Method and system for processing downlink bandwidth allocation information

Abstract

The present application provides a method and system for processing downlink bandwidth allocation information, which relates to the field of terminal equipment in FTTR networking. The method comprises: obtaining device data collected from IoT devices, obtaining network bandwidth data of the optical fiber network where the local device resides, and obtaining network traffic data of the optical fiber network; performing local preprocessing on the device data, network bandwidth data, and network traffic data to obtain first data to be analyzed; obtaining a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information, and controlling the slave device to transmit data in the transmission time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction. The above technical solution solves the problem in the related art of inability to flexibly adjust bandwidth during bandwidth allocation, resulting in low service efficiency.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of terminal equipment in FTTR networking, and more specifically, to a method and system for processing downlink bandwidth allocation information. Background Art

[0002] With the continuous development of optical networks, broadband network services are gradually moving towards the F5.5G era with FTTR (Fiber to the Room) and 50G-PON as the mainstream. Compared with previous generations of fixed access technologies, F5.5G has a series of excellent features such as enhanced fixed bandwidth, all-optical connection, and real-time resilient connection. Figure 1 As shown, the F5.5G home private network builds on fiber-to-the-home (FTTH) by extending fiber to every room, achieving all-optical networking within the home. Combining 10G-PON, 50G-PON, Wi-Fi 6, and Wi-Fi 7 technologies, it delivers gigabit coverage throughout the home, resolving issues such as insufficient Wi-Fi signal coverage and substandard speeds, and providing secure and reliable gigabit coverage throughout the home. The signal coverage distance for home networking is 50 or 100 meters. The signal coverage distance for commercial micro networking is 200 meters, and the signal coverage distance for Wi-Fi access is 10 meters.

[0003] FTTR consists of a master device, slave devices, and an indoor fiber-optic distributed network. Based on a fiber-optic P2MP (Point to Multiple Point) physical topology, FTTR deploys a master device at the access point of a home or small or micro enterprise, creating a fully optical network for that home or small or micro enterprise. The master device connects to an OLTPON (Optical Line Terminal Passive Optical Network) port at the central office. Multiple slave devices are then connected indoors. These slave devices can be deployed to the desired area based on the layout of the home or small or micro enterprise, providing network coverage and high-quality network quality in each area.

[0004] FTTR currently uses the ITU-T G.fin and G.Xfin standards internationally, including G.fin SA, G.fin DLL / G.Xfin DLL, G.fin PHY / G.Xfin PHY. Domestically, standards for the Fiber-to-the-Room (FTTR) data link layer and FTTR physical layer are used. The FTTR data link layer requires dynamic bandwidth allocation (DBA), which includes three methods: SR-DBA, TM-DBA, and CO-DBA. SR-DBA allocates bandwidth based on status reports. The Main FTTR Unit (MFU) requests status reports from the Main FTTR Unit (MFU), which then responds to the Sub FTTR Unit (SFU). TM-DBA allocates bandwidth based on traffic monitoring. The MFU observes idle FEM / XFEM frames and compares them with the corresponding bandwidth mapping to allocate bandwidth. In the CO-DBA mode, collaborative bandwidth allocation is performed by the MFU based on the management functions outside the DLL layer or the information provided to the MFU by the application.

[0005] Regarding the CO-DBA bandwidth allocation method, the current FTTR standard does not specify how external management functions or applications at the DLL layer provide information to the MFU. This relatively fixed bandwidth allocation method lacks flexibility in bandwidth adjustment. Furthermore, when there are many devices connected to the MFU or traffic is congested, high-priority services may experience packet loss or increased latency, reducing service efficiency and impacting user experience. This indicates that the inability to flexibly adjust bandwidth during bandwidth allocation in related technologies leads to low service efficiency.

[0006] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the Invention

[0007] The embodiments of the present application provide a method and system for processing downlink bandwidth allocation information, so as to at least solve the problem of low service efficiency caused by the inability to flexibly adjust bandwidth during bandwidth allocation.

[0008] According to one aspect of an embodiment of the present application, a method for processing downlink bandwidth allocation information is provided, comprising: obtaining device data collected from an Internet of Things device, obtaining network bandwidth data of a fiber optic network where the local device is located, and obtaining network traffic data of the fiber optic network; performing local preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed; obtaining a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information, and controlling the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction.

[0009] In an exemplary embodiment, obtaining device data collected from an IoT device includes: collecting data from the IoT device through the master device to obtain device data; or collecting data from the IoT device through the master device and at least two slave devices to obtain device data.

[0010] In an exemplary embodiment, obtaining network bandwidth data of the optical fiber network where the local device is located includes: collecting bandwidth configuration data of the optical fiber network through the master device; and / or collecting bandwidth jitter data of the optical fiber network through at least two of the slave devices.

[0011] In an exemplary embodiment, the device data, the network bandwidth data, and the network traffic data are locally preprocessed to obtain first data to be analyzed, including: locally preprocessing the device data, the network bandwidth data, and the network traffic data according to a preprocessing step, and determining the first data to be analyzed based on the processing result of the local preprocessing; the preprocessing step at least includes: performing data cleaning on the device data, the network bandwidth data, and the network traffic data to obtain cleaned data; performing data format conversion on the cleaned data to obtain converted data; performing data standardization on the converted data to obtain standard data; performing data feature extraction on the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; and / or performing data anomaly detection on the standard data to obtain abnormal data and normal data, and data storing the normal data.

[0012] In an exemplary embodiment, controlling the slave device to transmit data within the allocated bandwidth indicated by the first downlink bandwidth allocation information according to the first decision instruction includes: sending the first decision instruction to the slave device through the master device, so that the slave device parses the first downlink bandwidth allocation information to obtain an allocation indicator corresponding to the device number of the slave device, wherein the allocation indicator corresponds to a sending time slot, and uplink physical data is sent within a time slot window of the sending time slot.

[0013] In an exemplary embodiment, a dynamic link library management channel for transmitting data is established between the master device and the slave device, and the dynamic link library management channel includes sub-management channels of different channel types, which are used to encapsulate data according to a data format corresponding to a predetermined channel type, and transmit the encapsulated data through the sub-management channel of the predetermined channel type.

[0014] In an exemplary embodiment, the sub-management channels of different channel types include a wavelength division multiplexing management communication channel (WMCC) management channel, a fiber management communication channel (FMCC) management channel, a forward path linear organization access method (F-PLOAM) management channel, and an embedded operation, administration, and maintenance (OAM) management channel. Before encapsulating data according to a data format corresponding to a predetermined channel type and transmitting the encapsulated data through the sub-management channel of the predetermined channel type, the method further includes: determining the WMCC management channel, FMCC management channel, and F-PLOAM management channel as sub-management channels of the predetermined channel type.

[0015] The present application also proposes a method for processing downlink bandwidth allocation information, which is applied to a master device, wherein the master device is connected to at least two slave devices, and the master device and the slave devices are both local devices, including: determining bandwidth allocation data based on device data collected from an Internet of Things device, network bandwidth data of a fiber optic network where the local device is located, and network traffic data of the fiber optic network, wherein the Internet of Things device is connected to the slave device; obtaining first data to be analyzed obtained after local preprocessing of the bandwidth allocation data, and obtaining a first decision instruction output by a local analysis model based on the first data to be analyzed, wherein the first decision instruction carries first downlink bandwidth allocation information; and sending the bandwidth allocation data to a cloud platform, obtaining a second decision instruction generated by the cloud platform based on the bandwidth allocation data; and controlling the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction and the second decision instruction.

[0016] In an exemplary embodiment, before sending the bandwidth allocation data to the cloud platform and obtaining the second decision instruction generated by the cloud platform based on the bandwidth allocation data, the method also includes: establishing a data transmission channel for transmitting data between the main device of the local device and the cloud platform; sending the bandwidth allocation data to the cloud platform through the data transmission channel, and obtaining the second decision instruction generated by the cloud platform based on the bandwidth allocation data through the data transmission channel.

[0017] The present application also proposes a method for processing downlink bandwidth allocation information, which is applied to a cloud platform, comprising: obtaining bandwidth allocation data from a local device, the bandwidth allocation data at least including device data collected by the local device from an IoT device, obtaining network bandwidth data of a fiber optic network where the local device is located, and obtaining network traffic data of the fiber optic network; performing cloud-based preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed; obtaining a second decision instruction output by a cloud-based analysis model based on the second data to be analyzed, the second decision instruction carrying second downlink bandwidth allocation information, and sending the second decision instruction to the local device, the local device being configured to control, according to the second decision instruction and the first decision instruction, a slave device in the local device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information.

[0018] In an exemplary embodiment, obtaining bandwidth allocation data from a local device includes: receiving the bandwidth allocation data through a data transmission channel, where the data transmission channel represents a channel for transmitting data between a master device of the local device and the cloud platform.

