Gateway configuration method, device and equipment based on PON (Passive Optical Network), and storage medium

Abstract

The invention relates to the technical field of computer networks, in particular to a gateway configuration method and device based on a PON network, computer equipment and a storage medium. In addition, the gateway configuration method based on the PON network can also be applied to fiber to room (Fiber To Room, FTTR), enterprise-level fiber to room (Fiber To Room-Bus, FTTR-B) and broadband fusion terminal products. According to the method, hardware state information and flow state information of gateway equipment and user behavior data of client equipment are obtained and input into a parameter prediction model, and multi-dimensional hardware operation parameters of a hardware resource use state of the client equipment in a future time period are predicted through the parameter prediction model; the multi-dimensional hardware operation parameters comprise operation parameters of different hardware in the gateway equipment; compared with the prior art, dynamic parameter configuration of the gateway equipment can be achieved, and the operation efficiency of the gateway equipment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer networks, and in particular to a gateway configuration method, a gateway configuration device, a computer device, and a computer-readable storage medium based on a PON network. Background Art

[0002] A Passive Optical Network (PON) is a telecommunications network that transmits data over optical fiber lines. It is a key technology supporting network access technologies such as Fiber to the Room (FTTR) and Fiber to the Room-Business (FTTR-B). FTTR / FTTR-B extend optical fiber to every room, installing fiber optic terminal equipment in each room to convert optical signals into electrical signals. Typically, fiber optic terminal equipment functions as a gateway device, performing data forwarding and protocol conversion. However, in related technologies, gateway devices suffer from poor operational efficiency. Summary of the Invention

[0003] The embodiments of the present invention provide a gateway configuration method, apparatus, computer equipment and storage medium based on a PON network, which realize dynamic parameter configuration of the gateway device and improve the operation efficiency of the gateway device.

[0004] On the one hand, the present invention provides a gateway configuration method based on a PON network, comprising: Obtain hardware status information, traffic status information of gateway devices and user behavior data of client devices; Inputting hardware status information, traffic status information, and user behavior data into a parameter prediction model, the parameter prediction model is used to predict multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in the future time period. The multi-dimensional hardware operating parameters include operating parameters of different hardware in the gateway device. Configure the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters.

[0005] Optionally, in one embodiment, the parameter prediction model includes a demand prediction sub-model and a parameter decision sub-model. Hardware status information, traffic status information, and user behavior data are input into the parameter prediction model. The parameter prediction model is used to predict multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in the future time period, including: Inputting hardware status information, traffic status information, and user behavior data into the demand prediction sub-model, and using the demand prediction sub-model to predict the hardware resource usage status of the client device in the future time period; According to the hardware resource usage status, parameter decisions are made through the parameter decision sub-model to obtain multi-dimensional hardware operation parameters.

[0006] Optionally, in one embodiment, according to the hardware resource usage status, parameter decision is performed through a parameter decision sub-model to obtain multi-dimensional hardware operating parameters, including: With the goal of matching hardware resource usage and minimizing gateway device power consumption, an online reinforcement learning algorithm is used to update the parameter decision sub-model. According to the hardware resource usage status, parameter decisions are made through the updated parameter decision sub-model to obtain multi-dimensional hardware operation parameters.

[0007] Optionally, in one embodiment, before obtaining the hardware status information and traffic status information of the gateway device and the user behavior data of the client device, the method further includes: Obtain historical hardware status information, historical traffic status information of gateway devices, and historical user behavior data of client devices; Obtaining labeled multi-dimensional hardware operating parameters corresponding to historical hardware status information, historical traffic status information, and historical user behavior data; Input historical hardware status information, historical traffic status information, and historical user behavior data into the parameter prediction model to obtain reference multi-dimensional hardware operating parameters; Model parameters of the parameter prediction model are updated according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters.

[0008] Optionally, in one embodiment, after updating the model parameters of the parameter prediction model according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the method further includes: According to the preset model compression strategy, the parameter prediction model is compressed.

[0009] Optionally, in one embodiment, before updating the model parameters of the parameter prediction model according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the method further includes: Obtaining initial model parameters corresponding to the parameter prediction model from the cloud server; Based on the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the model parameters of the parameter prediction model are updated, including: updating the initial model parameters according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters; Transmitting the updated model parameters to the cloud server, so that the cloud server aggregates the target model parameters based on the updated model parameters corresponding to the gateway device and the updated model parameters of other gateway devices, and returns the target model parameters; Update the model parameters of the parameter prediction model to the target model parameters.

