FTTR network resource planning method, system, device and medium

Abstract

The invention discloses an FTTR network resource planning method, system and device and a medium, and relates to the technical field of networks, and the method comprises the steps: determining a target terminal according to the current position information; determining a signal intensity sequence according to the current position information, and determining a target node according to the signal intensity sequence; obtaining the current network speed and behavior data of each user terminal, and determining a target network speed sequence according to the current network speed and the behavior data; setting a bandwidth constraint condition and a performance constraint condition, calling a network speed prediction model, and determining a planned bandwidth sequence and a planned performance sequence corresponding to the planned bandwidth sequence according to a gradient descent algorithm; and converting the Internet access node of the target terminal into the target node, and adjusting the bandwidth of each network node to the corresponding planned bandwidth according to the planned bandwidth sequence. The method and the device have the effect of improving the flexibility of the FTTR network.

Description

Technical Field

[0001] This application relates to the technical field of networks, and in particular to an FTTR network resource planning method, system, device and medium. Background Technology

[0002] FTTR (Fiber to the Room) and FTTR B (Building / Broadband) technologies are designed to meet the growing demand for high-speed internet connectivity in modern home and business environments. FTTR provides extremely low latency and high bandwidth network services by directly connecting fiber optic cables to each room, ensuring a superior experience for users in various applications such as high-definition video streaming, online gaming, and remote work. FTTR B, on the other hand, emphasizes the effectiveness of fiber optic cabling within buildings, particularly in multi-family residential and commercial buildings, to ensure comprehensive coverage and reliable broadband access. The combination of these two technologies provides a solid network foundation for the construction of smart homes and smart cities, driving digital transformation and the efficient use of resources.

[0003] Regarding network bandwidth resource planning, current FTTR networks often employ a fixed allocation method. This method cannot be dynamically adjusted according to users' actual needs, potentially leading to insufficient bandwidth for the optical modem (ONT) or idle bandwidth resources. Furthermore, the lack of real-time monitoring and traffic analysis capabilities makes it difficult for network administrators to effectively optimize bandwidth allocation, thus impacting user experience.

[0004] Therefore, in practice, FTTR networks suffer from problems such as inflexible bandwidth resource allocation and difficulty in network adjustment, resulting in relatively low flexibility and intelligence levels, making it difficult to fully adapt to users' ever-changing network needs. Summary of the Invention

[0005] To improve the flexibility of FTTR networks, this application provides an FTTR network resource planning method, system, device, and medium.

[0006] Firstly, this application provides an FTTR network resource planning method, which adopts the following technical solution: An FTTR network resource planning method, comprising: The current location information of each user terminal in the FTTR network is monitored at a preset frequency, and the target terminal is determined based on the current location information. The target terminal is a user terminal whose distance between the current location and the previous location exceeds a preset distance. The signal strength sequence is determined based on the current location information, and the target node is determined based on the signal strength sequence, wherein the signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem; The current network speed and behavior data of each user terminal are obtained, and a target network speed sequence is determined based on the current network speed and the behavior data, wherein the target network speed sequence includes a target network speed that is compatible with the behavior of each user terminal. Set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the planned performance sequence corresponding to the planned bandwidth sequence according to the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth of each network node. The target terminal's internet access node is converted into the target node, and the bandwidth of each network node is adjusted to the corresponding planned bandwidth according to the planned bandwidth sequence.

[0007] By adopting the above technical solution, the current location information of each user terminal in the FTTR network is first monitored at a preset frequency, and the target terminal is determined based on the current location information. The target terminal is a user terminal whose distance from the current location to the previous location exceeds a preset distance. Then, a signal strength sequence is determined based on the current location information, and the target node is determined based on the signal strength sequence. The signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a primary optical modem and a secondary optical modem. Finally, the current network speed and behavior data of each user terminal are acquired, and the target network speed sequence is determined based on the current network speed and behavior data. The process involves several steps. First, the target network speed sequence includes the target network speed adapted to the behavior of each user terminal. Then, bandwidth and performance constraints are set. A network speed prediction model is invoked, and a gradient descent algorithm is used to determine the planned bandwidth sequence and its corresponding planned performance sequence. This ensures that the difference between the predicted network speed and the target network speed for each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth for each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth for each network node. Finally, the target terminal's internet access node is converted to the target node, and the bandwidth of each network node is adjusted to the corresponding planned bandwidth according to the planned bandwidth sequence. This method allows for automatic switching of the user terminal's internet access node based on signal strength when the user moves, while simultaneously replanning the bandwidth of each network node. This maintains stable network speeds for each user terminal in the FTTR network, improving user experience and enhancing the flexibility and intelligence of the FTTR network, thus better meeting evolving network demands.