[0019] In an exemplary embodiment, the device data, the network bandwidth data, and the network traffic data are pre-processed in the cloud to obtain second data to be analyzed, including: pre-processing the device data, the network bandwidth data, and the network traffic data in the cloud according to the pre-processing steps, and determining the second data to be analyzed based on the processing results of the cloud pre-processing; the pre-processing steps at least include: cleaning the device data, the network bandwidth data, and the network traffic data to obtain cleaned data; converting the data format of the cleaned data to obtain converted data; standardizing the converted data to obtain standard data; extracting data features from the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; and / or detecting data anomaly on the standard data to obtain abnormal data and normal data, and storing the normal data.

[0020] According to another aspect of an embodiment of the present application, a system for processing downlink bandwidth allocation information is also provided, including a local device; the local device includes a master device and a slave device; the master device and the slave device are used to jointly obtain device data collected by the slave device from the Internet of Things device, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network; the master device and the slave device are also used to jointly perform local pre-processing on the device data, the network bandwidth data and the network traffic data to obtain first data to be analyzed; the master device is also used to obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information; the slave device is also used to control the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction.

[0021] According to another aspect of an embodiment of the present application, a system for processing downlink bandwidth allocation information is also provided, including: a data acquisition module arranged in a local device, the data acquisition module is connected to a monitoring feedback module, and is used to collect IoT devices to obtain device data, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network from the monitoring feedback module; a data preprocessing module, connected to the data acquisition module, and is used to locally preprocess the device data, the network bandwidth data, and the network traffic data from the data acquisition module to obtain first data to be analyzed; a model training and inference module located in the local device, a local analysis model pre-installed in the model training and inference module, the model training and inference module is connected to the data preprocessing module, and is used to receive a first decision instruction output by the local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information; a bandwidth allocation module, connected to the model training and inference module, and is used to control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction from the model training and inference module.

[0022] According to another aspect of an embodiment of the present application, a system for processing downlink bandwidth allocation information is further provided, including: a cloud platform and a local device; the cloud platform is configured to obtain bandwidth allocation data from the local device; perform cloud preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed; obtain a second decision instruction output by a cloud analysis model based on the second data to be analyzed, the second decision instruction carrying second downlink bandwidth allocation information, and send the second decision instruction to the local device; the local device is configured to collect data from an Internet of Things device to obtain device data, obtain network bandwidth data of a fiber optic network where the local device is located and network traffic data of the fiber optic network, determine the bandwidth allocation data based on the device data, the network bandwidth data, and the network traffic data, obtain first data to be analyzed obtained after local preprocessing of the bandwidth allocation data, obtain a first decision instruction output by the local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information, and control the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information based on the second decision instruction and the first decision instruction.

[0023] According to another aspect of the embodiment of the present application, a system for processing downlink bandwidth allocation information is further provided, including: a data acquisition module arranged in a local device, the data acquisition module is connected to a monitoring feedback module, and is used to collect data from IoT devices to obtain device data, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network from the monitoring feedback module; a data preprocessing module, including a local data preprocessing unit located in the local device and a cloud platform data preprocessing unit located in the cloud platform, the local data preprocessing unit is connected to the data acquisition module, and is used to locally preprocess the device data, the network bandwidth data and the network traffic data from the data acquisition module to obtain first data to be analyzed; the cloud platform data preprocessing unit is connected to the data acquisition module, and is used to preprocess the device data, the network bandwidth data and the network traffic data from the data acquisition module The network traffic data is pre-processed in the cloud to obtain the second data to be analyzed; a model training and inference module includes a local analysis model located on the local device and a cloud analysis model located on the cloud platform, the local analysis model is connected to the data pre-processing module, and is used to receive the first data to be analyzed, and output a first decision instruction based on the first data to be analyzed, the first decision instruction carries first downlink bandwidth allocation information; the cloud analysis model is connected to the data pre-processing module, and is used to receive the second data to be analyzed, and output a second decision instruction based on the second data to be analyzed, the second decision instruction carries second downlink bandwidth allocation information; a bandwidth allocation module is connected to the model training and inference module, and is used to control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information and the second downlink bandwidth allocation information according to the first decision instruction and the second decision instruction from the model training and inference module.

[0024] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the above-mentioned method for processing downlink bandwidth allocation information when running.

[0025] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for processing downlink bandwidth allocation information through the computer program.

[0026] According to another aspect of the present application, a computer program product is provided, including a computer program, which implements the steps in any one of the above method embodiments when executed by a processor.

[0027] The solution provided in this application realizes the efficient collection and processing of IoT device data through local devices. At the same time, it dynamically adjusts the downlink bandwidth allocation according to the real-time bandwidth and traffic data of the optical fiber network, reducing the burden of data transmission. It combines the local analysis model to output decision instructions, improves the accuracy and real-time performance of decision-making, solves the problem of low business efficiency caused by the inability to flexibly adjust bandwidth during the bandwidth allocation process, improves the efficiency and stability of data transmission, and provides a strong guarantee for the efficient operation of IoT devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The exemplary embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0029] Figure 1 This is a schematic diagram of the FTTR networking architecture in related technologies;

[0030] Figure 2 This is a hardware structure block diagram of a computer terminal for the method for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0031] Figure 3 is a flowchart (1) of a method for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0032] Figure 4 This is a flowchart (1) of the FTTR network AI DBA decision issuance according to an embodiment of the present application;

[0033] Figure 5 This is a flowchart (II) of the FTTR network AI DBA decision-making process according to an embodiment of the present application;

[0034] Figure 6 Schematic diagram (1) of FTTR network AI DBA data preprocessing according to an embodiment of the present application;

[0035] Figure 7 Schematic diagram of AI DBA data transmission in FTTR network according to an embodiment of the present application;

[0036] Figure 8 Schematic diagram (2) of FTTR network AI DBA data preprocessing according to an embodiment of the present application;

[0037] Figure 9 Schematic diagram of dynamic bandwidth adjustment of AI DBA in FTTR networking according to an embodiment of the present application;

[0038] Figure 10 is a flowchart (II) of a method for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0039] Figure 11 is a flowchart (3) of a method for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0040] Figure 12 This is a schematic diagram of the principle of collaborative processing of the FTTR network AI DBA model according to an embodiment of the present application;

[0041] Figure 13 Schematic diagram of AI DBA monitoring feedback for FTTR networking according to an embodiment of the present application;

[0042] Figure 14 is a structural block diagram of a system for processing downlink bandwidth allocation information according to an embodiment of the present application (I);

[0043] Figure 15 is a structural block diagram (II) of a system for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0044] Figure 16 is a structural block diagram (III) of a system for processing downlink bandwidth allocation information according to an embodiment of the present application;

[0045] Figure 17 This is a structural block diagram (four) of a system for processing downlink bandwidth allocation information according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or similar computing device. Taking running on a computer terminal as an example, Figure 2 1 is a hardware structure block diagram of a computer terminal of the method for processing downlink bandwidth allocation information according to an embodiment of the present application. Figure 2 As shown, the computer terminal may include one or more ( Figure 2 Only one is shown in the figure) a processor 202 (the processor 202 may include but is not limited to a microprocessor (Microprocessor Unit, referred to as MPU) or a programmable logic device (Programmable logic device, referred to as PLD)) and a memory 204 for storing data. In an exemplary embodiment, the computer terminal may also include a transmission device 206 for communication functions and an input / output device 208. It will be understood by those skilled in the art that Figure 2 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. For example, the computer terminal may also include Figure 2 More or fewer components than shown, or with Figure 2 Equivalent functions or comparisons shown Figure 2 Shown are different configurations with more functionality.

[0049] The memory 204 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the method for processing downlink bandwidth allocation information in the embodiments of the present application. The processor 202 executes the computer program stored in the memory 204 to execute various functional applications and data processing, thereby implementing the above-mentioned method. The memory 204 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 204 may further include a memory remotely located relative to the processor 202, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0050] Transmission device 206 is used to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by a computer terminal's communications provider. In one embodiment, transmission device 206 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, transmission device 206 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0051] Figure 3 This is a flowchart of a method for processing downlink bandwidth allocation information according to an embodiment of the present application (I), and the execution subject may be a local device, such as Figure 3 As shown, the steps of the method include:

[0052] Step S302: Acquire device data collected from the IoT device, acquire network bandwidth data of the optical fiber network where the local device is located, and acquire network traffic data of the optical fiber network.

[0053] Step S304: locally pre-process the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed.

[0054] Step S306: Obtain a first decision instruction output by the local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information, and control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction.