[0010] Optionally, in one embodiment, after configuring the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters, the method further includes: Receive input parameter adjustment operations, and adjust hardware operating parameters of different hardware configurations according to the parameter adjustment operations; The adjusted multi-dimensional hardware operating parameters are obtained, and the model parameters of the parameter prediction model are updated according to the hardware status information, the traffic status information, the user behavior data, and the adjusted multi-dimensional hardware operating parameters.

[0011] In a second aspect, the present invention provides a gateway configuration device based on a PON network, comprising: Data collection module, used to obtain hardware status information and traffic status information of gateway devices and user behavior data of client devices; A parameter prediction module is used to input hardware status information, traffic status information, and user behavior data into a parameter prediction model. The parameter prediction model is used to predict multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in the future time period. The multi-dimensional hardware operating parameters include the operating parameters of different hardware in the gateway device; The parameter configuration module is used to configure the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters.

[0012] In a third aspect, the computer device provided by the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the gateway configuration method based on the PON network provided by the present invention is implemented.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the gateway configuration method based on a PON network provided by the present invention.

[0014] The present invention provides a gateway resource configuration solution based on a PON network. The solution inputs the acquired gateway device hardware status information, traffic status information, and user behavior data of a client device into a parameter prediction model, predicts the hardware resource usage status in a future time period, makes parameter decisions, and obtains multi-dimensional hardware operating parameters. This solution achieves dynamic configuration of different hardware operating parameters in the gateway device, avoids parameter rigidity, enables the gateway device to adapt to dynamically changing network loads, and improves the operating efficiency of the gateway device. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0016] Figure 1 1 is a flow chart of a gateway configuration method based on a PON network provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a model training process provided in an embodiment of the present invention; Figure 3 1 is a schematic structural diagram of a gateway configuration device based on a PON network provided by an embodiment of the present invention; Figure 4 It is a structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0019] It will also be understood that the term "and / or" used in the present description and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0020] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0021] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present invention. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in yet other embodiments" appearing in various places throughout this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically stated. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically stated.

[0023] The present invention provides a gateway configuration method, device, computer equipment and storage medium based on a PON network, wherein hardware status information and traffic status information of a gateway device and user behavior data of a client device are obtained; the hardware status information, traffic status information and user behavior data are input into a parameter prediction model, and multidimensional hardware operating parameters of the hardware resource usage status of the client device in a future time period are predicted by the parameter prediction model, the multidimensional hardware operating parameters including the operating parameters of different hardware in the gateway device; and operating parameters of different hardware in the gateway device are configured according to the multidimensional hardware operating parameters.

[0024] Please refer to Figure 1 , is a flow chart of a gateway configuration method based on a PON network disclosed in an embodiment of the present invention. This method is applicable to the scenario of parameter configuration of gateway devices in a network environment with multiple concurrent users and high dynamic load. Figure 1 As shown, the process of the gateway configuration method based on the PON network can be as follows: In 110 , hardware status information and traffic status information of the gateway device and user behavior data of the client device are obtained.

[0025] A gateway device is a node that connects different networks and implements data forwarding and protocol conversion. In network communications, a gateway device is located between the local area network (LAN) and the external network, and is used to receive, process, and distribute data traffic from various client devices. It not only has basic routing functions, but also integrates traffic scheduling, resource management, access control, and other functions to achieve unified management and resource coordination of multiple client devices. For example, in a home or office scenario, a gateway device can simultaneously support the access of multiple client devices such as mobile phones and computers, providing network access services for these client devices.

[0026] Hardware status information describes the usage and load of different hardware components in the gateway device, including but not limited to the status of key hardware components such as the processor, memory, and storage. Embodiments of the present invention use this status information as an important basis for subsequent parameter prediction. For example, the acquired hardware status information may include the gateway device's CPU utilization, memory occupancy, memory read and write times, power supply voltage fluctuations, and temperature sensor readings.

[0027] Traffic status information is a collection of network metrics about the data transmission handled by the gateway device over different time periods, reflecting the current network load and communication quality. For example, traffic status information obtained includes data throughput, uplink / downlink bandwidth utilization, average latency, maximum latency, packet loss rate, and number of retransmissions.

