device in the home collaborative scenario type according to the device priority, so as to obtain the network resource allocation strategy.

[0026] Optionally, the dynamic allocation device further includes: a monitoring unit, configured to monitor the home network after allocating network resources to each of the smart devices according to the network resource allocation strategy, to obtain a monitoring result of the home network; an adjustment unit, configured to adjust the resource allocation strategy when the monitoring result indicates that the home network is in a congested state, to obtain an adjusted resource allocation strategy; and an allocation unit, configured to allocate network resources to each of the smart devices according to the adjusted resource allocation strategy.

[0027] Optionally, the monitoring unit includes a monitoring module for monitoring the bandwidth utilization, packet loss rate, and latency jitter of the home network to obtain the monitoring results.

[0028] Optionally, the adjustment unit includes: a third calculation module, used to calculate the congestion index of the home network based on the bandwidth utilization, the packet loss rate, and the latency jitter; and an adjustment module, used to adjust the resource allocation strategy based on the congestion index and the scenario priority to obtain the adjusted resource allocation strategy.

[0029] Optionally, the dynamic allocation device further includes: the acquisition unit, which is further configured to acquire the resource allocation strategy from the ontology cache of the home network when the home network is detected to be in a weak network or offline state.

[0030] Optionally, the dynamic allocation device further includes: a collection unit, configured to collect feedback data from each of the intelligent devices after allocating network resources to each of the intelligent devices according to the network resource allocation strategy, wherein the feedback data includes: actual bandwidth usage data and latency index data; and an optimization unit, configured to optimize the scene recognition model and the resource allocation strategy based on the feedback data.

[0031] According to another aspect of the present invention, a smart home system is also provided, which uses the dynamic allocation method of network resources described in any one of the above embodiments.

[0032] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes the dynamic allocation method for network resources described in any one of the above embodiments.

[0033] According to another aspect of the present invention, a processor is also provided, the processor being configured to run a program, wherein the program, when running, executes the dynamic allocation method for network resources described in any one of the preceding embodiments.

[0034] According to another aspect of the present invention, a computer program product is also provided, including computer instructions that, when executed by a processor, perform the dynamic allocation method for network resources described in any one of the above embodiments.

[0035] In this embodiment of the invention, device data, user behavior data, and environmental data of each smart device in the home network are acquired; the device data, user behavior data, and environmental data are processed by a scene recognition model to obtain the home collaboration scene type of the home network. The scene recognition model is a model trained using multiple sets of training data through machine learning. Each set of training data includes: sample device data, sample user behavior data, sample environmental data, and a sample home collaboration scene type corresponding to the sample device data, user behavior data, and sample environmental data; the scene priority of the home collaboration scene type is calculated based on the home collaboration scene type, user identity, and time sensitivity; a network resource allocation strategy is generated based on the scene priority and device requirement information; and network resources are allocated to each smart device according to the network resource allocation strategy. The technical solution provided by this invention achieves the goal of dynamically allocating network resources based on scene priority and device demand, by capturing the operating status of all smart devices in real time, analyzing user habits and environmental indicators, identifying scenes based on the captured information, and calculating the priority of the identified scene type in combination with user identity and time sensitivity. This improves the sensitivity to changes in the home network environment. Moreover, through real-time monitoring and scene identification, the system achieves accurate resource allocation, avoids unnecessary bandwidth waste, and improves the overall utilization efficiency of home network resources. This solves the technical problem in related technologies where resource allocation based on static QoS configuration or simple traffic classification cannot adapt to the needs of multi-device collaboration and dynamic changes in home networks, leading to a disconnect between resource allocation and actual needs. Attached Figure Description

[0036] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0037] Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of dynamically allocating network resources according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart of a method for dynamically allocating network resources according to an embodiment of the present invention;

[0039] Figure 3 This is a flowchart of an optional dynamic allocation method for network resources according to an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of a dynamic allocation device for network resources according to an embodiment of the present invention.

[0041] The above figures include the following reference numerals:

[0042] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0043] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] As described in the background section, the resource allocation methods based on static QoS configuration or simple traffic classification in related technologies cannot adapt to the needs of multi-device collaboration and dynamic changes in home networks, resulting in a disconnect between resource allocation and actual needs. Specifically, static QoS configuration is the most basic QoS management method. By pre-setting fixed bandwidth and priorities for specific devices or applications, it ensures that these devices or applications can obtain the necessary network resources under normal network conditions. However, this method has a significant drawback: when the network environment changes or device usage scenarios switch, the preset QoS parameters may no longer be applicable, leading to uneven resource allocation or critical applications being affected by network congestion. Simple traffic classification: In home networks, traffic is classified based on device type or application protocol, and different QoS policies are applied to different categories. For example, video streams are classified as high-priority traffic, and file transfers are classified as low-priority traffic. While this method can reasonably allocate resources for some common applications, it ignores the complex needs of inter-device collaboration and the degree of impact of network congestion on different applications. Therefore, embodiments of the present invention provide a method and apparatus for dynamic allocation of network resources, a smart home system, a computer-readable storage medium, a processor, and a computer program product.

[0046] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0047] The methods and embodiments provided in this invention can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a dynamic allocation method of network resources according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0048] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the dynamic allocation method of network resources in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-described networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the above-described networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0049] Example 1

[0050] According to an embodiment of the present invention, a method embodiment of a dynamic allocation method for network resources is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0051] Figure 2 This is a flowchart of a dynamic allocation method for network resources according to an embodiment of the present invention, such as... Figure 2 As shown, the method includes the following steps:

[0052] Step S202: Obtain device data, user behavior data, and environmental data of each smart device in the home network.

[0053] Optionally, the above-mentioned device data may include, but is not limited to: bandwidth requirements, latency sensitivity, current network status (such as bandwidth utilization, packet loss rate), etc.

[0054] Optionally, the aforementioned user behavior data may include, but is not limited to, data such as device usage frequency and device usage time.

[0055] Optionally, the above environmental data may include, but is not limited to, data such as temperature, humidity, and light.