[0055] The embodiment of the present application obtains device data collected from the IoT device, obtains network bandwidth data of the optical fiber network where the local device is located, and obtains network traffic data of the optical fiber network; performs local preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed; obtains a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carries first downlink bandwidth allocation information, and controls the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction, thereby achieving efficient collection and processing of IoT device data based on local devices, and at the same time, dynamically adjusts downlink bandwidth allocation according to real-time bandwidth and traffic data of the optical fiber network, reducing the burden of data transmission, and outputs decision instructions in combination with the local analysis model, thereby improving the accuracy and real-time performance of decisions, solving the problem of low business efficiency caused by the inability to flexibly adjust bandwidth during bandwidth allocation, improving the efficiency and stability of data transmission, and providing a strong guarantee for the efficient operation of IoT devices.

[0056] Optionally, IoT devices include, but are not limited to, various sensors, actuators, or other smart devices. IoT devices are connected to the optical fiber network via slave devices to collect environmental data, device status, and other information.

[0057] Among them, the network bandwidth data of the fiber optic network reflects the actual available bandwidth of the network, while the network traffic data represents the amount of data transmitted in the current network.

[0058] It should be noted that the local analysis model can predict the optimal downlink bandwidth allocation strategy based on the pre-processed data. The training process for the above local analysis model is as follows:

[0059] Obtain training data: Use the MFU's data acquisition module to collect device operating status information and environmental data, as well as obtain device anomaly reports, user behavior data, and other data from the monitoring feedback module.

[0060] Data preprocessing, formatting, feature extraction, and data storage are performed sequentially. Data preprocessing includes local cleaning to remove invalid or erroneous data to ensure model training quality. Formatting standardizes data formats for easier model access. Feature extraction selects key features to reduce computational complexity. For example, when processing image data using MobileNetV2, only edge information or color distribution features can be retained. Data storage stores processed data in a local database for easy access.

[0061] Then, a simple model such as MobileNetV2 is used for training. Due to resource constraints, iterative training with small batches of data is typically used. Finally, performance, such as accuracy and response time, is tested on an independent dataset. The model that meets the performance requirements is selected as the local analysis model.

[0062] The method of this embodiment corresponds to the scenario where the cloud platform is disconnected or does not include the cloud platform in the FTTR network AI DBA (Artificial Intelligence Dynamic Bandwidth Allocation), such as Figure 4 As shown in steps 4.1 to 4.9, MFU ( Figure 4 For example, if two SFUs are connected to the main device of the MFU, MFU and SFU1 ( Figure 4 Slave 1) and SFU2 ( Figure 4 Slave device 2) establishes DLL layer channels respectively, SFU1 and SFU2 periodically report bandwidth allocation related data through their respective established DLL channels, and MFU also collects local bandwidth allocation related data and network performance monitoring data. MFU and SFU pre-process the bandwidth allocation related data and monitoring data locally, and send them to the local analysis model for analysis, reasoning and decision-making. MFU adjusts the BWmap field according to the decision and sends it to SFU1 and SFU2. After receiving the BWmap field, SFU1 and SFU2 will respond to the bandwidth for data transmission respectively to ensure the bandwidth and latency of high-priority services in FTTR networking, and improve bandwidth utilization and user experience.

[0063] Among them, steps 4.1 to 4.9 are as follows:

[0064] 4.1 Complete the establishment of the DLL layer channel between MFU and SFU1.

[0065] 4.2 Complete the establishment of the DLL layer channel between MFU and SFU2.

[0066] 4.3 SFU1 bandwidth allocation related data is reported periodically.

[0067] 4.4 SFU2 bandwidth allocation related data is reported periodically.

[0068] 4.5 MFU and SFU collect data, pre-process monitoring data locally, and send it to the local model for analysis, reasoning, and decision-making.

[0069] 4.6 The MFU adjusts the BWmap domain based on the decision and issues it.

[0070] 4.7 SFU1 transmits data to MFU.

[0071] 4.8 The MFU adjusts the BWmap domain based on the decision and issues it.

[0072] 4.9 SFU2 transmits data to MFU.

[0073] Accordingly, the following embodiments using the cloud platform and local devices as the execution entities correspond to the scenario where the cloud platform is included in the FTTR network AI DBA. Figure 5 As shown in steps 5.1 to 5.14, MFU ( Figure 5 For example, if two SFUs are connected to the main device of the MFU and the cloud platform, MFU and SFU1 ( Figure 5 Slave device 1) and ( Figure 5Slave device 1 and SFU2 establish DLL layer channels. SFU1 and SFU2 periodically report bandwidth allocation data through their respective DLL channels. The MFU also collects local bandwidth allocation data and network performance monitoring data. The MFU and SFU bandwidth allocation and monitoring data are pre-processed locally and fed into the local analysis model for analysis and inference. The MFU and the cloud platform also establish cloud-based channels. The MFU periodically reports its bandwidth allocation data to the cloud platform. The cloud platform collects user experience monitoring data, pre-processes the bandwidth allocation and monitoring data in the cloud, and feeds the cloud-based analysis model for analysis and inference. The cloud platform then sends the analysis and inference results locally. The MFU makes a comprehensive decision based on the analysis and inference results of the cloud and local analysis models, adjusts the BWmap field accordingly, and sends it to SFU1 and SFU2. Upon receiving the BWmap field, SFU1 and SFU2 respond with their respective bandwidths for data transmission, ensuring bandwidth and latency for high-priority services in the FTTR network, improving bandwidth utilization and user experience.

[0074] Among them, steps 5.1 to 5.14 are as follows:

[0075] 5.1 Complete the establishment of the cloud channel between the cloud platform and MFU.

[0076] 5.2 Complete the establishment of the DLL layer channel between MFU and SFU1.

[0077] 5.3 Complete the establishment of the DLL layer channel between MFU and SFU2.

[0078] 5.4 SFU1 bandwidth allocation related data is reported periodically.

[0079] 5.5 SFU2 bandwidth allocation related data is reported periodically.

[0080] 5.6 Bandwidth allocation related data of MFU and SFU is reported periodically.

[0081] 5.7 The collected data and monitoring data of MFU and SFU are pre-processed locally and sent to the local model for analysis and reasoning.

[0082] 5.8 The collected data from MFU and SFU are pre-processed in the cloud and then sent to the cloud model for analysis and reasoning.

[0083] 5.9 The cloud platform sends the analysis and reasoning results to the MFU.

[0084] 5.10 The MFU makes comprehensive decisions based on the analysis and reasoning results of the cloud model and the local model.

[0085] 5.11 MFU adjusts the BWmap city based on the decision and sends it to SFU1.

[0086] 5.12 SFU1 transmits data to MFU.

[0087] 5.13 The MFU adjusts the BWmap domain based on the decision and sends it to SFU2.

[0088] 5.14 SFU2 transmits data to MFU.

[0089] In an exemplary embodiment, obtaining device data collected from an IoT device may include: collecting data from the IoT device through the master device to obtain device data; or collecting data from the IoT device through the master device and at least two of the slave devices to obtain device data.

[0090] The device data may include, but is not limited to, various physical or environmental parameters such as temperature, humidity, light intensity, and device health status.

[0091] In this embodiment, data collection, either through a master device alone or through collaboration between the master device and multiple slave devices, ensures data comprehensiveness and accuracy. This also increases system redundancy. Even if a slave device fails, other devices can continue data collection, ensuring continuous system operation. This flexible data collection approach adapts to IoT application scenarios of varying scale and complexity, improving the reliability and efficiency of data collection.

[0092] In an exemplary embodiment, obtaining network bandwidth data of the optical fiber network where the local device is located includes: collecting bandwidth configuration data of the optical fiber network through the master device; and / or collecting bandwidth jitter data of the optical fiber network through at least two of the slave devices.

[0093] Bandwidth configuration data, such as DBRu, idle frame percentage, and bandwidth allocation, represents basic network capacity information, while bandwidth jitter data reflects fluctuations in network bandwidth. This data is crucial for assessing network carrying capacity and stability. Joint data collection by master and slave devices more accurately reflects the true state of the network, aiding in the formulation of appropriate bandwidth allocation strategies, avoiding data transmission failures or delays caused by network fluctuations, and improving data transmission continuity and efficiency.

[0094] like Figure 6As shown, both the master and slave devices include data acquisition modules to collect data related to the downstream IOT (Internet of Things) devices or bandwidth. IOT devices cover a wide range, such as mobile terminals, smart door locks, mobile phones, and cameras. The data generated by IOT devices mainly includes service data and device data. Service data includes data generated by games, audio, IPTV (Internet Protocol Television), and OTT (Over The Top), while device data includes data generated by fingerprint authentication and device actions. These data can be collected by the MFU and SFU data acquisition modules, or collected uniformly by the MFU data acquisition module. In addition to IOT device data, the SFU data acquisition module also needs to collect bandwidth-related data such as DBRu and bandwidth jitter. The MFU data acquisition module also needs to collect related data such as DBRu, idle frame ratio, bandwidth configuration, and network traffic. After collection, the data will be transmitted to the data preprocessing module for unified processing.