[0028] User behavior data is used to reflect the usage patterns and access characteristics of client devices in the network. For example, it usually includes information such as the client device's request type, access frequency, and connection duration.

[0029] For example, an embedded agent can be used to obtain hardware and traffic status information for a gateway device. The embedded agent is deployed within the gateway device and periodically calls a system interface to collect relevant operational data, which is then organized and reported in a pre-set format. For example, the embedded agent obtains the current CPU utilization and memory usage of the gateway device based on the system call interface, monitors the uplink / downlink bandwidth usage and packet loss rate within a certain time window through the network protocol stack, and reports the collected data to the gateway device in a log file or structured data format.

[0030] In addition, with user authorization, the gateway device can collect user behavior data from client devices through deep packet inspection (DPI), application layer protocol identification, or log analysis.

[0031] In 120, hardware status information, traffic status information and user behavior data are input into a parameter prediction model, and the parameter prediction model is used to predict multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in the future time period. The multi-dimensional hardware operating parameters include operating parameters of different hardware in the gateway device.

[0032] It should be noted that in this embodiment of the present invention, a pre-trained parameter prediction model is configured to take as input the hardware status information and traffic status information of the gateway device, as well as the user behavior data of the client device, and to output as output the multi-dimensional hardware operating parameters corresponding to the hardware resource usage status of the gateway device required by the client device in a future time period. The model price and training method of this parameter prediction model are not specifically limited herein.

[0033] Accordingly, the gateway device can perform standardization, normalization, denoising and other pre-processing on the acquired hardware status information, traffic status information and user behavior data, and then extract time series features (such as sliding window statistics) and event features (such as burst traffic markers) and input them into the parameter prediction model. The parameter prediction model can predict the multi-dimensional hardware operating parameters corresponding to the hardware resource usage status of the gateway device required by the client device in the future time period. Exemplarily, the multi-dimensional hardware operating parameters include the CPU frequency of the gateway device, memory usage strategy, network interface bandwidth allocation ratio, etc.

[0034] In 130 , operating parameters of different hardware in the gateway device are configured according to the multi-dimensional hardware operating parameters.

[0035] According to the multi-dimensional hardware operating parameters, corresponding operation configuration operations are performed on each hardware in the gateway device. This process covers the parameter configuration of different hardware. For example, for the processor module, the CPU's main frequency and voltage can be configured, some cores can be enabled or disabled, or switched to low-power mode, etc., according to the relevant operating parameters of the CPU in the multi-dimensional hardware operating parameters, so as to control the balance between computing resources and energy consumption; for the memory module, the memory allocation strategy, cache cleaning mechanism or memory resource reservation can be configured according to the relevant operating parameters of the memory in the multi-dimensional hardware operating parameters; for the RF antenna, the working channel and frequency band mode of the RF antenna can be configured according to the relevant operating parameters of the RF antenna in the multi-dimensional hardware operating parameters, etc.; for the memory module, the memory module can be configured to idle standby according to the relevant operating parameters of the memory module in the multi-dimensional hardware operating parameters, etc.

[0036] Optionally, in one embodiment, the parameter prediction model includes a demand prediction sub-model and a parameter decision sub-model. Hardware status information, traffic status information, and user behavior data are input into the parameter prediction model. The parameter prediction model is used to predict multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in the future time period, including: Inputting hardware status information, traffic status information, and user behavior data into the demand prediction sub-model, and using the demand prediction sub-model to predict the hardware resource usage status of the client device in the future time period; According to the hardware resource usage status, parameter decisions are made through the parameter decision sub-model to obtain multi-dimensional hardware operation parameters.

[0037] In an embodiment of the present invention, in the first stage, the gateway device inputs hardware status information, traffic status information, and user behavior data into the demand forecasting sub-model. This demand forecasting sub-model can use time series modeling to model time-related resource change trends. For example, a long short-term memory (LSTM) network is used to extract temporal dependencies and short-term mutations from the input sequence, thereby predicting the usage status of various hardware resources in the future time period. Furthermore, the LSTM structure can be updated using the following formula: ; in, Represents the input features at the tth time step, including hardware usage status information, traffic status information, and user behavior data; Represents the hidden state of the previous moment, which contains the context information in the historical time series; It is the hidden state at the current moment, that is, the joint modeling of current and historical data, which is used as the input basis for subsequent predictions.