[0056] In this embodiment, the system uses smart components in the home network, such as sensors and data analysis modules, to capture the operating status of all smart devices in real time (including bandwidth consumption, latency tolerance, network connection status, etc.), analyze user habits (such as device usage frequency and time), and environmental indicators (such as light, temperature and humidity).

[0057] This phase utilizes IoT technology and data science to collect multi-dimensional information, laying a data foundation for subsequent scene identification and resource allocation.

[0058] The comprehensive and real-time data collection ensured a keen awareness of changes in the home network environment, providing an accurate basis for resource allocation.

[0059] It should be noted that this data collection is conducted through network protocols to ensure the security and efficiency of data transmission.

[0060] Step S204: Process device data, user behavior data, and environmental data through a scene recognition model to obtain the home collaboration scene type of the home network. The scene recognition model is a model trained by machine learning using multiple sets of training data. Each set of training data includes: sample device data, sample user behavior data, sample environmental data, and sample home collaboration scene type corresponding to the sample device data, user behavior data, and sample environmental data.

[0061] In this embodiment, the collected data is fed into a scene recognition model, which is trained on a large amount of historical data and is able to identify specific collaborative scenarios (e.g., "family movie watching" or "nighttime sleep mode") that are currently taking place in the home network. The model uses machine learning techniques, particularly advanced algorithms for contextual understanding.

[0062] Here, the scene recognition principle is based on the pattern recognition capability of machine learning. By comparing the input data with the training dataset inside the model, the most matching home collaboration scene type is identified.

[0063] High-precision scene recognition enables more targeted resource allocation, improving the overall performance and user experience of home networks in specific scenarios.

[0064] Step S206: Calculate the scenario priority of the family collaboration scenario type based on the family collaboration scenario type, user identity, and time sensitivity.

[0065] In this embodiment, once a family collaboration scenario is identified, the system dynamically calculates the scenario priority based on the scenario type, user identity (such as elderly or child), and the urgency of the current time.

[0066] When calculating priorities, the system references a pre-defined weight matrix, where different user identities and scenario types have corresponding weight values. These weights reflect the importance of the scenario under various circumstances.

[0067] By dynamically calculating scenario priorities, it is ensured that, given limited resources, more important scenarios receive priority in terms of service quality.

[0068] Step S208: Generate a network resource allocation strategy based on scenario priority and device requirement information.

[0069] In this embodiment, based on calculated scenario priorities and predefined device requirements (such as minimum bandwidth and latency requirements), the system generates and implements a dynamic network resource allocation strategy. This strategy ensures that bandwidth and priority are rationally allocated to each device to maximize user experience and network efficiency.

[0070] The above resource allocation strategy employs advanced network management algorithms, such as Weighted Fair Queuing (WFQ), to ensure that bandwidth allocation among devices is both fair and effective in a dynamic network environment.

[0071] Through intelligent resource allocation, the high performance of critical equipment and services can be maintained even during peak hours or network congestion, while balancing the needs of other equipment.

[0072] Step S210: Allocate network resources to each smart device according to the network resource allocation strategy.

[0073] In this embodiment, according to the generated network resource allocation strategy, the home network smart controller adjusts the network parameters of each smart device in real time, such as bandwidth limits and data packet priorities, to ensure that each device can obtain the network resources most suitable for its needs.

[0074] The above dynamic allocation principle takes into account real-time monitoring and prediction of network resources, and allows for immediate adjustment of resource allocation based on network conditions to cope with sudden congestion or changes in equipment demand.

[0075] Dynamic resource allocation can effectively prevent and alleviate network bottlenecks, ensuring the stable operation and efficient use of home networks under any circumstances.

[0076] Figure 3 This is a flowchart of a dynamic allocation method for network resources according to an embodiment of the present invention, such as... Figure 3As shown in this embodiment of the invention, in scene-driven bandwidth allocation, the system dynamically adjusts the bandwidth allocation ratio based on the priority of the current family collaboration scene (such as "health monitoring" or "family movie watching"). For example, in the "health monitoring" scene, the system prioritizes allocating bandwidth to health devices, while in the "family movie watching" scene, it allocates more resources to video devices. For scene priority determination, the system comprehensively judges scene priority based on user identity (such as elderly / child), time sensitivity (such as real-time needs), and scene type (such as security / entertainment). For example, if the user is an elderly person and the scene is "health monitoring," the scene priority weight is significantly increased.

[0077] As described above, in this embodiment of the invention, device data, user behavior data, and environmental data of each smart device in the home network are acquired; the device data, user behavior data, and environmental data are processed by a scene recognition model to obtain the home collaboration scene type of the home network. The scene recognition model is a model trained using multiple sets of training data through machine learning. Each set of training data includes: sample device data, sample user behavior data, sample environmental data, and a sample home collaboration scene type corresponding to the sample device data, user behavior data, and sample environmental data; the scene priority of the home collaboration scene type is calculated based on the home collaboration scene type, user identity, and time sensitivity; and network resource allocation is generated based on the scene priority and device requirement information. The strategy involves allocating network resources to each smart device according to a network resource allocation policy. This achieves the goal of dynamically allocating network resources based on real-time capture of the operating status of all smart devices (including bandwidth consumption, latency tolerance, network connection status, etc.), analysis of user habits (such as device usage frequency and time), and environmental indicators. Based on the captured information, the system identifies scenarios and calculates their priority according to user identity and time sensitivity. This improves the sensitivity to changes in the home network environment. Furthermore, through real-time monitoring and scenario identification, the system achieves precise resource allocation, avoids unnecessary bandwidth waste, and improves the overall utilization efficiency of home network resources.

[0078] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem that the resource allocation method based on static QoS configuration or simple traffic classification in the related technology cannot adapt to the needs of multi-device collaboration and dynamic changes in home networks, resulting in a disconnect between resource allocation and actual needs.