[0095] Optionally, the master device and the slave device may communicate with each other via an IFDN (Integrated Frame Data Network) protocol, for example.

[0096] The "DBRu" (Database Request Unit) is the smallest unit of data read and write operations in a database. For example, DBRu is used to measure the read and write capacity of a disk; one DBRu corresponds to a certain number of data read and write operations.

[0097] In network communications, data is transmitted at the physical layer in frames. Idle frames, also known as idle frames, are sent by network devices to maintain link activity when no data is being transmitted. The idle frame percentage refers to the proportion of idle frames among all frames sent over a period of time. A high idle frame percentage may indicate low network utilization and inefficient data transmission.

[0098] In data communications, bandwidth jitter refers to the instability of a network connection's bandwidth over time. Typically, bandwidth should be constant, but in practice, due to factors such as network congestion, device performance variations, and signal interference, the actual bandwidth may fluctuate within a certain range. Excessive bandwidth jitter can affect data transmission quality, causing, for example, video streaming to freeze.

[0099] Bandwidth configuration refers to the data transmission speed settings of network devices or services. In wired or wireless networks, bandwidth is typically measured in Mbps (megabits per second) or Gbps (gigabits per second). Bandwidth configuration determines the maximum transmission speed of the network and how network resources are allocated between different services or applications.

[0100] In an exemplary embodiment, a dynamic link library management channel for transmitting data is established between the master device and the slave device, and the dynamic link library management channel includes sub-management channels of different channel types, which are used to encapsulate data according to the data format corresponding to the predetermined channel type, and transmit the encapsulated data through the sub-management channel of the predetermined channel type. The establishment of the dynamic link library management channel allows the master device and the slave device to exchange data and control instructions efficiently and flexibly. By using sub-management channels of different channel types, the most appropriate transmission method can be selected according to the data type and transmission requirements. For example, high-priority data can be transmitted through a high-speed channel, while low-frequency data can be transmitted through a low-speed but stable channel. This embodiment not only improves the efficiency of data transmission, but also enhances the stability and security of data transmission, and is suitable for various complex network environments and data transmission requirements.

[0101] Among them, the dynamic link library management channel corresponds to Figure 7 DDL management channel.

[0102] In an exemplary embodiment, the sub-management channels of different channel types include a wavelength division multiplexing management communication channel (WMCC) management channel, a fiber management communication channel (FMCC) management channel, a forward path linear organization access method (F-PLOAM) management channel, and an embedded operation, administration, and maintenance (OAM) management channel. Before encapsulating data according to a data format corresponding to a predetermined channel type and transmitting the encapsulated data through the sub-management channel of the predetermined channel type, the method further includes: determining the WMCC management channel, FMCC management channel, and F-PLOAM management channel as sub-management channels of the predetermined channel type.

[0103] Among them, the WMCC management channel uses wavelength division multiplexing technology to allow optical signals of different wavelengths to be transmitted on the same optical fiber, thereby improving the utilization rate of the optical fiber; the FMCC management channel is specifically used for management communications of the optical fiber network, ensuring the rapid transmission of control instructions; the F-PLOAM management channel is a management channel used for linear organization access of the forward path, and is suitable for chain network structures. Figure 7As shown in the figure, the DLL management channel between the MFU and SFU includes four channels: the WMCC (Wavelength Division Multiplexed Management Channel), the FMCC (Fiber Management Channel), the F-PLOAM (Forward Path Linearly Organized Access Method), and the embedded OAM (Operations, Administration, and Maintenance) channel. Data collected by the SFU data acquisition module can be transmitted through any of the WMCC, FMCC, or F-PLOAM management channels and encapsulated in the appropriate message format, such as WMCI, FMCI, or F-PLOAM messages. After transmission to the MFU, data can be processed locally or further to the cloud. By predetermining the use of these sub-management channels, the most appropriate channel can be selected for data transmission based on different data types and transmission requirements, thereby improving data transmission efficiency and network resource utilization.

[0104] In an exemplary embodiment, the device data, the network bandwidth data, and the network traffic data are locally preprocessed to obtain first data to be analyzed, including: locally preprocessing the device data, the network bandwidth data, and the network traffic data according to a preprocessing step, and determining the first data to be analyzed based on the processing result of the local preprocessing; the preprocessing step at least includes: performing data cleaning on the device data, the network bandwidth data, and the network traffic data to obtain cleaned data; performing data format conversion on the cleaned data to obtain converted data; performing data standardization on the converted data to obtain standard data; performing data feature extraction on the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; and / or performing data anomaly detection on the standard data to obtain abnormal data and normal data, and data storing the normal data.

[0105] Data cleansing removes invalid or erroneous data, data format conversion ensures data consistency, data standardization enables data from different sources to be compared on a consistent basis, and data feature extraction identifies key factors influencing bandwidth allocation. This preprocessing method significantly improves the accuracy and efficiency of subsequent data analysis, enabling rapid response to network changes and reducing decision-making biases caused by data quality issues.

[0106] like Figure 8 As shown in the figure, the data preprocessing module includes steps such as data cleaning, format conversion, standardization, feature extraction, anomaly detection, and data storage. Data cleaning primarily removes noise (outliers, erroneous data), fills missing values, and removes duplicates from the collected data. Format conversion primarily unifies the data format and timestamps. Standardization primarily unifies standards, such as standardizing key metrics like bandwidth and latency, ensuring consistency across all feature scales. Feature extraction extracts key features (bandwidth, latency, etc.) and simultaneously generates new features (peak bandwidth features, user behavior features). Anomaly detection primarily detects, marks, and removes anomalous data. Data storage primarily stores preprocessed data in a database for local and cloud-based analytical models to access, analyze, and infer. Preprocessing ensures the quality of data input to local analytical models, thereby improving the model's predictive accuracy.

[0107] In an exemplary embodiment, controlling the slave device to transmit data within the allocated bandwidth indicated by the first downlink bandwidth allocation information according to the first decision instruction specifically includes: sending the first decision instruction to the slave device through the master device, so that the slave device parses the first downlink bandwidth allocation information to obtain an allocation indicator corresponding to the device number of the slave device, wherein the allocation indicator corresponds to a sending time slot, and uplink physical data is sent within the time slot window of the sending time slot.

[0108] The first decision instruction contains information about the bandwidth resources that each slave device should use within a specific time window. By sending these instructions to the slaves via the master device, the slaves can ensure that they adhere to the optimized bandwidth allocation strategy for data transmission. The setting of transmission time slots prevents multiple devices from competing for limited bandwidth resources at the same time, thereby reducing data collisions and retransmissions, improving network utilization and data transmission efficiency.

[0109] like Figure 9As shown in the figure, after the MFU receives the AI ​​DBA decision and instructions, the bandwidth allocation module dynamically adjusts the bandwidth, primarily adjusting the G.fin or G.Xfin TAmap. The G.fin TAmap consists of the Alloc-ID (Allocation Identifier), StartTime (Start Time), and StopTime (End Time). The Alloc-ID is bound to a specific TCONT, and the bandwidth allocation timeslot is determined by the StartTime and StopTime. The G.Xfin TAmap consists of the Alloc-ID, StartTime, and GrantSize. The Alloc-ID is bound to a specific TCONT, and the bandwidth allocation timeslot is determined by the StartTime and GrantSize. After receiving the TAmap, the SFU parses it, determines the transmit timeslot based on the Alloc-ID, and sends an uplink PHY burst in the corresponding transmit timeslot.

[0110] Next, the processing of downlink bandwidth allocation information is described with the local device and the cloud platform as the execution entities respectively.

[0111] In this embodiment, the master device is connected to at least two slave devices, such as Figure 10 As shown, the process of processing the downlink bandwidth allocation information with the local device as the execution subject includes:

[0112] S1002, determining bandwidth allocation data based on device data collected from the IoT device, network bandwidth data of the optical fiber network where the local device is located, and network traffic data of the optical fiber network, wherein the IoT device is connected to the slave device;

[0113] S1004: Obtain first data to be analyzed obtained by locally preprocessing the bandwidth allocation data, and obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information;

[0114] S1006, and sending the bandwidth allocation data to the cloud platform, and obtaining a second decision instruction generated by the cloud platform based on the bandwidth allocation data;

[0115] S1008 : Control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction and the second decision instruction.

[0116] The master device determines a preliminary bandwidth allocation plan through a comprehensive analysis of IoT device data, network bandwidth data, and network traffic data. The master device then sends this data to the cloud platform, which leverages its powerful computing power for in-depth analysis and generates a more optimized bandwidth allocation strategy. This solves the problem of inefficient bandwidth allocation, which can lead to inefficient business operations, and improves the efficiency and stability of data transmission.