[0038] In the second stage, the predicted hardware resource usage status is input into the parameter decision sub-model as a decision basis. This sub-model makes parameter decisions and outputs a set of multidimensional hardware operating parameters that match the hardware resource usage status. The parameter decision sub-model can be constructed using a sampling transformer block.

[0039] Optionally, in one embodiment, according to the hardware resource usage status, parameter decision is performed through a parameter decision sub-model to obtain multi-dimensional hardware operating parameters, including: With the goal of matching the hardware resource usage status and minimizing the power consumption of gateway devices, an online reinforcement learning algorithm is used to update the parameter decision sub-model.

[0040] According to the hardware resource usage status, parameter decisions are made through the updated parameter decision sub-model to obtain multi-dimensional hardware operation parameters.

[0041] In an embodiment of the present invention, the gateway device dynamically updates the parameter decision submodel using an online reinforcement learning algorithm to match hardware resource usage and minimize power consumption. For example, the online reinforcement learning algorithm can utilize the Proximal Policy Optimization (PPO) algorithm. By introducing a policy change restriction mechanism, it avoids drastic fluctuations during the policy update process, such as policy jumps, extreme deviations in action selection distribution, and training oscillations. PPO controls the magnitude of each policy change using the following function: ; in, Represents the objective function of the PPO algorithm; Represents the expected value of the sampled data at time step t, that is, the average over multiple samples; Represents the probability ratio of the current strategy to the old strategy; represents the advantage function estimate; is the preset cutoff coefficient; Represents the probability ratio The cropping operation is limited to [ ] interval.

[0042] In actual implementation, the gateway device uses hardware resource usage as observation input, selects multidimensional hardware operating parameters based on the current policy (i.e., the parameter decision sub-model), and optimizes the policy based on system feedback from the executed actions (e.g., changes in power consumption, changes in hardware resource usage, etc.), thereby updating the model parameters of the parameter decision sub-model. Through continuous interaction, sampling, and parameter updates, the gateway device ultimately makes parameter decisions based on the hardware resource usage using the updated parameter decision sub-model, resulting in multidimensional hardware operating parameters that can be deployed and used to configure each hardware to match the hardware resource usage.

[0043] Optionally, in one embodiment, before obtaining the hardware status information and traffic status information of the gateway device and the user behavior data of the client device, the method further includes: Obtain historical hardware status information, historical traffic status information of gateway devices, and historical user behavior data of client devices; Obtaining labeled multi-dimensional hardware operating parameters corresponding to historical hardware status information, historical traffic status information, and historical user behavior data; Input historical hardware status information, historical traffic status information, and historical user behavior data into the parameter prediction model to obtain reference multi-dimensional hardware operating parameters; Model parameters of the parameter prediction model are updated according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters.

[0044] In this embodiment of the present invention, a gateway device first obtains historical hardware status information (e.g., CPU usage, memory usage, etc.), historical traffic status information (e.g., throughput, number of connections, etc.), and historical user behavior data (e.g., request type, connection frequency, etc.) as input samples, and then obtains labeled multidimensional hardware operating parameters corresponding to the historical hardware status information, historical traffic status information, and historical user behavior data as output labels. The labeled multidimensional hardware operating parameters are multidimensional hardware operating parameters corresponding to the historical hardware status information, historical traffic status information, and historical user behavior data, and are used by the gateway device in actual operation under the optimal hardware resource utilization state. They serve as supervised learning targets for the parameter prediction model.

[0045] The gateway device inputs the acquired historical hardware status information, historical traffic status information, and historical user behavior data into the parameter prediction model to obtain reference multi-dimensional hardware operating parameters. Please refer to the relevant description in the above embodiment for details, which will not be repeated here.

[0046] Furthermore, the gateway device obtains the difference between the reference multidimensional hardware operating parameters and the label multidimensional hardware operating parameters, and updates the model parameters of the parameter prediction model based on the difference until the preset stop condition is met. There is no specific restriction on what loss function to use to measure the difference between the reference multidimensional hardware operating parameters and the label multidimensional hardware operating parameters, and it can be configured by technical personnel in this field according to actual needs.