[0079] It should be noted that the smart home system provided in this embodiment of the invention may include: a home network perception layer, a home collaborative scene recognition engine, a dynamic QoS allocation algorithm module, a network state adaptive mechanism module, and a resource allocation execution and feedback module. The following description is based on specific embodiments.

[0080] According to the above embodiments of the present invention, acquiring device data, user behavior data, and environmental data of each smart device in a home network includes: collecting bandwidth demand data, latency sensitivity data, and network status data from each smart device through a network resource sensing component in the home network to obtain device data; analyzing user interaction logs to obtain user behavior data; and collecting environmental data through an environmental sensing component in the home network.

[0081] In this embodiment, by integrating various sensors and network devices within the home, the system continuously collects bandwidth requirements, latency sensitivity, and network status data for each smart device. Simultaneously, it analyzes user behavior using user interaction logs and monitors changes in the indoor environment using environmental sensors, forming a comprehensive dataset of devices, behaviors, and the environment.

[0082] The aforementioned home network perception layer may include: a device status monitoring module: collecting real-time data on bandwidth requirements, latency sensitivity, and current network status (such as bandwidth utilization and packet loss rate) for each device in the home network; a scene feature extraction module: extracting scene features from multiple dimensions such as device status, user behavior (such as device usage frequency and time), and environmental data (such as light intensity and temperature); and a network quality assessment module: assessing the current network quality based on real-time network monitoring data (such as latency, jitter, and bandwidth) and generating a network status report.

[0083] The aforementioned data collection forms the foundation for intelligent scene recognition. By analyzing the data using technologies such as deep learning, subtle connections between device usage patterns and home collaboration scenarios can be captured. The fusion of time-series characteristics of user behavior with environmental data makes scene recognition more accurate.

[0084] Detailed data collection improves the accuracy of scene recognition, ensuring that resource allocation is closer to the actual needs of users, thereby improving device efficiency and user experience.

[0085] According to the above embodiments of the present invention, the home network collaborative scene type is obtained by processing device data, user behavior data, and environmental data through a scene recognition model. This includes: preprocessing device data, user behavior data, and environmental data to obtain preprocessed device data, user behavior data, and environmental data; and inputting the preprocessed device data, user behavior data, and environmental data into the scene recognition model to process the preprocessed device data, user behavior data, and environmental data using the scene recognition model to obtain the home collaborative scene type.

[0086] In this embodiment, the system preprocesses the raw data before inputting it into the scene recognition model, including feature selection, data cleaning, and format conversion, to ensure that the model can process the data efficiently. The preprocessed data is then input into the scene recognition model trained on historical data.

[0087] Furthermore, in this embodiment, scene recognition can be performed using a home collaboration scene recognition engine. Specifically, common home collaboration scenes and their characteristics can be predefined through a scene feature library (e.g., "Home Movie Viewing": TV, audio, and lighting coordination; "Health Monitoring": health device data upload). The scene recognition model uses machine learning algorithms (e.g., random forest, neural networks) to classify the input device status, user behavior, and other features, outputting the current collaboration scene type. A scene priority evaluation module can also dynamically calculate scene priority weights based on scene type, user identity (e.g., elderly / children), and time sensitivity (e.g., real-time needs).

[0088] The preprocessing steps described above are crucial for improving the model's recognition accuracy. By standardizing the data and extracting features, noise interference is eliminated, and the expressive power of the data is enhanced. The scene recognition model, based on machine learning, can identify the current type of home collaboration scene from the input feature vector.

[0089] By combining preprocessing with model application, the efficiency and accuracy of scene recognition are improved, laying a solid foundation for subsequent dynamic resource allocation.

[0090] According to the above embodiments of the present invention, the home network collaborative scene type is obtained by processing device data, user behavior data and environmental data through a scene recognition model, including: converting user behavior data into time series feature data; fusing environmental data and device data to generate scene feature vectors; and classifying the scene feature vectors according to the time series feature data to obtain the home collaborative scene type.

[0091] In this embodiment, the system transforms user behavior data into time-series features, which are then fused with environmental and device data to form a feature vector representing a home collaboration scenario. In this way, the model can capture long-term trends and periodic patterns, further improving the accuracy of scene recognition.

[0092] The aforementioned time series features reflect the patterns of device usage over time, while scene vector fusion integrates multi-dimensional data so that the model can fully understand the subtle differences in the home environment.

[0093] This fusion approach enables the system to finely differentiate between different scenarios, accurately identifying different family collaboration patterns even under similar environmental conditions, thus ensuring the refinement of resource allocation strategies.

[0094] According to the above embodiments of the present invention, a network resource allocation strategy is generated based on scenario priority and device requirement information, including: extracting bandwidth requirement data and latency sensitivity data of each smart device from a scenario device mapping table to obtain device requirement information, wherein the scenario device mapping table is pre-generated to record the bandwidth required by each smart device in different collaborative scenarios; calculating the device priority of each smart device according to the device type and scenario priority of each smart device; and calculating the bandwidth allocation ratio of each smart device in the home collaborative scenario type according to the device priority to obtain the network resource allocation strategy.

[0095] In this embodiment, the system extracts information from the scene device mapping table based on scene priority and device requirements (such as bandwidth and latency requirements), dynamically calculates device priority and bandwidth allocation ratio, and formulates network resource allocation strategies.

[0096] Here, the dynamic QoS allocation algorithm module can utilize the scenario device mapping table to define participating devices and their QoS requirements (such as bandwidth and latency) for each collaborative scenario. The resource allocation strategy engine generates bandwidth and priority allocation schemes based on scenario priority, device requirements, and network status. For example, the Weighted Fair Queuing (WFQ) algorithm can be used to dynamically allocate bandwidth according to device priority, ensuring that high-priority scenarios and devices receive resources.

[0097] Furthermore, in this embodiment of the invention, the device priority calculation involves a multi-factor comprehensive evaluation to dynamically adjust device priorities, ensuring that critical devices receive higher weight in resource allocation. The system can perform multi-dimensional priority evaluation, dynamically calculating device priorities by comprehensively considering device type (e.g., security / non-security), latency sensitivity (e.g., real-time requirements), user identity (e.g., elderly / children), and current scenario priority. Device type weight: High-priority devices (e.g., health monitoring devices, smart door locks): Due to their involvement in user safety or core functions, their priority weight is significantly higher than that of ordinary devices.