[0008] Optionally, the step of determining the signal strength sequence based on the current location information includes: The location information of each network node is obtained, and a distance sequence is determined based on the location information and the current location, wherein the distance sequence includes the distance between the target terminal and each network node; Based on the signal strength prediction model and according to the distance sequence, the signal strength sequence is determined.

[0009] By adopting the above technical solution, in order to determine the signal strength sequence, the location information of each network node is obtained, and the distance sequence is determined based on the location information and the current location. The distance sequence includes the distance between the target terminal and each network node. Then, based on the signal strength prediction model and the distance sequence, the signal strength sequence is determined.

[0010] Optionally, the step of determining the target network speed sequence based on the current network speed and the behavioral data includes: For each user terminal, the target behavior type and the cumulative number of times corresponding to the target behavior type are determined based on the behavior data. The target behavior type is matched in a preset table to obtain the target network speed required for the target behavior type. The preset table includes user behavior type data and network speed required data, and the preset table is used to represent the mapping relationship between the user behavior type data and the network speed required data. The user terminal's required network speed is calculated based on the required network speed and the cumulative number of times, and it is determined whether the current network speed is greater than the required network speed. If so, the current network speed is taken as the target network speed of the user terminal; otherwise, the required network speed is taken as the target network speed of the user terminal. A target network speed sequence is generated based on the target network speed of each user terminal.

[0011] By adopting the above technical solution, in order to determine the target network speed sequence, for each user terminal, the target behavior type and the cumulative number of times corresponding to the target behavior type are determined based on the behavior data. Then, the target behavior type is matched in a preset table to obtain the target network speed required for the target behavior type. The preset table includes user behavior type data and network speed required data, and is used to represent the mapping relationship between user behavior type data and network speed required data. Then, the network speed required for the user terminal is calculated based on the network speed required and the cumulative number of times. It is then determined whether the current network speed is greater than the network speed required. If the current network speed is greater than the network speed required, the current network speed is taken as the target network speed for the user terminal. If the current network speed is less than the network speed required, the network speed required is taken as the target network speed for the user terminal. Finally, a target network speed sequence is generated based on the target network speeds of each user terminal.

[0012] Optionally, the method for generating the network speed prediction model includes: Obtain an evaluation parameter set, wherein the evaluation parameter set includes first historical bandwidth data and first historical performance data corresponding to the first historical bandwidth data; Calculate the Pearson coefficients between the parameters in the evaluation parameter set, and evaluate the mutual exclusivity between the parameters in the parameter set based on the Pearson coefficients to determine the mutually exclusive parameter group; Perform a linear transformation on the mutual exclusion parameter set to obtain the corresponding merging parameters; Based on the decision tree algorithm, and according to the merging parameters and the evaluation parameter set, an initial decision tree model is obtained; The initial decision tree model is trained and tested based on the model training data to obtain the network speed prediction model.

[0013] By adopting the above technical solution, in order to generate a network speed prediction model, an evaluation parameter set is obtained. The evaluation parameter set includes first historical bandwidth data and first historical performance data corresponding to the first historical bandwidth data. Then, the Pearson coefficient between each parameter in the evaluation parameter set is calculated, and the mutual exclusivity between each parameter in the parameter set is evaluated based on the Pearson coefficient to determine the mutually exclusive parameter group. Then, the mutually exclusive parameter group is linearly transformed to obtain the corresponding merging parameters. Then, based on the decision tree algorithm, and according to the merging parameters and the evaluation parameter set, a model is constructed to obtain an initial decision tree model. Finally, the initial decision tree model is trained and tested according to the model training data to obtain the network speed prediction model.

[0014] Optionally, the step of training and testing the initial decision tree model based on the model training data to obtain the network speed prediction model includes: Acquire model training data, which includes second historical bandwidth data, second historical performance data corresponding to the second historical bandwidth data, and historical network speed data. Divide the model training data into a training set and a test set according to a preset ratio. The hyperparameters of the initial decision tree model are set according to the genetic algorithm, and the root mean square error (RMSE) and coefficient of determination (R²) are set. 2 As an evaluation indicator; The initial decision tree model is trained based on the training set to obtain a trained decision tree model; The trained decision tree model is tested according to the test set, and the error is judged according to the evaluation index to see if it is within the preset range. If so, the trained decision tree model is used as the network speed prediction model.