[0047] Optionally, in one embodiment, after updating the model parameters of the parameter prediction model according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the method further includes: According to the preset model compression strategy, the parameter prediction model is compressed.

[0048] The embodiments of the present invention do not impose specific limitations on the configuration of the preset model compression strategy. For example, the preset model compression strategy can be configured to perform neural network pruning on the parameter prediction model. Pruning involves identifying and removing redundant or low-contributing neuronal connections or channels in the model while maintaining the model's predictive power. Examples include weight threshold-based pruning (e.g., removing weights with small absolute values), structural sparsity-based channel pruning (e.g., deleting certain convolution kernels or node outputs), and dynamic pruning based on sensitivity assessment. The preset model compression strategy can also be configured to perform quantization on the parameter prediction model, mapping the floating-point weight parameters and activation values ​​in the original model to low-bit-width integers (e.g., converting 32-bit floating-point numbers to 8-bit fixed-point numbers) to reduce the model's storage space and computational complexity. Examples include symmetric quantization, asymmetric quantization, or perceptual training quantization. This ensures a high compression ratio within a manageable accuracy loss, thereby improving the operational efficiency of the parameter prediction model on the gateway device.

[0049] Optionally, in one embodiment, before updating the model parameters of the parameter prediction model according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the method further includes: Obtaining initial model parameters corresponding to the parameter prediction model from the cloud server; Based on the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, the model parameters of the parameter prediction model are updated, including: updating the initial model parameters according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters; Transmitting the updated model parameters to the cloud server, so that the cloud server aggregates the target model parameters based on the updated model parameters corresponding to the gateway device and the updated model parameters of other gateway devices, and returns the target model parameters; Update the model parameters of the parameter prediction model to the target model parameters.

[0050] In the embodiment of the present invention, in order to further improve the generalization ability and adaptability of the model, the gateway device can be Figure 2 The parameter prediction model is trained using the training method shown: Different gateway devices obtain initial model parameters corresponding to the parameter prediction model from the cloud server. The initial model parameters can be obtained by pre-training the parameter prediction model by the cloud server.

[0051] During the local training phase, each gateway device obtains the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, calculates the gradient of the local parameter prediction model, iteratively optimizes the initial model parameters of the parameter prediction model, obtains the locally updated model parameters, and encrypts the updated model parameters and uploads them to the cloud server; another invention is that the cloud server receives updated model parameters from multiple different gateway devices, adopts a federal averaging or weighted aggregation strategy, aggregates the target model parameters, and then sends the aggregated target model parameters to each gateway device.

[0052] After receiving the target model parameters from the cloud server, each gateway device updates the model parameters of the local parameter prediction model to the target model parameters accordingly.

[0053] Optionally, in one embodiment, after configuring the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters, the method further includes: Receive input parameter adjustment operations, and adjust hardware operating parameters of different hardware configurations according to the parameter adjustment operations; The adjusted multi-dimensional hardware operating parameters are obtained, and the model parameters of the parameter prediction model are updated according to the hardware status information, the traffic status information, the user behavior data, and the adjusted multi-dimensional hardware operating parameters.

[0054] The gateway device can receive parameter adjustment operations input by users, such as hardware configuration modification requests entered through the management interface, remote configuration commands, or app control instructions. Parameter adjustment operations can specify desired operating parameters or constraint policies for specific hardware modules (e.g., processors, wireless modules, memory). Examples include manually setting the CPU operating frequency, forcing a Wi-Fi channel switch, or adjusting the memory usage limit. Upon receiving such operations, the gateway device adjusts the currently configured multi-dimensional hardware operating parameters based on the user input.

[0055] A new set of training samples is constructed based on the adjusted multi-dimensional hardware operating parameters, current hardware status information, traffic status information, and user behavior data. This is used as self-supervisory feedback input to fine-tune the parameters of the parameter prediction model to improve the model's adaptability to user behavior, avoid the system from repeatedly covering user preferences in the subsequent automatic optimization process, and meet the user's personalized needs.

[0056] From the above, it can be seen that the present invention provides a gateway resource configuration solution based on the PON network, which inputs the obtained gateway device hardware status information, traffic status information and user behavior data of the client device into the parameter prediction model, predicts the hardware resource usage status in the future time period, makes parameter decisions, and obtains multi-dimensional hardware operating parameters. In this way, dynamic configuration of different hardware operating parameters in the gateway device is realized, parameter rigidity is avoided, and the gateway device can adapt to the dynamically changing network load, thereby improving the operating efficiency of the gateway device.