[0098] Low-priority devices (such as smart bulbs and temperature / humidity sensors): Their priority can be temporarily lowered when resources are scarce. Latency sensitivity adjustment: Devices sensitive to latency (such as health monitoring devices) will have their priority dynamically increased to ensure real-time data transmission. For example, when the latency sensitivity of a health device is adjusted from "medium" to "high," its priority weight increases by 20%.

[0099] The above resource allocation strategy is generated based on device priority calculation. By comprehensively considering device type, scenario priority and dynamic network status, it ensures that resource allocation is both fair and efficient.

[0100] Dynamic QoS policies can be adjusted in real time, ensuring that critical devices can still obtain the necessary network resources when the scenario changes, thereby improving network response speed and resource utilization.

[0101] According to the above embodiments of the present invention, after allocating network resources to each smart device according to the network resource allocation strategy, the dynamic allocation method further includes: monitoring the home network to obtain the monitoring results of the home network; when the monitoring results indicate that the home network is in a congested state, adjusting the resource allocation strategy to obtain the adjusted resource allocation strategy; and allocating network resources to each smart device according to the adjusted resource allocation strategy.

[0102] In this embodiment, in addition to conventional resource allocation, the system also monitors the home network status, such as bandwidth utilization, packet loss rate, and latency jitter. When the monitoring results indicate network congestion, the system dynamically adjusts the resource allocation strategy.

[0103] Here, the congestion detection module within the network state adaptive mechanism can be utilized to monitor network congestion status in real time (such as bandwidth utilization and packet loss rate) and generate a congestion index. The dynamic adjustment module adjusts device resource allocation based on the congestion index and scenario priority, prioritizing high-priority scenarios.

[0104] like Figure 3As shown, in this embodiment of the invention, device requirements are adapted to network conditions in real time: the system not only considers the bandwidth requirements of devices (e.g., video devices require high bandwidth), but also dynamically adjusts the allocation strategy based on network conditions (e.g., current bandwidth utilization, packet loss rate). Furthermore, this embodiment of the invention also establishes dynamic bandwidth adjustment rules: Normal state: Bandwidth is allocated according to scenario priority and device requirements to ensure that critical devices receive sufficient resources. Congested state: If network bandwidth utilization exceeds 80%, the system will prioritize bandwidth requirements for high-priority scenarios while reducing the bandwidth allocation ratio for low-priority scenarios. For example, when "health monitoring" and "home movie watching" occur simultaneously and the network is congested, the bandwidth allocation ratio for health monitoring devices will increase from 30% to 50%, while the bandwidth allocation ratio for video devices will decrease from 70% to 50%. Local coordination priority strategy: In weak network or network outage environments, the system will prioritize coordinating device resources within the local network to reduce reliance on the cloud and ensure the bandwidth requirements of critical devices (e.g., health monitoring devices).

[0105] In addition, this embodiment of the invention also includes a hierarchical bandwidth allocation mechanism, mainly comprising: basic bandwidth allocation: allocating a minimum guaranteed bandwidth to each device to ensure its basic functions operate normally (e.g., the minimum bandwidth requirement for a healthy device is 10Mbps). Dynamic bandwidth expansion: dynamically expanding the bandwidth based on scenario priority and device requirements. For example, a video device can be temporarily expanded to 50Mbps in a "home movie viewing" scenario, while maintaining the basic bandwidth in other scenarios.

[0106] It should be noted that in this embodiment of the invention, in the scene-driven dynamic priority adjustment, the system dynamically adjusts the device priority according to the priority of the current scene. For example, in the "home movie viewing" scene, the priority of the video device will be temporarily increased, while the priority of the health monitoring device will be adjusted appropriately according to the needs of the scene. For scene priority weight superposition: if a device participates in multiple scenes (such as a health monitoring device participating in both "health monitoring" and "home movie viewing"), the system will calculate the device priority based on the superposition of scene priorities. For example, the priority weight of the "health monitoring" scene is 1.2, while that of "home movie viewing" is 0.8, and the total priority weight of the device is 1.2 + 0.8 = 2.0.

[0107] In addition, in user behavior and device status feedback, the system dynamically adjusts priorities based on user behavior (such as device usage frequency and time) and device status (such as whether it is online or active). For example, if a user frequently uses a video device, the system will temporarily increase its priority to ensure a smooth experience.

[0108] Furthermore, this embodiment of the invention also includes a priority dynamic recycling mechanism. That is, when resources are scarce, the system will prioritize recycling bandwidth resources from low-priority devices. For example, if the network is congested and the health monitoring device has already obtained sufficient bandwidth, the system will gradually reduce the bandwidth allocation for non-critical devices (such as smart switches) to ensure the stability of high-priority scenarios.

[0109] Here, the congestion detection mechanism, combined with scenario priority, can reallocate bandwidth when network resources are scarce, ensuring the continuity and stability of high-priority scenarios.

[0110] The network status monitoring and policy adjustment mechanism enables the system to effectively cope with sudden network congestion, avoid critical service interruptions, and improve the overall network service quality.

[0111] According to the above embodiments of the present invention, monitoring a home network to obtain monitoring results includes monitoring the bandwidth utilization, packet loss rate, and latency jitter of the home network to obtain monitoring results.

[0112] In this embodiment, the system monitors key indicators of the home network in real time, including bandwidth utilization, packet loss rate, and latency jitter, in order to quantify the degree of network congestion.

[0113] Here, the bandwidth allocation executor in the resource allocation execution and feedback module can be used to send the calculation results to the target device and adjust its bandwidth and priority.

[0114] By calculating the average, standard deviation, or threshold trigger of these indicators, the system can accurately determine whether the network is in a congested state, providing a quantitative basis for resource allocation adjustments during congestion.