[0015] By adopting the above technical solution, in order to train and test the initial decision tree model, model training data is first obtained. The model training data includes second historical bandwidth data, second historical performance data corresponding to the second historical bandwidth data, and historical network speed data. The model training data is divided into training set and test set according to a preset ratio. Then, the hyperparameters of the initial decision tree model are set according to the genetic algorithm, and the root mean square error (RMSE) and coefficient of determination (R²) are used as evaluation indicators. Then, the initial decision tree model is trained according to the training set to obtain a trained decision tree model. Finally, the trained decision tree model is tested according to the test set, and the error is judged according to the evaluation indicators to see if they are all within the preset range. If so, the trained decision tree model is used as the network speed prediction model.

[0016] Optionally, after the step of testing the trained decision tree model according to the test set and determining whether the errors are all within a preset range according to the evaluation index, and if so, using the trained decision tree model as the network speed prediction model, the method further includes: Acquire target bandwidth data and target performance data, wherein the target performance data includes target packet loss rate, target throughput, target number of connections, and target latency; The target bandwidth data and the performance data are input into the network speed prediction model to obtain the corresponding predicted network speed.

[0017] By adopting the above technical solution, in order to predict network speed, target bandwidth data and target performance data are first obtained. The target performance data includes target packet loss rate, target throughput, target number of connections, and target latency. Then, the target bandwidth data and performance data are input into the network speed prediction model to obtain the corresponding predicted network speed.

[0018] Optionally, the step of determining the planned bandwidth sequence and the corresponding planned performance sequence based on the gradient descent algorithm, so that the network speed of each network node is closest to the target network speed, includes: Obtain the merging logic of the merging parameters, and define the search parameter space based on the merging logic, the bandwidth constraints, and the performance constraints; Search space mapping is performed based on the merging parameters and the evaluation parameter set, and parameter space distance is defined based on Euclidean distance; Set the initial search value and initial search direction, and set the initial search; Based on the iteration point and search direction Perform iterations, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient in Euclidean space; The network speed is predicted according to the network speed prediction model, and the prediction result is obtained. The termination condition is determined according to the prediction result. If the result is met, it means that the planned bandwidth of each network node and the planned performance corresponding to the planned bandwidth have been found. A planned bandwidth sequence is generated based on the planned bandwidth of each network node, and a planned performance sequence is generated based on the planned performance of each network node.

[0019] By adopting the above technical solution, in order to determine the planned bandwidth sequence and the corresponding planned performance sequence, the merging logic of the merging parameters is first obtained, and the search parameter space is defined according to the merging logic, bandwidth constraints, and performance constraints. Then, the search space is mapped according to the merging parameters and the evaluation parameter set, and the parameter space distance is defined according to Euclidean distance. Then, the initial search value and initial search direction are set, and the initial search is performed based on the iteration points. and search direction Perform iterations, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient is given in Euclidean space. Then, the network speed is predicted according to the network speed prediction model. The prediction result is obtained, and it is determined whether the termination condition is met based on the prediction result. If so, it means that the planned bandwidth and the planned performance corresponding to the planned bandwidth of each network node have been found. Finally, a planned bandwidth sequence is generated based on the planned bandwidth of each network node, and a planned performance sequence is generated based on the planned performance of each network node.

[0020] Secondly, this application also provides an FTTR network resource planning system, which adopts the following technical solution: An FTTR network resource planning system, comprising: The target terminal determination module is used to monitor the current location information of each user terminal in the FTTR network at a preset frequency, and determine the target terminal based on the current location information. The target terminal is a user terminal whose distance between the current location and the previous location exceeds a preset distance. The target node determination module is used to determine a signal strength sequence based on the current location information and to determine a target node based on the signal strength sequence, wherein the signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem; The target network speed sequence determination module is used to acquire the current network speed and behavior data of each user terminal, and determine the target network speed sequence based on the current network speed and the behavior data, wherein the target network speed sequence includes the target network speed that is adapted to the behavior of each user terminal; The planning module is used to set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the planned performance sequence corresponding to the planned bandwidth sequence according to the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth of each network node. The network adjustment module is used to convert the Internet access node of the target terminal into the target node, and adjust the bandwidth of each network node to the corresponding planned bandwidth according to the planned bandwidth sequence.

[0021] Thirdly, this application also provides a computer device, which adopts the following technical solution: A computer device includes a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the method described in the first aspect.

[0022] Fourthly, this application also provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the method described in the first aspect.