[0057] To facilitate better implementation of the above-mentioned PON-based gateway configuration method, an embodiment of the present invention further provides a corresponding PON-based gateway configuration device. The meanings of the terms herein are the same as those in the above-mentioned PON-based gateway configuration method. For specific implementation details, please refer to the description in the above-mentioned method embodiment.

[0058] Please refer to Figure 3 The gateway configuration device based on the PON network may include a data acquisition module 210, a parameter prediction module 220 and a parameter configuration module 230. The functional modules are described in detail as follows: The data collection module 210 is used to obtain the hardware status information and traffic status information of the gateway device and the user behavior data of the client device; Parameter prediction module 220, configured to input hardware status information, traffic status information, and user behavior data into a parameter prediction model, and predict, using the parameter prediction model, multi-dimensional hardware operating parameters of the client device's hardware resource usage status in a future time period. The multi-dimensional hardware operating parameters include operating parameters of different hardware components in the gateway device. The parameter configuration module 230 is used to configure the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters.

[0059] Optionally, in one embodiment, the parameter prediction model includes a demand prediction sub-model and a parameter decision sub-model. The parameter prediction module 220 is used to input hardware status information, traffic status information and user behavior data into the demand prediction sub-model, and predict the hardware resource usage status of the client device in the future time period through the demand prediction sub-model; according to the hardware resource usage status, parameter decisions are made through the parameter decision sub-model to obtain multi-dimensional hardware operation parameters.

[0060] Optionally, in one embodiment, the parameter prediction module 220 is used to update the parameter decision sub-model using an online reinforcement learning algorithm with the goal of matching the hardware resource usage status and minimizing the power consumption of the gateway device; according to the hardware resource usage status, parameter decisions are made through the updated parameter decision sub-model to obtain reference multi-dimensional hardware operating parameters.

[0061] Optionally, in one embodiment, the PON network-based gateway configuration rule provided by the present invention also includes a machine learning module for obtaining historical hardware status information, historical traffic status information and historical user behavior data of the gateway device; obtaining labeled multi-dimensional hardware operating parameters corresponding to the historical hardware status information, historical traffic status information and historical user behavior data; inputting the historical hardware status information, historical traffic status information and historical user behavior data into a parameter prediction model to obtain reference multi-dimensional hardware operating parameters; and updating the model parameters of the parameter prediction model based on the difference between the reference multi-dimensional hardware operating parameters and the labeled multi-dimensional hardware operating parameters.

[0062] Optionally, in one embodiment, the machine learning module is further used to perform model compression processing on the parameter prediction model according to a preset model compression strategy.

[0063] Optionally, in one embodiment, the machine learning module is also used to obtain initial model parameters corresponding to the parameter prediction model from the cloud server; update the initial model parameters based on the difference between the reference multidimensional hardware operating parameters and the label multidimensional hardware operating parameters; transmit the updated model parameters to the cloud server, so that the cloud server aggregates the target model parameters based on the updated model parameters corresponding to the gateway device and the updated model parameters of other gateway devices, and returns the target model parameters; and update the model parameters of the parameter prediction model to the target model parameters.

[0064] Optionally, in one embodiment, the parameter configuration module 230 is also used to receive input parameter adjustment operations and adjust the hardware operating parameters of different hardware configurations according to the parameter adjustment operations; the data acquisition module 210 is also used to obtain the adjusted multi-dimensional hardware operating parameters; the machine learning module is also used to update the model parameters of the parameter prediction model based on hardware status information, traffic status information, user behavior data and the adjusted multi-dimensional hardware operating parameters.

[0065] The specific definitions of the PON-based gateway configuration device can be found in the definitions of the PON-based gateway configuration method described above and will not be repeated here. Each module in the PON-based gateway configuration device described above can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0066] In one embodiment, a computer device is provided. The computer device may be a gateway device, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to connect to an external wireless client and provide wireless network access services for the connected wireless client. When the computer program is executed by the processor, it implements the gateway configuration method based on the PON network provided by the present invention.

[0067] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the gateway configuration method based on the PON network in the above embodiment is implemented.

[0068] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the gateway configuration method based on the PON network in the above embodiment is implemented.