[0115] The aforementioned quantitative monitoring of network congestion enables the system to respond promptly to network fluctuations and take effective resource adjustment measures, preventing service degradation caused by network congestion.

[0116] According to the above embodiments of the present invention, the resource allocation strategy is adjusted to obtain an adjusted resource allocation strategy, including: calculating the congestion index of the home network based on bandwidth utilization, packet loss rate, and latency jitter; and adjusting the resource allocation strategy based on the congestion index and scenario priority to obtain the adjusted resource allocation strategy.

[0117] In this embodiment, when network congestion is detected, the system automatically adjusts the resource allocation strategy and recalculates the bandwidth allocation ratio of the devices based on the real-time congestion index and scenario priority.

[0118] Here, the feedback optimization module can be used to collect device feedback data (such as actual bandwidth usage and latency metrics) and continuously optimize the scene recognition model and resource allocation strategy.

[0119] The aforementioned dynamic adjustment mechanism is based on network congestion levels and scenario priorities. It adjusts QoS policies through real-time feedback to achieve dynamic reallocation of resources.

[0120] Dynamic adjustments to resource allocation strategies ensure that high-priority scenarios are prioritized for network performance even under congestion, thereby enhancing user satisfaction and trust.

[0121] According to the above embodiments of the present invention, the dynamic allocation method further includes: when the home network is detected to be in a weak network or offline state, obtaining the resource allocation strategy from the home network's ontology cache.

[0122] In this embodiment, when the home network is in a weak or offline state, the system reads a preset resource allocation strategy from the local cache to maintain the operation of critical devices.

[0123] That is, in the embodiments of the present invention, a local coordination priority strategy is also included, which coordinates device resources through local cache state when the network is weak or disconnected, thereby reducing the dependence on the cloud.

[0124] The aforementioned local caching strategy reduces reliance on external networks and ensures device collaboration within the home network even during internet outages by pre-storing resource allocation strategies.

[0125] The local caching strategy ensures that the home network remains functional and available even in extreme situations, enhancing the system's robustness and the continuity of the user experience.

[0126] According to the above embodiments of the present invention, after allocating network resources to each smart device according to the network resource allocation strategy, the dynamic allocation method further includes: collecting feedback data from each smart device, wherein the feedback data includes: actual bandwidth usage data and latency index data; and optimizing the scene recognition model and resource allocation strategy based on the feedback data.

[0127] In this embodiment, after performing resource allocation, the system continuously collects device feedback data, including actual bandwidth usage and latency metrics, and then optimizes the scene recognition model and resource allocation strategy based on this feedback data.

[0128] The aforementioned feedback optimization mechanism is based on the principle of continuous learning. By collecting actual performance data, the system can self-correct and improve the accuracy of scene recognition and resource allocation.

[0129] This closed-loop feedback optimization enables the system to continuously evolve, adapting to changes in the home network environment and fluctuations in device demands, thereby maintaining the effectiveness of resource allocation strategies in the long term and improving the level of intelligent management of the home network.

[0130] like Figure 3As shown, during the data acquisition phase, the home network sensing layer uses sensors and protocol analysis modules to collect real-time data on bandwidth requirements, latency sensitivity, and network status (such as bandwidth utilization and packet loss rate) of various devices in the home network. Simultaneously, it extracts scene features through user behavior records (such as device usage frequency and time) and environmental data (such as light intensity and temperature). Data sources include: Device status: Bandwidth requirements and latency sensitivity are collected from smart TVs, speakers, health monitoring devices, etc. User behavior: User intent is analyzed through user interaction logs (such as device startup time and usage frequency). Environmental data: Environmental information is obtained through temperature and humidity sensors, light sensors, etc. Data transmission: All data is transmitted through the home network to the scene recognition engine, where it is preprocessed by the scene feature extraction module (such as standardization and feature normalization).

[0131] In the scene recognition phase, the scene recognition model inputs collected device status, user behavior, and environmental data, and uses machine learning algorithms (such as random forests) to determine the current home collaboration scene. For example, after the system recognizes the "home movie viewing" scene, it calls the scene-device mapping table to determine the participating devices (such as TV, audio equipment, and lighting) and their QoS requirements. For feature processing: user behavior data (such as device startup time and usage frequency) is transformed into time-series features. Environmental data (such as light intensity) is combined with device status data (such as TV resolution) to generate scene feature vectors. For model inference: a pre-trained scene recognition model (such as a deep learning-based classifier) ​​is used to classify the feature vectors and output the scene type (such as "home movie viewing" or "health monitoring").

[0132] Furthermore, for scene priority calculation: Scene priority is dynamically calculated based on scene type, user identity (e.g., elderly / child), and time sensitivity (e.g., real-time requirements). For example, if the user is elderly and the scene is "health monitoring," the scene priority weight is 1.2. For the resource allocation phase: The dynamic QoS allocation algorithm module generates a bandwidth and priority allocation scheme based on scene priority and device requirements. For example, in the "health monitoring" scene, health devices are assigned the highest priority, while video devices have a lower priority. The system allocates bandwidth proportionally to each device using the WFQ algorithm. 1. Device requirement analysis: Extract the device's bandwidth requirements (e.g., 10Mbps for health devices, 50Mbps for video devices) and latency sensitivity (e.g., 10ms for health devices, 50ms for video devices) from the scene-device mapping table. Priority calculation: Calculate device priority based on device type (e.g., smart lock = 1.5, smart bulb = 0.5) and scene priority weight. For example, health devices have a priority of 10, and video devices have a priority of 5. For bandwidth allocation: Calculate the bandwidth allocation ratio for each device using the WFQ algorithm. For example, if the priority of health devices is 10, video devices are 5, and the total priority is 15, then the allocation for health devices is 66.7, and the allocation for video devices is 33.3. Network state adaptive adjustment phase: The network state adaptive mechanism module monitors network congestion in real time. If congestion is detected, the system dynamically adjusts resource allocation. For example, when "home movie watching" and "health monitoring" occur simultaneously and network congestion occurs, the bandwidth allocation ratio for health monitoring devices increases from 30% to 50%, while the bandwidth for video devices decreases from 70% to 50%.