[0117] In an exemplary embodiment, before sending the bandwidth allocation data to the cloud platform and obtaining the second decision instruction generated by the cloud platform based on the bandwidth allocation data, a data transmission channel for transmitting data can also be established between the main device of the local device and the cloud platform; the bandwidth allocation data is sent to the cloud platform through the data transmission channel, and the second decision instruction generated by the cloud platform based on the bandwidth allocation data is obtained through the data transmission channel. By establishing the data transmission channel, the security and efficiency of data transmission between the main device and the cloud platform are ensured. Through this channel, the main device can transmit the collected device data, network bandwidth data and network traffic data losslessly to the cloud platform for in-depth analysis by the cloud analysis model. At the same time, the cloud platform can also use this channel to promptly feedback the analysis results, i.e., the second decision instruction, to the main device to guide it to perform bandwidth allocation. This embodiment can effectively shorten the decision cycle and improve the real-time nature of decision-making.

[0118] It should be noted that, in this embodiment, other solutions implemented by the master device and the slave device may refer to the above-mentioned method for processing downlink bandwidth allocation information with the local device as the execution subject, and this application will not elaborate on them here.

[0119] This application also proposes a method for processing downlink bandwidth allocation information applied to a cloud platform, such as Figure 11 Shown, including:

[0120] S1102: Obtain bandwidth allocation data from a local device, the bandwidth allocation data including at least device data collected by the local device from the IoT device, network bandwidth data of the optical fiber network where the local device is located, and network traffic data of the optical fiber network.

[0121] S1104, performing cloud pre-processing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed;

[0122] S1106, obtaining a second decision instruction output by the cloud analysis model based on the second data to be analyzed, where the second decision instruction carries second downlink bandwidth allocation information, and sending the second decision instruction to the local device. The local device is used to control the slave device in the local device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the second decision instruction and the first decision instruction.

[0123] In this embodiment, the cloud platform receives bandwidth allocation data, including IoT device status information, real-time network bandwidth, and traffic data. After cloud-based preprocessing, the data is input into a cloud-based analysis model to generate an optimized bandwidth allocation strategy. By sending the second decision instruction back to the local device, the cloud platform can assist the local device in more efficient bandwidth resource allocation, improving data transmission fluidity and network stability.

[0124] The training process for the cloud-based analysis model includes collecting device status information and other relevant data through the data acquisition module. Data preprocessing, formatting, feature extraction, and data storage are sequentially performed. Data preprocessing can be a cloud-based cleansing process. Subsequently, training is performed using a complex model based on Transformer technology. This allows the local analysis model to rapidly process real-time data, providing preliminary analysis results and immediate decision recommendations. Simultaneously, the cloud-based analysis model receives local preprocessed data and feedback for in-depth analysis, providing more accurate decision-making. When both are available, the master device (MFU) integrates the results of the local and cloud-based analysis models to make a comprehensive decision, improving decision accuracy. If the cloud connection is lost, the local analysis model assumes full decision-making responsibility. While accuracy may be reduced, it ensures that the device's basic functionality remains uninterrupted. The combination of local and cloud-based analysis models ensures real-time and robust device operation while improving the accuracy and flexibility of analytical decisions. This represents a key technology combination for implementing intelligent IoT applications.

[0125] In an exemplary embodiment, obtaining bandwidth allocation data from a local device includes: receiving the bandwidth allocation data through a data transmission channel, wherein the data transmission channel represents a channel for transmitting data between the master device of the local device and the cloud platform. By establishing a data transmission channel, a basis is provided for data transmission between the local device and the cloud platform. Through this channel, the cloud platform can receive bandwidth allocation data from the local device in real time, including device data, network bandwidth data, and network traffic data, so that it can conduct in-depth analysis and generate optimized bandwidth allocation strategies in a timely manner. This embodiment can effectively shorten the decision-making cycle and improve the real-time nature of decision-making.

[0126] like Figure 12As shown, the AI ​​model includes a local analysis model and a cloud analysis model, wherein the local analysis model is a simple model, such as MobileNetV2, YoLoV5 and ResNet50, and the cloud analysis model is a complex model such as DeepSeek, Qwen and OpenAI. Local data includes data collected by the MFU and SFU data acquisition modules and data from the monitoring feedback module. After the collection is completed, local pre-processing will be performed, including local cleaning, formatting, feature extraction and data storage, etc. The local pre-processed data will be transmitted to the local analysis model for analysis and reasoning. Cloud data includes data collected by the MFU and SFU data acquisition modules or local pre-processed data and monitoring module feedback data. After the collection is completed, cloud pre-processing will be performed, including cloud cleaning, formatting, feature extraction and data storage, etc. The cloud pre-processed data will be transmitted to the cloud analysis model for analysis and reasoning, and the analysis and reasoning results will be sent to the local. The MFU will make comprehensive decisions based on the inference results of the local analysis model and refer to the inference results of the cloud analysis model. When the cloud is disconnected, the local analysis model will perform analysis and reasoning and decision-making. Among them, Figure 13 As shown in Figure 1, the monitoring feedback module monitors key performance of local devices such as bandwidth utilization, packet loss rate, and alarms, and collects user experience data from the cloud.

[0127] In addition, by establishing a dynamic link library management channel and a data transmission channel, the data interaction between the master device and the slave device and the cloud platform is optimized, further enhancing the flexibility and reliability of the system.

[0128] In one exemplary embodiment, cloud-based preprocessing of the device data, network bandwidth data, and network traffic data to obtain second data to be analyzed may include: performing cloud-based preprocessing on the device data, network bandwidth data, and network traffic data according to the preprocessing steps, and determining the second data to be analyzed based on the results of the cloud-based preprocessing; the preprocessing steps at least including: data cleansing of the device data, network bandwidth data, and network traffic data to obtain cleaned data; data format conversion of the cleaned data to obtain converted data; data standardization of the converted data to obtain standard data; data feature extraction of the standard data to obtain a data feature vector, and determining the processing results based on the data feature vector; and / or data anomaly detection of the standard data to obtain anomaly data and normal data, and data storage of the normal data. The cloud-based preprocessing steps are similar to local preprocessing, but cloud platforms typically have more powerful computing resources and can handle more complex data preprocessing tasks. Data cleansing, format conversion, standardization, and feature extraction ensure the quality of data input to the cloud-based analysis model, thereby improving the model's prediction accuracy. Data anomaly detection helps cloud platforms identify network anomalies, such as equipment failures or cyberattacks, and take appropriate measures to protect network security. Through cloud-based preprocessing, cloud platforms can more effectively analyze and process large amounts of data, making more accurate decisions. This approach is applicable to a variety of scenarios requiring advanced data analysis capabilities, including but not limited to big data analysis, artificial intelligence applications, and network security monitoring, ensuring high system performance and reliability.

[0129] In addition, the above embodiments of the present application are not only applicable to FTTR systems, but also have certain applicability to FTTH systems, such as 10G-PON, 50G-PON, and VHSP systems. In the future, it is possible to combine with WIFI8 technology to provide a better user experience.

[0130] Next, the method for processing downlink bandwidth allocation information is further described in conjunction with the following embodiments.

[0131] Based on the above embodiment, the following further provides the AI ​​DBA processing process in FTTR networking:

[0132] 1. The data acquisition module performs data acquisition.

[0133] Both the MFU and SFU include data collection modules to collect data from connected IoT devices or bandwidth-related data. IoT device data can be collected by both the MFU and SFU data collection modules, or the MFU data collection module can collect IoT device data uniformly. In addition to IoT device data, the SFU data collection module also collects bandwidth-related data such as DBRu and bandwidth jitter. The MFU data collection module also collects data such as DBRu, idle frame ratio, and bandwidth configuration. After collection, the data is transmitted to the data preprocessing module for unified processing.

[0134] Among them, IOT devices have a wider range, such as PCs, smart door locks, mobile phones and cameras. IOT device data mainly includes service data and device data. Service data such as Game, IPTV, Voice, OTT, etc. generate data, and device data such as fingerprint authentication, device action, etc. generate data.

[0135] 2. Use DLL to manage the channel to transmit the collected data.

[0136] like Figure 7 As shown in the figure, the DLL (Dynamic Link Library) management channels established by the MFU and SFU respectively include four channels: WMCC (WDM Management Communication Channel), FMCC (Fiber Management Communication Channel), F-PLOAM (Forward Path Linearly Organized Access Method), and Embedded Operations, Administration and Maintenance (Embedded OAM).