[0023] In summary, this application includes at least the following beneficial technical effects: First, it monitors the current location information of each user terminal in the FTTR network at a preset frequency, and determines the target terminal based on the current location information. The target terminal is a user terminal whose distance from the current location to the previous location exceeds a preset distance. Then, it determines the signal strength sequence based on the current location information and determines the target node based on the signal strength sequence. The signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a primary optical modem and a secondary optical modem. Finally, it acquires the current network speed and behavior data of each user terminal and determines the target node based on the current network speed and behavior data. The FTTR network uses a target network speed sequence, where the target network speed sequence includes the target network speed adapted to the behavior of each user terminal. Then, bandwidth and performance constraints are set, a network speed prediction model is invoked, and a planned bandwidth sequence and its corresponding planned performance sequence are determined using a gradient descent algorithm. This ensures that the difference between the predicted network speed and the target network speed for each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth for each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth for each network node. Finally, the target terminal's internet access node is converted to the target node, and the bandwidth of each network node is adjusted to the corresponding planned bandwidth according to the planned bandwidth sequence. Through this method, when a user moves, the user terminal's internet access node can be automatically switched based on signal strength, while simultaneously replanning the bandwidth of each network node. This maintains stable network speeds for each user terminal in the FTTR network, improving user experience and enhancing the flexibility and intelligence of the FTTR network, better meeting ever-changing network demands. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall process of an embodiment of this application.

[0025] Figure 2 This is a schematic diagram of the system structure of this application.

[0026] Figure 3 This is a structural block diagram of the computer device described in this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] This application discloses an FTTR network resource planning method.

[0029] Reference Figure 1A method for FTTR network resource planning, characterized by comprising: Step S11: Monitor the current location information of each user terminal in the FTTR network at a preset frequency, and determine the target terminal based on the current location information.

[0030] The target terminal is a user terminal whose distance from the location of this data collection to the location of the previous data collection exceeds a preset distance.

[0031] It should be noted that in step S11, the location information of all user terminals in the FTTR network can be monitored periodically (e.g., every minute, every 10 seconds) by setting a timer. GPS, Wi-Fi or other positioning technologies can be used for precise positioning. Then, the current location is compared with the last location and the distance is calculated. If the distance exceeds the preset distance (e.g., 5 meters or 10 meters), the user terminal is marked as the target terminal. The Euclidean distance between user locations can be calculated using a formula to ensure efficiency.

[0032] Step S12: Determine the signal strength sequence based on the current location information, and determine the target node based on the signal strength sequence.

[0033] The signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem.

[0034] It should be noted that in step S12, the signal strength from the target terminal to the main optical modem and each slave optical modem is calculated and recorded in real time. Data is collected using network management tools or SNMP protocol, and all signal strength data is organized into a sequence. The signal strength values ​​can be sorted from high to low, and the node with the strongest signal strength is selected as the target node to ensure optimal network connection quality.

[0035] Step S13: Obtain the current network speed and behavior data of each user terminal, and determine the target network speed sequence based on the current network speed and behavior data.

[0036] The target network speed sequence includes target network speeds that are adapted to the behavior of each user terminal.

[0037] It should be noted that in step S13, the system obtains the current actual network speed and network behavior data of each user terminal in real time (such as the specific application type of ongoing video conferencing, file downloading, or web browsing), and intelligently calculates and sets an idealized target network speed that best matches its current behavior for each terminal according to a set of preset strategies aimed at optimizing the overall experience; finally, the personalized target network speed values ​​of all terminals are combined into an ordered set, which constitutes the target network speed sequence on which subsequent bandwidth allocation is based.

[0038] Step S14: Set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the planned performance sequence corresponding to the planned bandwidth sequence according to the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within the preset range.

[0039] The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth of each network node.

[0040] It should be noted that in step S14, based on network requirements, user needs, and user network packages, a lower and upper limit for bandwidth (e.g., min2Mbps, max100Mbps) are set. Furthermore, based on the configuration of each network node and its inherent performance limitations, upper and lower limits for various performance indicators are set. Then, a network speed prediction model is built using machine learning algorithms (e.g., linear regression or deep learning), inputting historical network speeds and other relevant factors for training. Finally, a gradient descent algorithm is applied to optimize bandwidth allocation. Each iteration adjusts the planned bandwidth sequence to ensure that the adjusted network speed is as close as possible to the target network speed.

[0041] Step S15: Convert the target terminal's Internet access node into a target node, and adjust the bandwidth of each network node to the corresponding planned bandwidth according to the planned bandwidth sequence.

[0042] It should be noted that in step S15, the network connection of the target terminal is set to the target node identified in step S12 to ensure that the new network connection is effective. Based on the planned bandwidth sequence, the bandwidth of each network node is adjusted one by one through the network manager (such as network control protocol or SNMP) until it meets the planned bandwidth setting.