[0133] For congestion detection: Real-time monitoring of network bandwidth utilization (e.g., whether it exceeds 90%), packet loss rate (e.g., whether it exceeds 5%), and latency jitter (e.g., whether it exceeds 20ms). 2. Dynamic adjustment: Dynamically adjust device resource allocation based on the congestion index (e.g., 0.7) and scenario priority. For example, the priority of healthy devices is adjusted from 10 to 6.5, and the priority of video devices is adjusted from 5 to 3.5. Local collaborative priority strategy: In the event of weak network or network outage, device resources are coordinated through local cached states. For example, bandwidth allocation for healthy devices is directly executed by local policies without waiting for cloud instructions.

[0134] Furthermore, during the resource allocation execution and feedback phase: the bandwidth allocation executor sends the calculation results to the target device, adjusting its bandwidth and priority. For example, a healthy device receives a fixed bandwidth (e.g., 10Mbps), while video devices dynamically adjust their bandwidth based on real-time demands. The feedback optimization module collects device feedback data (such as actual bandwidth usage and latency metrics) to continuously optimize the scene recognition model and resource allocation strategy. Specifically, the bandwidth allocation results can be sent to the target device via network protocols (such as MQTT and HTTP) to adjust its network parameters (such as bandwidth limits and priority flags). Actual bandwidth usage and latency data are recorded in device logs and fed back to the feedback optimization module.

[0135] Finally, the optimization strategy can update the training data of the scene recognition model based on feedback data to improve the accuracy of scene classification. The weight calculation logic of the WFQ algorithm can also be optimized to adapt to new network environments and user needs.

[0136] As described above, the technical solution provided by this invention dynamically allocates bandwidth and priority based on real-time perception of home scenarios (such as "family movie watching" and "multi-person video conferencing") and device requirements and scenario priorities. Scenario-driven resource allocation: For the first time, home collaboration scenarios are used as the core basis for QoS allocation, rather than relying on device type or traffic classification; Multi-dimensional scenario recognition: Data such as device status, user behavior, time, and location are integrated to accurately identify collaboration scenarios; Dynamic scenario-device mapping: Participating devices and their QoS requirements are automatically matched according to scenario type, achieving efficient resource utilization. Furthermore, it improves the accuracy of resource allocation in multi-device collaboration scenarios, avoiding resource waste; and ensures resource requirements for high-priority scenarios (such as health monitoring), improving user experience. Meanwhile, it has the following beneficial effects: Congestion state perception and response: By monitoring network bandwidth, latency, packet loss rate and other indicators in real time, it can quickly identify congestion and trigger adjustments; Scenario priority-driven resource preemption: When congested, it prioritizes the bandwidth and latency requirements of high-priority scenarios (such as health monitoring); Local collaborative priority strategy: Reduces dependence on the cloud and improves the efficiency of resource allocation in weak network environments; Technical effects: Improves the stability of critical devices when the network is congested, avoids health data loss or video stuttering; Ensures QoS of high latency-sensitive services (such as video calls and health monitoring), meeting the actual needs of home users.

[0137] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0138] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to 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 device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0139] Example 2

[0140] According to embodiments of the present invention, a dynamic allocation apparatus for network resources is also provided for implementing the above-described dynamic allocation method for network resources. Figure 4 This is a schematic diagram of a dynamic allocation device for network resources according to an embodiment of the present invention, such as... Figure 4 As shown, the dynamic allocation device for network resources includes: an acquisition unit 401, a processing unit 403, a calculation unit 405, a generation unit 407, and an allocation unit 409. The dynamic allocation device will be described below.

[0141] The acquisition unit 401 is used to acquire device data, user behavior data, and environmental data of each smart device in the home network.

[0142] The processing unit 403 is used to process device data, user behavior data and environmental data through a scene recognition model to obtain the home collaboration scene type of the home network. The scene recognition model is a model trained by machine learning using multiple sets of training data. Each set of training data includes: sample device data, sample user behavior data and sample environmental data, and the sample home collaboration scene type corresponding to the sample device data, user behavior data and sample environmental data.

[0143] The calculation unit 405 is used to calculate the scenario priority of the family collaboration scenario type based on the family collaboration scenario type, user identity, and time sensitivity.

[0144] The generation unit 407 is used to generate network resource allocation strategies based on scenario priority and device requirement information.

[0145] Allocation unit 409 is used to allocate network resources to each smart device according to the network resource allocation strategy.

[0146] It should be noted that the above-mentioned acquisition unit 401, processing unit 403, calculation unit 405, generation unit 407 and allocation unit 409 correspond to steps S202 to S210 in the above embodiments. The five units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments.

[0147] As can be seen from the above, in the solution described in the above embodiments of the present invention, the acquisition unit can acquire device data, user behavior data, and environmental data of each smart device in the home network; the processing unit can process the device data, user behavior data, and environmental data through a scene recognition model to obtain the home collaboration scene type of the home network. The scene recognition model is a model trained using multiple sets of training data through machine learning. Each set of training data includes: sample device data, sample user behavior data, sample environmental data, and a sample home collaboration scene type corresponding to the sample device data, user behavior data, and sample environmental data; the calculation unit can calculate the scene priority of the home collaboration scene type based on the home collaboration scene type, user identity, and time sensitivity; and the generation unit can calculate the scene priority based on the scene priority and device behavior data. The system generates network resource allocation strategies based on demand information. Using an allocation unit, it allocates network resources to each smart device according to these strategies. This achieves the goal of dynamically allocating network resources based on real-time capture of the operating status of all smart devices (including bandwidth consumption, latency tolerance, and network connection status), analysis of user habits (such as device usage frequency and time), and environmental indicators. Based on the captured information, it performs scene identification and calculates the priority of the identified scene type in conjunction with user identity and time sensitivity. This improves the sensitivity to changes in the home network environment and, through real-time monitoring and scene identification, enables precise resource allocation, avoids unnecessary bandwidth waste, and improves the overall utilization efficiency of home network resources.