[0137] The information collected by the SFU data acquisition module can be transmitted through any of the WMCC management channels, FMCC management channels, and F-PLOAM management channels, and encapsulated in the corresponding message format, such as WMCI messages, FMCI messages, or F-PLOAM messages, and then transmitted to the MFU for processing locally or further transmitted to the cloud.

[0138] 3. Make decisions and issue them.

[0139] The AI ​​models in the model training and inference module include local analysis models and cloud analysis models. Local analysis models represent simple models such as MobileNetV2, YoLoV5 and ResNet50 based on DNN, CNN and RNN technologies, while cloud analysis models represent complex models such as DeepSeek, Qwen and OpenAI based on Transformer technology.

[0140] Local data includes data collected by the MFU and SFU data acquisition modules and data fed back by the monitoring module. After collection is completed, local preprocessing will be performed, including local cleaning, formatting, feature extraction and data storage. The local preprocessed data will be transmitted to the local analysis model for analysis and reasoning.

[0141] Cloud data includes data collected by the MFU and SFU data acquisition modules or data pre-processed locally, as well as data fed back by the monitoring module. After collection is completed, cloud pre-processing will be performed. Cloud pre-processing includes cloud cleaning, formatting, feature extraction, and data storage.

[0142] The cloud-based pre-processed data will be transmitted to the cloud-based analysis model for analysis and reasoning, and the analysis and reasoning results will be sent locally. The MFU will make comprehensive decisions based on the reasoning results of the local analysis model and refer to the reasoning results of the cloud-based analysis model. When the cloud is disconnected, the local analysis model will perform analysis and reasoning and decision-making.

[0143] 4. Execute decisions and issue instructions to dynamically adjust bandwidth.

[0144] After the MFU receives the decision, the bandwidth allocation module dynamically adjusts the bandwidth, primarily adjusting the G.fin or G.XfinTAmap. "G.Fin" refers to the Gigabit-Capable Passive Optical Network (GPON) frame boundary indicator signal, used to synchronize the frame boundaries between the optical line terminal (OLT) and the optical network unit (ONU) to ensure correct data transmission. "G.Xfin" is a GPON control signal used by the OLT to send specific control information to the optical network unit (ONU), such as activating or deactivating the ONU or sending system parameter updates.

[0145] The Transmission Allocation Map (TAmap) is a mechanism used to allocate downstream bandwidth in GPON. Sent by the OLT, the TAmap contains information about the downstream bandwidth allocation for each ONU, including the allocated time slot, bandwidth size, and priority.

[0146] like Figure 9 As shown, G.fin TAmap is mainly composed of Alloc-ID (Allocation Identifier, allocation identifier, such as Figure 9 Alloc-ID is bound to a specific TCONT (Transmission Container). The bandwidth allocation time slot is determined by StartTime and StopTime. G.Xfin TAmap mainly consists of Alloc-ID (such as Figure 9 Alloc-ID is bound to a specific TCONT, and the bandwidth allocation time slot is determined by StartTime and GrantSize (grant size).

[0147] After receiving the TAmap, the slave device will analyze it, determine the transmission time slot according to the Alloc-ID, and send a PHY burst in the corresponding transmission time slot (corresponding to Figure 9 A PHY burst is a concept used in wireless communication systems, particularly in cellular networks, for data transmission. Uplink transmission refers to sending data from user devices (such as mobile phones, tablets, and IoT devices) to the base station. "PHY" refers to the physical layer, the lowest layer of the network protocol stack that handles hardware transmission and reception of signals. "Burst" refers to the concentrated transmission of large amounts of data in a short period of time. An uplink PHY burst can include data from multiple physical channels, such as control information (PUCCH, Physical Uplink Control Channel) and user data (PUSCH, Physical Uplink Shared Channel). The device bundles this data and sends it within a short time window to improve transmission efficiency and minimize interference.

[0148] 5. The monitoring feedback module collects user experience information.

[0149] like Figure 12 As shown in the figure, the monitoring and feedback module locally monitors network performance, collects user experience data, and optimizes feedback. Network performance monitoring is performed by the master device on the local network, primarily monitoring and collecting data on key performance indicators such as bandwidth utilization, packet loss rate, and alarms. User experience collection is performed in the cloud (also known as the cloud platform), primarily collecting user experience data, including user surveys and mobile app feedback. Network performance monitoring data and user experience data collected are regularly fed back to the local and cloud networks for preprocessing and model analysis and inference.

[0150] Through the above steps, data is collected from the MFU and SFU through the data collection module, collected data is transmitted through the DLL management channel, data analysis and reasoning and decision-making are issued through the local and cloud pre-processing modules and the model training and inference module, and the bandwidth allocation module executes the decision-making and issuing instructions to dynamically adjust the bandwidth. The monitoring feedback module collects and feedbacks user experience information after bandwidth changes to ensure high-priority service bandwidth and latency, improve bandwidth utilization and user experience. This application method is not only applicable to FTTR systems, but also to FTTH systems such as 10G-PON, 50G-PON, and VHSP systems.

[0151] This embodiment also provides a system for processing downlink bandwidth allocation information, which is used to implement the above-mentioned embodiments and preferred implementations. Details already described will not be repeated here. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0152] Figure 14 1 is a structural block diagram of a system for processing downlink bandwidth allocation information according to an embodiment of the present application. Figure 14 As shown, the system for processing downlink bandwidth allocation information includes a local device 1402;

[0153] The local device includes a master device and a slave device;

[0154] The master device and the slave device are used to jointly obtain device data collected by the slave device on the IoT device, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network;

[0155] The master device and the slave device are further configured to jointly perform local pre-processing on the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed;

[0156] The master device is further configured to obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information;

[0157] The slave device is further configured to control the slave device to perform data transmission in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction.

[0158] The embodiment of the present application obtains device data collected from the IoT device, obtains network bandwidth data of the optical fiber network where the local device is located, and obtains network traffic data of the optical fiber network; performs local preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed; obtains a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carries first downlink bandwidth allocation information, and controls the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction, thereby achieving efficient collection and processing of IoT device data based on local devices, and at the same time, dynamically adjusts downlink bandwidth allocation according to real-time bandwidth and traffic data of the optical fiber network, reducing the burden of data transmission, and outputs decision instructions in combination with the local analysis model, thereby improving the accuracy and real-time performance of decisions, solving the problem of low business efficiency caused by the inability to flexibly adjust bandwidth during bandwidth allocation, improving the efficiency and stability of data transmission, and providing a strong guarantee for the efficient operation of IoT devices.

[0159] This system architecture achieves efficient data collection and preprocessing of IoT device data through the collaborative work of master and slave devices. It also dynamically adjusts downlink bandwidth allocation based on the real-time bandwidth and traffic data of the fiber optic network, improving the efficiency and stability of data transmission. Local preprocessing and analysis can quickly respond to network changes and reduce data transmission delays, while slave devices transmit data based on decision instructions, ensuring the rational use of network resources. This embodiment is particularly suitable for scenarios with dense deployment of IoT devices and can effectively resolve bottlenecks in the data collection and transmission process, including but not limited to smart homes, industrial automation, and smart cities, ensuring efficient system operation and rational resource allocation.

[0160] In an exemplary embodiment, the system is further configured to collect data from the IoT device through the master device to obtain device data; or, to collect data from the IoT device through the master device and at least two of the slave devices to obtain device data.

[0161] In an exemplary embodiment, the system is further configured to collect bandwidth configuration data of the optical fiber network through the master device; and / or collect bandwidth jitter data of the optical fiber network through at least two of the slave devices.

[0162] In an exemplary embodiment, the local device is also used to perform local preprocessing on the device data, the network bandwidth data and the network traffic data according to a preprocessing step, and determine the first data to be analyzed based on the processing result of the local preprocessing; the preprocessing step at least includes: performing data cleaning on the device data, the network bandwidth data and the network traffic data to obtain cleaned data; performing data format conversion on the cleaned data to obtain converted data; performing data standardization on the converted data to obtain standard data; performing data feature extraction on the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; and / or performing data anomaly detection on the standard data to obtain abnormal data and normal data, and data storing the normal data.

[0163] In an exemplary embodiment, the system is further used to send the first decision instruction to the slave device through the master device, so that the slave device parses the first downlink bandwidth allocation information to obtain an allocation indicator corresponding to the device number of the slave device, wherein the allocation indicator corresponds to a sending time slot, and uplink physical data is sent within the time slot window of the sending time slot.

[0164] In an exemplary embodiment, a dynamic link library management channel for transmitting data is established between the master device and the slave device, and the dynamic link library management channel includes sub-management channels of different channel types, which are used to encapsulate data according to a data format corresponding to a predetermined channel type, and transmit the encapsulated data through the sub-management channel of the predetermined channel type.