[0043] In the above implementation, the current location information of each user terminal in the FTTR network is first monitored at a preset frequency, and the target terminal is determined based on the current location information. The target terminal is a user terminal whose distance from the current location to the previous location exceeds a preset distance. Then, a signal strength sequence is determined based on the current location information, and a target node is determined based on the signal strength sequence. The signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem. Finally, the current network speed and behavior data of each user terminal are acquired, and a target network speed sequence is determined based on the current network speed and behavior data. The process involves several steps: First, the target network speed sequence includes target network speeds adapted to the behavior of each user terminal. Then, bandwidth and performance constraints are set, a network speed prediction model is invoked, and a gradient descent algorithm is used to determine the planned bandwidth sequence and its corresponding planned performance sequence. This ensures that the difference between the predicted and target network speeds for each network node remains within a preset range. The planned bandwidth sequence includes the planned bandwidth for each network node, and the planned performance sequence includes the planned performance corresponding to that bandwidth. Finally, the target terminal's internet access node is converted to the target node, and the bandwidth of each network node is adjusted to the corresponding planned bandwidth based on the planned bandwidth sequence. This method allows for automatic switching of the user terminal's internet access node based on signal strength during user movement, while simultaneously replanning the bandwidth of each network node. This maintains stable network speeds for each user terminal in the FTTR network, improving user experience and enhancing the flexibility and intelligence of the FTTR network to better meet evolving network demands.

[0044] As a further implementation of the method, the step of determining the signal strength sequence based on the current location information includes: Step S21: Obtain the location information of each network node, and determine the distance sequence based on the location information and the current location. The distance sequence includes the distance between the target terminal and each network node.

[0045] It should be noted that in step S21, GPS, Wi-Fi positioning and other technologies are used to obtain the precise location information of all network nodes (such as the main optical modem, the slave optical modem, etc.). Each node can be recorded through the network management system (NMS) or manually. The obtained location information is then stored in the database for subsequent query and calculation. The distance between the target terminal and each network node is calculated using the Euclidean distance formula.

[0046] Step S22: Based on the signal strength prediction model and according to the distance sequence, determine the signal strength sequence.

[0047] In the above implementation, in order to determine the signal strength sequence, the location information of each network node is obtained, and the distance sequence is determined based on the location information and the current location. The distance sequence includes the distance between the target terminal and each network node. Then, based on the signal strength prediction model and the distance sequence, the signal strength sequence is determined.

[0048] As a further implementation of the method, the step of determining the target network speed sequence based on the current network speed and behavioral data includes: Step S31: For each user terminal, determine the target behavior type and the cumulative number of times corresponding to the target behavior type based on the behavior data.

[0049] It should be noted that in step S31, the system analyzes user behavior data (such as internet access time, application types used, etc.) to determine the target behavior type (e.g., video viewing, web browsing, etc.) and its corresponding cumulative frequency for each user terminal. This typically involves data mining and pattern recognition techniques, which can identify the user's main usage scenarios and needs. Through this analysis, the system can quantify user behavior, laying the foundation for subsequent network demand analysis. Clearly defining the behavioral patterns of different users helps to more accurately assess their network needs.

[0050] Step S32: Match the target behavior type with the preset table to obtain the target network speed required for the target behavior type.

[0051] The preset table includes user behavior type data and network speed demand data. The preset table is used to represent the mapping relationship between user behavior type data and network speed demand data.

[0052] It should be noted that in step S32, the system uses a preset mapping table to match the identified target behavior types with the corresponding required network speeds. The preset table contains the relationship between behavior type data and required network speed data, which is usually obtained through historical data statistics and experience summaries. This matching helps to formulate personalized network speed strategies, thereby improving the user experience.

[0053] Step S33: Calculate the user terminal's required network speed based on the required network speed and the cumulative number of times, and determine whether the current network speed is greater than the required network speed. If so, use the current network speed as the user terminal's target network speed; otherwise, use the required network speed as the user terminal's target network speed.

[0054] Specifically, the user terminal's required network speed is calculated based on the required network speed and the cumulative number of attempts. It is then determined whether the current network speed is greater than the required network speed. If the current network speed is greater than the required network speed, the current network speed is used as the target network speed for the user terminal. If the current network speed is not greater than the required network speed, the required network speed is used as the target network speed for the user terminal.

[0055] It should be noted that in step S33, the system calculates the required network speed for all target behavior types and compares it with the current actual network speed. This process may involve mathematical calculations and conditional judgments to ensure that the user terminal can obtain the optimal network speed setting. This dynamic adjustment mechanism can ensure that the network speed always meets the user's needs during use.