[0148] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem that the resource allocation method based on static QoS configuration or simple traffic classification in the related technology cannot adapt to the needs of multi-device collaboration and dynamic changes in home networks, resulting in a disconnect between resource allocation and actual needs.

[0149] Optionally, the acquisition unit includes: a first acquisition module, used to acquire bandwidth demand data, latency sensitivity data and network status data from various smart devices through network resource sensing components in the home network to obtain device data; an analysis module, used to analyze user interaction logs to obtain user behavior data; and a second acquisition module, used to acquire environmental data through environmental sensing components in the home network.

[0150] Optionally, the processing unit includes: a preprocessing module for preprocessing device data, user behavior data, and environmental data to obtain preprocessed device data, user behavior data, and environmental data; and a processing module for inputting the preprocessed device data, user behavior data, and environmental data into a scene recognition model to process the preprocessed device data, user behavior data, and environmental data using the scene recognition model to obtain a home collaboration scene type.

[0151] Optionally, the processing unit includes: a transformation module for transforming user behavior data into time-series feature data; a fusion module for fusing environmental data and device data to generate a scene feature vector; and a classification module for classifying the scene feature vector according to the time-series feature data to obtain the home collaboration scene type.

[0152] Optionally, the generation unit includes: an extraction module, used to extract bandwidth requirement data and latency sensitivity data of each smart device from a scene device mapping table to obtain device requirement information, wherein the scene device mapping table is pre-generated to record the bandwidth required by each smart device in different collaborative scenarios; a first calculation module, used to calculate the device priority of each smart device according to the device type and scene priority of each smart device; and a second calculation module, used to calculate the bandwidth allocation ratio of each smart device in the home collaborative scenario type according to the device priority, so as to obtain a network resource allocation strategy.

[0153] Optionally, the dynamic allocation device further includes: a monitoring unit, used to monitor the home network after allocating network resources to each smart device according to the network resource allocation strategy, so as to obtain the monitoring results of the home network; an adjustment unit, used to adjust the resource allocation strategy when the monitoring results indicate that the home network is in a congested state, so as to obtain the adjusted resource allocation strategy; and an allocation unit, used to allocate network resources to each smart device according to the adjusted resource allocation strategy.

[0154] Optionally, the monitoring unit includes: a monitoring module for monitoring the bandwidth utilization, packet loss rate, and latency jitter of the home network, and obtaining monitoring results.

[0155] Optionally, the adjustment unit includes: a third calculation module for calculating the congestion index of the home network based on bandwidth utilization, packet loss rate, and latency jitter; and an adjustment module for adjusting the resource allocation strategy based on the congestion index and scenario priority to obtain the adjusted resource allocation strategy.

[0156] Optionally, the dynamic allocation device further includes an acquisition unit, which is also used to acquire a resource allocation strategy from the home network's ontological cache when the home network is detected to be in a weak network or offline state.

[0157] Optionally, the dynamic allocation device further includes: a collection unit, used to collect feedback data from each smart device after allocating network resources to each smart device according to the network resource allocation strategy, wherein the feedback data includes: actual bandwidth usage data and latency index data; and an optimization unit, used to optimize the scene recognition model and resource allocation strategy based on the feedback data.

[0158] According to another aspect of the present invention, a smart home system is also provided, which uses the dynamic allocation method of network resources described above.

[0159] According to another aspect of the present invention, a processor is also provided, which is used to run a program, wherein the program executes the network resource dynamic allocation method described above during runtime.

[0160] According to another aspect of the present invention, a computer program product is also provided, including computer instructions, which, when executed by a processor, perform the method for dynamically allocating network resources as described above.

[0161] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes the network resource dynamic allocation method described above.

[0162] Optionally, in this embodiment, the computer-readable storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any communication device in a group of communication devices.

[0163] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: acquiring device data, user behavior data, and environmental data of each smart device in the home network; processing the device data, user behavior data, and environmental data through a scene recognition model to obtain the home collaboration scene type of the home network, wherein the scene recognition model is a model trained by machine learning using multiple sets of training data, each set of training data including: sample device data, sample user behavior data, sample environmental data, and sample home collaboration scene type corresponding to the sample device data, user behavior data, and sample environmental data; calculating the scene priority of the home collaboration scene type based on the home collaboration scene type, user identity, and time sensitivity; generating a network resource allocation strategy based on the scene priority and device demand information; and allocating network resources to each smart device according to the network resource allocation strategy.

[0164] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: collecting bandwidth demand data, latency sensitivity data, and network status data from various smart devices through network resource sensing components in the home network to obtain device data; analyzing user interaction logs to obtain user behavior data; and collecting environmental data through environmental sensing components in the home network.

[0165] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: preprocessing device data, user behavior data, and environmental data to obtain preprocessed device data, user behavior data, and environmental data; inputting the preprocessed device data, user behavior data, and environmental data into a scene recognition model to process the preprocessed device data, user behavior data, and environmental data using the scene recognition model to obtain a home collaboration scene type.

[0166] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: converting user behavior data into time-series feature data; fusing environmental data and device data to generate a scene feature vector; and classifying the scene feature vector according to the time-series feature data to obtain a home collaboration scene type.

[0167] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: extracting bandwidth requirement data and latency sensitivity data of each smart device from a scene device mapping table to obtain device requirement information, wherein the scene device mapping table is pre-generated to record the bandwidth required by each smart device in different collaborative scenarios; calculating the device priority of each smart device according to the device type and scene priority of each smart device; calculating the bandwidth allocation ratio of each smart device in the home collaborative scenario type according to the device priority to obtain a network resource allocation strategy.

[0168] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after allocating network resources to each smart device according to the network resource allocation strategy, monitoring the home network to obtain the monitoring results of the home network; when the monitoring results indicate that the home network is in a congested state, adjusting the resource allocation strategy to obtain the adjusted resource allocation strategy; and allocating network resources to each smart device according to the adjusted resource allocation strategy.