[0165] In an exemplary embodiment, the sub-management channels of different channel types include a wavelength division multiplexing management communication channel (WMCC) management channel, a fiber management communication channel (FMCC) management channel, a forward path linear organization access method (F-PLOAM) management channel, and an embedded operation, administration, and maintenance (OAM) management channel. Before encapsulating data in a data format corresponding to a predetermined channel type and transmitting the encapsulated data through the sub-management channel of the predetermined channel type, the system is further configured to determine the WMCC management channel, FMCC management channel, and F-PLOAM management channel as sub-management channels of the predetermined channel type.

[0166] Figure 15 1 is a structural block diagram of a system for processing downlink bandwidth allocation information according to an embodiment of the present application. Figure 15 As shown, the system for processing downlink bandwidth allocation information includes:

[0167] A data acquisition module 1501 is provided in the local device and is connected to the monitoring and feedback module 1505 to collect device data from the IoT device, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network from the monitoring and feedback module;

[0168] a data preprocessing module 1502 connected to the data acquisition module, configured to perform local preprocessing on the device data, the network bandwidth data, and the network traffic data from the data acquisition module to obtain first data to be analyzed;

[0169] a model training and reasoning module 1503 located in the local device, wherein a local analysis model is pre-installed in the model training and reasoning module, and the model training and reasoning module is connected to the data preprocessing module, and is configured to receive a first decision instruction output by the local analysis model based on the first data to be analyzed, wherein the first decision instruction carries first downlink bandwidth allocation information;

[0170] The bandwidth allocation module 1504 is connected to the model training and reasoning module, and is used to control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction from the model training and reasoning module.

[0171] Through the modular architecture of the above-mentioned system, the separation of data collection, preprocessing, analysis and decision-making is achieved, and the scalability and maintainability of the system are improved. Among them, the data collection module is responsible for collecting data from IoT devices, the monitoring feedback module monitors the real-time status of the network, the data preprocessing module cleans, converts and standardizes the collected data, and the model training and inference module analyzes the preprocessed data to generate a bandwidth allocation strategy. Finally, the bandwidth allocation module controls the data transmission from the device according to the strategy. This embodiment can effectively solve the problem of low business efficiency caused by the inability to flexibly adjust the bandwidth during the bandwidth allocation process, and improves the efficiency and stability of data transmission through intelligent bandwidth allocation. In addition, this modular design includes but is not limited to the use of a microservice architecture, which can flexibly add or subtract modules according to actual needs, thereby improving the adaptability and flexibility of the system.

[0172] Figure 16 1 is a structural block diagram of a system for processing downlink bandwidth allocation information according to an embodiment of the present application. Figure 16 As shown, the system for processing downlink bandwidth allocation information includes:

[0173] Cloud platform 1602 and local device 1604;

[0174] The cloud platform is configured to obtain bandwidth allocation data from the local device; perform cloud preprocessing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed; obtain a second decision instruction output by a cloud analysis model based on the second data to be analyzed, the second decision instruction carrying second downlink bandwidth allocation information, and send the second decision instruction to the local device;

[0175] The local device is used to collect data from IoT devices, obtain device data, obtain network bandwidth data of the optical fiber network where the local device is located and network traffic data of the optical fiber network, determine the bandwidth allocation data based on the device data, the network bandwidth data and the network traffic data, obtain first data to be analyzed obtained after local preprocessing of the bandwidth allocation data, and obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information, and control the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the second decision instruction and the first decision instruction.

[0176] Through the above-mentioned embodiments, efficient data processing and intelligent decision-making can be achieved through the collaborative work of local devices and cloud platforms. Local devices are responsible for the initial collection and preprocessing of data, while the cloud platform conducts in-depth analysis and generates an optimized bandwidth allocation strategy. By combining local and cloud decision-making instructions, a more accurate bandwidth allocation plan is formulated, which solves the problem of inefficient business operations caused by the inability to flexibly adjust bandwidth during bandwidth allocation, thereby ensuring the efficiency and stability of data transmission. In addition, this collaborative working mode includes but is not limited to the use of a hybrid cloud architecture, which can flexibly allocate computing resources between local and cloud according to actual needs, thereby improving the overall performance and reliability of the system.

[0177] The master device is connected to at least two slave devices. The system is also used to determine bandwidth allocation data based on device data collected from the Internet of Things device, network bandwidth data of the optical fiber network where the local device is located, and network traffic data of the optical fiber network, and the Internet of Things device is connected to the slave device; obtain first data to be analyzed obtained after local preprocessing of the bandwidth allocation data, and obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information; and send the bandwidth allocation data to a cloud platform, and obtain a second decision instruction generated by the cloud platform based on the bandwidth allocation data; and control the slave device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction and the second decision instruction.

[0178] In an exemplary embodiment, before sending the bandwidth allocation data to the cloud platform and obtaining the second decision instruction generated by the cloud platform based on the bandwidth allocation data, the system is also used to establish a data transmission channel for transmitting data between the main device of the local device and the cloud platform; send the bandwidth allocation data to the cloud platform through the data transmission channel, and obtain the second decision instruction generated by the cloud platform based on the bandwidth allocation data through the data transmission channel.

[0179] In an exemplary embodiment, the cloud platform is also used to obtain bandwidth allocation data from a local device, the bandwidth allocation data at least including device data collected by the local device from the IoT device, network bandwidth data of the optical fiber network where the local device is located, and network traffic data of the optical fiber network; perform cloud pre-processing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed; obtain a second decision instruction output by a cloud analysis model based on the second data to be analyzed, the second decision instruction carries second downlink bandwidth allocation information, and sends the second decision instruction to the local device, and the local device is used to control the slave device in the local device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the second decision instruction and the first decision instruction.

[0180] In an exemplary embodiment, the cloud platform is further configured to receive the bandwidth allocation data via a data transmission channel, where the data transmission channel represents a channel for transmitting data between a master device of the local device and the cloud platform.

[0181] In an exemplary embodiment, the cloud platform is also used to perform cloud-based preprocessing on the device data, the network bandwidth data, and the network traffic data according to a preprocessing step, and determine the second data to be analyzed based on the processing result of the cloud-based preprocessing; the preprocessing step at least includes: performing data cleaning on the device data, the network bandwidth data, and the network traffic data to obtain cleaned data; performing data format conversion on the cleaned data to obtain converted data; performing data standardization on the converted data to obtain standard data; performing data feature extraction on the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; and / or performing data anomaly detection on the standard data to obtain abnormal data and normal data, and data storing the normal data.

[0182] Figure 17 1 is a structural block diagram of a system for processing downlink bandwidth allocation information according to an embodiment of the present application. Figure 17 As shown, the system for processing downlink bandwidth allocation information includes:

[0183] A data acquisition module 1701 provided in the local device, connected to the monitoring and feedback module 1705, is used to collect device data from the IoT device, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network from the monitoring and feedback module;

[0184] The data preprocessing module 1702 includes a local data preprocessing unit located on the local device and a cloud platform data preprocessing unit located on the cloud platform. The local data preprocessing unit is connected to the data acquisition module and is used to perform local preprocessing on the device data, the network bandwidth data, and the network flow data from the data acquisition module to obtain first data to be analyzed. The cloud platform data preprocessing unit is connected to the data acquisition module and is used to perform cloud-based preprocessing on the device data, the network bandwidth data, and the network flow data from the data acquisition module to obtain second data to be analyzed.

[0185] The model training and inference module 1703 includes a local analysis model located on the local device and a cloud analysis model located on the cloud platform. The local analysis model is connected to the data preprocessing module and is used to receive the first data to be analyzed and output a first decision instruction based on the first data to be analyzed, wherein the first decision instruction carries first downlink bandwidth allocation information. The cloud analysis model is connected to the data preprocessing module and is used to receive the second data to be analyzed and output a second decision instruction based on the second data to be analyzed, wherein the second decision instruction carries second downlink bandwidth allocation information.

[0186] The bandwidth allocation module 1704 is connected to the model training and reasoning module, and is used to control the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information and the second downlink bandwidth allocation information according to the first decision instruction and the second decision instruction from the model training and reasoning module.

[0187] In this embodiment, the master device determines a preliminary bandwidth allocation plan through a comprehensive analysis of IoT device data, network bandwidth data, and network traffic data. The master device then sends this data to the cloud platform, which leverages its powerful computing power for in-depth analysis, generating a more optimized bandwidth allocation strategy and assisting local devices in allocating bandwidth resources more efficiently. By combining local and cloud analysis results, more precise bandwidth allocation decision instructions are formulated, addressing the inefficient bandwidth allocation process inherent in related technologies, thereby improving the efficiency and stability of data transmission.