[0056] Step S34: Generate a target network speed sequence based on the target network speed of each user terminal.

[0057] In the above implementation, in order to determine the target network speed sequence, for each user terminal, the target behavior type and the cumulative number of times corresponding to the target behavior type are determined based on the behavior data. Then, the target behavior type is matched in a preset table to obtain the target network speed required for the target behavior type. The preset table includes user behavior type data and network speed required data, and the preset table is used to represent the mapping relationship between user behavior type data and network speed required data. Then, the network speed required for the user terminal is calculated based on the network speed required and the cumulative number of times. It is then determined whether the current network speed is greater than the network speed required. If the current network speed is greater than the network speed required, the current network speed is taken as the target network speed for the user terminal. If the current network speed is less than the network speed required, the network speed required is taken as the target network speed for the user terminal. Finally, a target network speed sequence is generated based on the target network speeds of each user terminal.

[0058] As a further implementation of the method, the method for generating the network speed prediction model includes: Step S41: Obtain the evaluation parameter set, wherein the evaluation parameter set includes the first historical bandwidth data and the first historical performance data corresponding to the first historical bandwidth data.

[0059] Step S42: Calculate the Pearson coefficients between the parameters in the evaluation parameter set, and evaluate the mutual exclusivity between the parameters in the parameter set based on the Pearson coefficients to determine the mutually exclusive parameter group.

[0060] It should be noted that in step S42, a program (such as Python) is written to calculate the Pearson coefficients between all parameter pairs. Commonly used libraries include NumPy or Pandas. The results are saved to a matrix or data frame, and the parameters are determined to be mutually exclusive based on a set threshold.

[0061] Step S43: Perform a linear transformation on the mutual exclusion parameter group to obtain the corresponding merged parameters.

[0062] Step S44: Based on the decision tree algorithm, and according to the merged parameters and evaluation parameter set, the model is constructed to obtain the initial decision tree model.

[0063] Step S45: Train and test the initial decision tree model based on the model training data to obtain the network speed prediction model.

[0064] In the above implementation, in order to generate a network speed prediction model, an evaluation parameter set is obtained, which includes first historical bandwidth data and first historical performance data corresponding to the first historical bandwidth data. Then, the Pearson coefficient between each parameter in the evaluation parameter set is calculated, and the mutual exclusivity between each parameter in the evaluation parameter set is determined based on the Pearson coefficient. Then, a linear transformation is performed on the mutual exclusive parameter set to obtain the corresponding merging parameters. Then, based on the decision tree algorithm, and according to the merging parameters and the evaluation parameter set, a model is constructed to obtain an initial decision tree model. Finally, the initial decision tree model is trained and tested according to the model training data to obtain the network speed prediction model.

[0065] As a further implementation of the method, the step of training and testing the initial decision tree model based on the model training data to obtain the network speed prediction model includes: Step S51: Obtain model training data, which includes second historical bandwidth data, second historical performance data corresponding to the second historical bandwidth data, and historical network speed data. Divide the model training data into training set and test set according to a preset ratio.

[0066] Step S52: Set the hyperparameters of the initial decision tree model according to the genetic algorithm, and use the root mean square error (RMSE) and the coefficient of determination (R²) as evaluation indicators.

[0067] Step S53: Train the initial decision tree model based on the training set to obtain a trained decision tree model.

[0068] Step S54: Test the trained decision tree model according to the test set, and determine whether the errors are all within the preset range according to the evaluation index. If so, use the trained decision tree model as the network speed prediction model.

[0069] In the above implementation, in order to train and test the initial decision tree model, model training data is first acquired. The model training data includes second historical bandwidth data, second historical performance data corresponding to the second historical bandwidth data, and historical network speed data. The model training data is divided into a training set and a test set according to a preset ratio. Then, the hyperparameters of the initial decision tree model are set according to the genetic algorithm, and the root mean square error (RMSE) and coefficient of determination (R²) are used as evaluation indicators. The initial decision tree model is then trained according to the training set to obtain a trained decision tree model. Finally, the trained decision tree model is tested according to the test set, and the error is judged according to the evaluation indicators to see if they are all within the preset range. If so, the trained decision tree model is used as the network speed prediction model.

[0070] As a further implementation of the method, after testing the trained decision tree model against a test set and determining whether the errors are all within a preset range based on evaluation metrics, and if so, using the trained decision tree model as the network speed prediction model, the method further includes: Step S61: Obtain target bandwidth data and target performance data.