[0069] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0070] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0071] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A gateway configuration method based on a PON network, comprising: Obtain hardware status information, traffic status information of gateway devices and user behavior data of client devices; Inputting the hardware status information, the traffic status information, and the user behavior data into a parameter prediction model, and predicting multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in a future time period through the parameter prediction model, wherein the multi-dimensional hardware operating parameters include operating parameters of different hardware in the gateway device; The operating parameters of different hardware in the gateway device are configured according to the multi-dimensional hardware operating parameters.

2. The gateway configuration method based on a PON network according to claim 1, wherein the parameter prediction model includes a demand prediction sub-model and a parameter decision sub-model, and the inputting of the hardware status information, the traffic status information, and the user behavior data into the parameter prediction model to predict the multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in a future time period through the parameter prediction model includes: Inputting the hardware status information, the traffic status information, and the user behavior data into the demand prediction sub-model, and predicting the hardware resource usage status of the client device in a future time period through the demand prediction sub-model; According to the hardware resource usage status, parameter decision is performed through the parameter decision sub-model to obtain the multi-dimensional hardware operation parameters.

3. The gateway configuration method based on a PON network according to claim 2, wherein the step of performing parameter decision-making based on the hardware resource usage status by using the parameter decision sub-model to obtain the multi-dimensional hardware operating parameters comprises: With the goal of matching the hardware resource usage status and minimizing the power consumption of the gateway device, an online reinforcement learning algorithm is used to update the parameter decision sub-model; According to the hardware resource usage status, parameter decision is performed through the updated parameter decision sub-model to obtain the multi-dimensional hardware operation parameters.

4. The PON network-based gateway configuration method according to claim 1, before obtaining the hardware status information and traffic status information of the gateway device and the user behavior data of the client device, further comprising: Obtaining historical hardware status information, historical traffic status information of the gateway device and historical user behavior data of the client device; Obtaining label multi-dimensional hardware operating parameters corresponding to the historical hardware status information, the historical traffic status information, and the historical user behavior data; Inputting the historical hardware status information, the historical traffic status information, and the historical user behavior data into the parameter prediction model to obtain reference multi-dimensional hardware operating parameters; Model parameters of the parameter prediction model are updated according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters.

5. The PON network-based gateway configuration method according to claim 4, further comprising: after updating the model parameters of the parameter prediction model based on the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters; According to the preset model compression strategy, the parameter prediction model is subjected to model compression processing.

6. The PON network-based gateway configuration method according to claim 4, before updating the model parameters of the parameter prediction model based on the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters, further comprising: Obtaining initial model parameters corresponding to the parameter prediction model from a cloud server; The updating of the model parameters of the parameter prediction model according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters includes: updating the initial model parameters according to the difference between the reference multi-dimensional hardware operating parameters and the label multi-dimensional hardware operating parameters; Transmitting the updated model parameters to the cloud server, so that the cloud server aggregates target model parameters based on the updated model parameters corresponding to the gateway device and the updated model parameters of other gateway devices, and returns the target model parameters; The model parameters of the parameter prediction model are updated to the target model parameters.

7. The PON network-based gateway configuration method according to claim 1, after configuring the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters, further comprising: receiving input parameter adjustment operations, and adjusting hardware operating parameters of different hardware configurations according to the parameter adjustment operations; The adjusted multi-dimensional hardware operating parameters are obtained, and the model parameters of the parameter prediction model are updated according to the hardware status information, the traffic status information, the user behavior data, and the adjusted multi-dimensional hardware operating parameters.

8. A gateway configuration device based on a PON network, characterized in that: include: Data collection module, used to obtain hardware status information and traffic status information of gateway devices and user behavior data of client devices; a parameter prediction module, configured to input the hardware status information, the traffic status information, and the user behavior data into a parameter prediction model, and predict, using the parameter prediction model, multi-dimensional hardware operating parameters of the hardware resource usage status of the client device in a future time period, the multi-dimensional hardware operating parameters including operating parameters of different hardware in the gateway device; A parameter configuration module is used to configure the operating parameters of different hardware in the gateway device according to the multi-dimensional hardware operating parameters.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the gateway configuration method based on the PON network according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the gateway configuration method based on the PON network according to any one of claims 1 to 7 is implemented.

Images