[0169] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: monitoring the bandwidth utilization, packet loss rate, and latency jitter of the home network, and obtaining monitoring results.

[0170] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: calculating the congestion index of the home network based on bandwidth utilization, packet loss rate, and latency jitter; adjusting the resource allocation strategy based on the congestion index and scenario priority to obtain the adjusted resource allocation strategy.

[0171] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: when the home network is detected to be in a weak or offline state, a resource allocation strategy is obtained from the home network's ontology cache.

[0172] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after allocating network resources to each smart device according to the network resource allocation strategy, collecting feedback data from each smart device, wherein the feedback data includes: actual bandwidth usage data and latency index data; and optimizing the scene recognition model and resource allocation strategy based on the feedback data.

[0173] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0174] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0175] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0176] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0177] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0178] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0179] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0180] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic allocation of network resources, characterized in that, The method comprises the following steps: obtaining device data, user behavior data and environment data of each smart device in a home network; processing the device data, the user behavior data and the environment data through a scene recognition model to obtain a home collaborative scene type of the home network, wherein the scene recognition model is a model trained by machine learning using multiple sets of training data, each of the multiple sets of training data comprising sample device data, sample user behavior data, sample environment data and a sample home collaborative scene type corresponding to the sample device data, the sample user behavior data and the sample environment data; calculating a scene priority of the home collaborative scene type according to the home collaborative scene type, user identity and time sensitivity; generating a network resource allocation strategy according to the scene priority and device demand information; allocating network resources to each smart device according to the network resource allocation strategy.

2. The method of claim 1, wherein, Obtaining device data, user behavior data and environment data of each smart device in a home network comprises: collecting bandwidth demand data, delay sensitivity data and network state data from each smart device through a network resource sensing component in the home network to obtain the device data; analyzing user interaction logs to obtain the user behavior data; collecting the environment data through an environment sensing component in the home network.

3. The method of claim 1, wherein, Processing the device data, the user behavior data and the environment data through a scene recognition model to obtain a home collaborative scene type of the home network comprises: preprocessing the device data, the user behavior data and the environment data to obtain preprocessed device data, preprocessed user behavior data and preprocessed environment data; inputting the preprocessed device data, the preprocessed user behavior data and the preprocessed environment data into the scene recognition model to process the preprocessed device data, the preprocessed user behavior data and the preprocessed environment data using the scene recognition model to obtain the home collaborative scene type.

4. The method of claim 1, wherein, Processing the device data, the user behavior data and the environment data through a scene recognition model to obtain a home collaborative scene type of the home network comprises: converting the user behavior data into time series feature data; fusing the environment data and the device data to generate a scene feature vector; classifying the scene feature vector according to the time series feature data to obtain the home collaborative scene type.

5. The method of claim 1, wherein, Generating a network resource allocation strategy according to the scene priority and device demand information comprises: extracting bandwidth demand data and delay sensitivity data of each smart device from a scene device mapping table to obtain the device demand information, wherein the scene device mapping table is a pre-generated table recording the bandwidth required by each smart device in different collaborative scenes; calculating a device priority of each smart device according to the device type of each smart device and the scene priority; The device priority is used to calculate a bandwidth allocation ratio of each smart device in the home collaborative scenario type, so as to obtain the network resource allocation strategy.

6. The method of claim 1, wherein, After the network resources are allocated to each smart device according to the network resource allocation strategy, the dynamic allocation method further comprises: monitoring the home network to obtain a monitoring result of the home network; when the monitoring result indicates that the home network is in a congestion state, adjusting the resource allocation strategy to obtain an adjusted resource allocation strategy; allocating network resources to each smart device according to the adjusted resource allocation strategy.

7. The method of claim 6, wherein, The monitoring of the home network to obtain a monitoring result of the home network comprises: monitoring the bandwidth utilization, packet loss rate and delay jitter of the home network to obtain the monitoring result.

8. The method of claim 7, wherein, The adjustment of the resource allocation strategy to obtain an adjusted resource allocation strategy comprises: calculating a congestion index of the home network according to the bandwidth utilization, packet loss rate and delay jitter; adjusting the resource allocation strategy according to the congestion index and the scenario priority to obtain the adjusted resource allocation strategy.

9. The method of claim 1, wherein, The dynamic allocation method further comprises: when it is monitored that the home network is in a weak network or disconnected state, obtaining the resource allocation strategy from the ontology cache of the home network.

10. The method of claim 1-9, wherein, After the network resources are allocated to each smart device according to the network resource allocation strategy, the dynamic allocation method further comprises: collecting feedback data of each smart device, wherein the feedback data comprises actual bandwidth usage data and delay index data; optimizing the scenario identification model and the resource allocation strategy according to the feedback data.

11. A device for dynamic allocation of network resources, characterized in that, comprises: an acquisition unit configured to acquire device data, user behavior data and environment data of each smart device in a home network; a processing unit configured to process the device data, user behavior data and environment data through a scenario identification model to obtain a home collaborative scenario type of the home network, wherein the scenario identification model is a model trained by machine learning using a plurality of sets of training data, each set of the plurality of sets of training data comprising sample device data, sample user behavior data, sample environment data and a sample home collaborative scenario type corresponding to the sample device data, sample user behavior data and sample environment data; a calculation unit configured to calculate a scenario priority of the home collaborative scenario type according to the home collaborative scenario type, user identity and time sensitivity; a generation unit configured to generate a network resource allocation strategy according to the scenario priority and device demand information; an allocation unit configured to allocate network resources to each smart device according to the network resource allocation strategy.

12. A smart home system, characterized by The smart home system uses the dynamic allocation method of network resources of any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium comprises a stored program, wherein the program executes the dynamic allocation method of network resources of any one of claims 1 to 10.

14. A computer program product comprising computer instructions, characterized in that, The computer instructions, when executed by a processor, perform the method of claim 1-10 for dynamic allocation of network resources.

Images