[0071] The target performance data includes target packet loss rate, target throughput, target number of connections, and target latency.

[0072] Step S62: Input the target bandwidth data and performance data into the network speed prediction model to obtain the corresponding predicted network speed.

[0073] In the above implementation, in order to predict network speed, target bandwidth data and target performance data are first obtained. The target performance data includes target packet loss rate, target throughput, target number of connections, and target latency. Then, the target bandwidth data and performance data are input into the network speed prediction model to obtain the corresponding predicted network speed.

[0074] As a further implementation of the method, the step of determining the planned bandwidth sequence and the corresponding planned performance sequence based on the gradient descent algorithm to make the network speed of each network node as close as possible to the target network speed includes: Step S71: Obtain the merging logic of the merging parameters, and define the search parameter space based on the merging logic, bandwidth constraints, and performance constraints.

[0075] It should be noted that in step S71, it is necessary to obtain merging parameters related to the bandwidth and performance of network nodes. For example, traffic distribution between nodes, bandwidth limitations, and performance evaluation criteria (such as latency, packet loss rate, etc.) may be considered. When defining the search parameter space, the merging logic can simplify the parameter space, reduce noise and risk, and improve the performance of the model and the accuracy of the optimization results.

[0076] Step S72: Perform search space mapping based on the merged parameters and the evaluation parameter set, and define the parameter space distance based on Euclidean distance.

[0077] Step S73: Set the initial search value and initial search direction, and set the initial search.

[0078] Step S74, based on the iteration point and search direction Perform iterations, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient is in Euclidean space.

[0079] Step S75: Perform network speed prediction based on the network speed prediction model, obtain the prediction result, and determine whether the termination condition is met based on the prediction result. If so, it means that the planned bandwidth and the planned performance corresponding to the planned bandwidth of each network node have been searched.

[0080] Step S76: Generate a planned bandwidth sequence based on the planned bandwidth of each network node, and generate a planned performance sequence based on the planned performance of each network node.

[0081] In the above implementation, to determine the planned bandwidth sequence and the corresponding planned performance sequence, the merging logic of the merging parameters is first obtained, and a search parameter space is defined based on the merging logic, bandwidth constraints, and performance constraints. Then, the search space is mapped based on the merging parameters and the evaluation parameter set, and the parameter space distance is defined based on Euclidean distance. Next, initial search values ​​and initial search directions are set, and the initial search is performed based on iteration points. and search direction Perform iterations, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient is given in Euclidean space. Then, the network speed is predicted according to the network speed prediction model. The prediction result is obtained, and it is determined whether the termination condition is met based on the prediction result. If so, it means that the planned bandwidth and the planned performance corresponding to the planned bandwidth of each network node have been found. Finally, a planned bandwidth sequence is generated based on the planned bandwidth of each network node, and a planned performance sequence is generated based on the planned performance of each network node.

[0082] This application also discloses an FTTR network resource planning system.

[0083] refer to Figure 2 An FTTR network resource planning system, comprising: The target terminal determination module is used to monitor the current location information of each user terminal in the FTTR network at a preset frequency, and determine the target terminal based on the current location information. The target terminal is a user terminal whose distance between the current location and the previous location exceeds a preset distance. The target node determination module is used to determine the signal strength sequence based on the current location information and to determine the target node based on the signal strength sequence. The signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem. The target network speed sequence determination module is used to obtain the current network speed and behavior data of each user terminal, and determine the target network speed sequence based on the current network speed and behavior data. The target network speed sequence includes the target network speed that matches the behavior of each user terminal. The planning module is used to set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the corresponding planned performance sequence based on the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth of each network node. The network adjustment module is used to convert the target terminal's Internet access node into a target node, and adjust the bandwidth of each network node to the corresponding planned bandwidth according to the planned bandwidth sequence.

[0084] The FTTR network resource planning method of the present invention can implement any method in an FTTR network resource planning system, and the specific working process of the FTTR network resource planning method of the present invention can refer to the corresponding process in the above-mentioned FTTR network resource planning system.

[0085] This application also discloses a computer device.

[0086] refer to Figure 3 A computer device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement any of the above-described FTTR network resource planning methods.

[0087] This application also discloses a computer-readable storage medium.

[0088] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing any of the above-described FTTR network resource planning methods.