[0188] like Figure 17As shown, the data acquisition module (either the MFU or the SFU) performs data collection. The data preprocessing module and the model training and inference module, both locally and in the cloud, enable collaborative decision-making through local and cloud-based analysis models. The bandwidth allocation module specifically allocates bandwidth through the MFU, with the SFU providing bandwidth response. The monitoring feedback module primarily collects network performance and user experience information after bandwidth changes and feeds it back to the data preprocessing module for processing.

[0189] Through the description of the above 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 the necessary general hardware platform. 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 the present application, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a readable storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0190] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0191] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0192] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0193] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0194] S1, obtaining device data collected from the IoT device, obtaining network bandwidth data of the optical fiber network where the local device is located, and obtaining network traffic data of the optical fiber network;

[0195] S2, locally preprocessing the device data, the network bandwidth data, and the network traffic data to obtain first data to be analyzed;

[0196] S3, obtaining a first decision instruction output by the local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information, and controlling the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction.

[0197] Alternatively, the processor may be further configured to execute the following steps through a computer program:

[0198] S1, determining bandwidth allocation data based on device data collected from an IoT device, network bandwidth data of a fiber optic network where the local device is located, and network traffic data of the fiber optic network, wherein the IoT device is connected to the slave device;

[0199] S2, obtaining first data to be analyzed obtained by locally preprocessing the bandwidth allocation data, and obtaining a first decision instruction output by a local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information;

[0200] S3, and sending the bandwidth allocation data to the cloud platform, obtaining the second decision instruction generated by the cloud platform based on the bandwidth allocation data; controlling the slave device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction and the second decision instruction.

[0201] Alternatively, the processor may be further configured to execute the following steps through a computer program:

[0202] S1, obtaining bandwidth allocation data from a local device, the bandwidth allocation data including at least device data collected by the local device from an IoT device, network bandwidth data of a fiber optic network where the local device is located, and network traffic data of the fiber optic network;

[0203] S2, performing cloud pre-processing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed;

[0204] S3, obtaining a second decision instruction output by the cloud analysis model based on the second data to be analyzed, where the second decision instruction carries second downlink bandwidth allocation information, and sending the second decision instruction to the local device, where the local device is used to control the slave device in the local device to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information according to the second decision instruction and the first decision instruction.

[0205] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0206] Optionally, in this embodiment, the electronic device may also be configured to execute steps S1, S2 and S3 via a computer program.

[0207] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0208] An embodiment of the present application further provides another computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above method embodiments are implemented.

[0209] An embodiment of the present application also provides a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any of the above method embodiments.

[0210] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0211] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices, they can be implemented using program code executable by the computing device, and thus, they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be performed in a different order than herein, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0212] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for processing downlink bandwidth allocation information, characterized in that: Applied to a master device, the master device is connected to at least two slave devices, and the master device and the slave devices are both local devices, including: Determining bandwidth allocation data based on device data collected from an IoT device, network bandwidth data of an optical fiber network where the local device is located, and network traffic data of the optical fiber network, wherein the IoT device is connected to the slave device; Acquire first data to be analyzed obtained by locally preprocessing the bandwidth allocation data, and acquire a first decision instruction output by a local analysis model based on the first data to be analyzed, where the first decision instruction carries first downlink bandwidth allocation information; and sending the bandwidth allocation data to a cloud platform to obtain a second decision instruction generated by the cloud platform based on the bandwidth allocation data; According to the first decision instruction and the second decision instruction, the slave device is controlled to transmit data in the sending time slot indicated by the first downlink bandwidth allocation information.

2. The method according to claim 1, characterized in that Before sending the bandwidth allocation data to the cloud platform and obtaining a second decision instruction generated by the cloud platform based on the bandwidth allocation data, the method further includes: Establishing a data transmission channel for transmitting data between the master device of the local device and the cloud platform; The bandwidth allocation data is sent to the cloud platform through the data transmission channel, and a second decision instruction generated by the cloud platform based on the bandwidth allocation data is obtained through the data transmission channel.

3. A method for processing downlink bandwidth allocation information, characterized in that: Applied to cloud platforms, including: Obtaining bandwidth allocation data from a local device, the bandwidth allocation data including at least device data collected by the local device from the IoT device, network bandwidth data of a fiber optic network where the local device is located, and network traffic data of the fiber optic network; Performing cloud pre-processing on the device data, the network bandwidth data, and the network traffic data to obtain second data to be analyzed; Obtaining a second decision instruction output by the cloud analysis model based on the second data to be analyzed, where the second decision instruction carries second downlink bandwidth allocation information, and sending the second decision instruction to the local device, where the local device is configured to control a slave device in the local device to perform data transmission in a transmission time slot indicated by the first downlink bandwidth allocation information according to the second decision instruction and the first decision instruction; Among them, the local device is used to locally preprocess the device data, the network bandwidth data and the network traffic data to obtain first data to be analyzed; obtain the first decision instruction output by the local analysis model based on the first data to be analyzed, and the first decision instruction carries the first downlink bandwidth allocation information.

4. The method according to claim 3, characterized in that Get bandwidth allocation data from local devices, including: The bandwidth allocation data is received via a data transmission channel, where the data transmission channel represents a channel for transmitting data between a master device of the local device and the cloud platform.

5. The method according to claim 3, characterized in that The device data, the network bandwidth data, and the network traffic data are pre-processed in the cloud to obtain second data to be analyzed, including: Performing cloud preprocessing on the device data, the network bandwidth data, and the network traffic data according to the preprocessing step, and determining second data to be analyzed based on the processing result of the cloud preprocessing; The pre-processing step at least comprises: Cleaning the device data, the network bandwidth data, and the network traffic data to obtain cleaned data; Performing data format conversion on the cleaned data to obtain converted data; performing data standardization processing on the converted data to obtain standard data; Extracting data features from the standard data to obtain a data feature vector, and determining the processing result based on the data feature vector; And / or, performing data anomaly detection on the standard data to obtain abnormal data and normal data, and storing the normal data.

6. A system for processing downlink bandwidth allocation information, characterized in that: include: Cloud platforms and local devices; The local device is used to collect data from IoT devices to obtain device data, obtain network bandwidth data of the optical fiber network where the local device is located and network traffic data of the optical fiber network, determine bandwidth allocation data based on the device data, the network bandwidth data, and the network traffic data, obtain first data to be analyzed obtained after local preprocessing of the bandwidth allocation data, and obtain a first decision instruction output by a local analysis model based on the first data to be analyzed, the first decision instruction carrying first downlink bandwidth allocation information, and control a slave device included in the local device to transmit data in a sending time slot indicated by the first downlink bandwidth allocation information based on a second decision instruction and the first decision instruction; The cloud platform is used to obtain bandwidth allocation data from the local device; The device data, the network bandwidth data, and the network traffic data are pre-processed in the cloud to obtain second data to be analyzed; a second decision instruction output by a cloud analysis model based on the second data to be analyzed is obtained, the second decision instruction carries second downlink bandwidth allocation information, and the second decision instruction is sent to the local device.

7. A system for processing downlink bandwidth allocation information, characterized in that: include: A data acquisition module provided in the local device, the data acquisition module being connected to the monitoring and feedback module, and configured to acquire device data from the IoT device, obtain network bandwidth data of the optical fiber network where the local device is located, and obtain network traffic data of the optical fiber network from the monitoring and feedback module; a data preprocessing module, comprising a local data preprocessing unit located on the local device and a cloud platform data preprocessing unit located on the cloud platform, the local data preprocessing unit being connected to the data acquisition module and configured to perform local preprocessing on the device data, the network bandwidth data, and the network traffic data from the data acquisition module to obtain first data to be analyzed; The cloud platform data preprocessing unit is connected to the data acquisition module and is used to perform cloud preprocessing on the device data, the network bandwidth data, and the network traffic data from the data acquisition module to obtain second data to be analyzed; a model training and inference module, comprising a local analysis model located on the local device and a cloud-based analysis model located on the cloud platform, wherein the local analysis model is connected to the data preprocessing module and is configured to receive the first data to be analyzed and output a first decision instruction based on the first data to be analyzed, wherein the first decision instruction carries first downlink bandwidth allocation information; The cloud analysis model is connected to the data preprocessing module, and is configured to receive the second data to be analyzed, and output a second decision instruction based on the second data to be analyzed, wherein the second decision instruction carries second downlink bandwidth allocation information; A bandwidth allocation module is connected to the model training and reasoning module, and is used to control the slave device in the local device to transmit data in the transmission time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction from the model training and reasoning module, or to control the slave device in the local device to transmit data in the transmission time slot indicated by the first downlink bandwidth allocation information according to the first decision instruction and the second decision instruction from the model training and reasoning module.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the method according to any one of claims 1 to 5 through the computer program.

10. A computer program product, characterized in that The method comprises a computer program, which implements the method according to any one of claims 1 to 5 when the computer program is executed by a processor.

Images