[0089] The computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0090] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for FTTR network resource planning, characterized in that, include: The current location information of each user terminal in the FTTR network is monitored at a preset frequency, and the target terminal is determined based on the current location information. The target terminal is a user terminal whose distance between the current location and the previous location exceeds a preset distance. The signal strength sequence is determined based on the current location information, and the target node is determined based on the signal strength sequence, wherein the signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem; The current network speed and behavior data of each user terminal are obtained, and a target network speed sequence is determined based on the current network speed and the behavior data, wherein the target network speed sequence includes a target network speed that is compatible with the behavior of each user terminal. Set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the planned performance sequence corresponding to the planned bandwidth sequence according to the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance corresponding to the planned bandwidth of each network node. The target terminal's internet access node is converted into the target node, and the bandwidth of each network node is adjusted to the corresponding planned bandwidth according to the planned bandwidth sequence.

2. The FTTR network resource planning method according to claim 1, characterized in that, The step of determining the signal strength sequence based on the current location information includes: The location information of each network node is obtained, and a distance sequence is determined based on the location information and the current location, wherein the distance sequence includes the distance between the target terminal and each network node; Based on the signal strength prediction model and according to the distance sequence, the signal strength sequence is determined.

3. The FTTR network resource planning method according to claim 1, characterized in that, The step of determining the target network speed sequence based on the current network speed and the behavioral data includes: For each user terminal, the target behavior type and the cumulative number of times corresponding to the target behavior type are determined based on the behavior data. The target behavior type is matched in a preset table to obtain the target network speed required for the target behavior type. The preset table includes user behavior type data and network speed required data, and the preset table is used to represent the mapping relationship between the user behavior type data and the network speed required data. The user terminal's required network speed is calculated based on the required network speed and the cumulative number of times, and it is determined whether the current network speed is greater than the required network speed. If so, the current network speed is taken as the target network speed of the user terminal; otherwise, the required network speed is taken as the target network speed of the user terminal. A target network speed sequence is generated based on the target network speed of each user terminal.

4. The FTTR network resource planning method according to claim 1, characterized in that, The method for generating the network speed prediction model includes: Obtain an evaluation parameter set, wherein the evaluation parameter set includes first historical bandwidth data and first historical performance data corresponding to the first historical bandwidth data; Calculate the Pearson coefficients between the parameters in the evaluation parameter set, and evaluate the mutual exclusivity between the parameters in the parameter set based on the Pearson coefficients to determine the mutually exclusive parameter group; Perform a linear transformation on the mutual exclusion parameter set to obtain the corresponding merging parameters; Based on the decision tree algorithm, and according to the merging parameters and the evaluation parameter set, an initial decision tree model is obtained; The initial decision tree model is trained and tested based on the model training data to obtain the network speed prediction model.

5. The FTTR network resource planning method according to claim 4, characterized in that, The step of training and testing the initial decision tree model based on the model training data to obtain the network speed prediction model includes: Acquire model training data, which includes second historical bandwidth data, second historical performance data corresponding to the second historical bandwidth data, and historical network speed data. Divide the model training data into a training set and a test set according to a preset ratio. The hyperparameters of the initial decision tree model are set according to the genetic algorithm, and the root mean square error (RMSE) and coefficient of determination (R²) are set. 2 As an evaluation indicator; The initial decision tree model is trained based on the training set to obtain a trained decision tree model; The trained decision tree model is tested according to the test set, and the error is judged according to the evaluation index to see if it is within the preset range. If so, the trained decision tree model is used as the network speed prediction model.

6. The FTTR network resource planning method according to claim 5, characterized in that, After the steps of testing the trained decision tree model according to the test set and determining whether the errors are all within a preset range according to the evaluation index, and if so, using the trained decision tree model as the network speed prediction model, the method further includes: Acquire target bandwidth data and target performance data, wherein the target performance data includes target packet loss rate, target throughput, target number of connections, and target latency; The target bandwidth data and the performance data are input into the network speed prediction model to obtain the corresponding predicted network speed.

7. The FTTR network resource planning method according to claim 4, characterized in that, The step of determining the planned bandwidth sequence and the corresponding planned performance sequence based on the gradient descent algorithm, so that the network speed of each network node is closest to the target network speed, includes: Obtain the merging logic of the merging parameters, and define the search parameter space based on the merging logic, the bandwidth constraints, and the performance constraints; Search space mapping is performed based on the merging parameters and the evaluation parameter set, and parameter space distance is defined based on Euclidean distance; Set the initial search value and initial search direction, and set the initial search; Based on the iteration point and search direction Perform iterations, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient in Euclidean space; The network speed is predicted according to the network speed prediction model, and the prediction result is obtained. The termination condition is determined according to the prediction result. If the result is met, it means that the planned bandwidth of each network node and the planned performance corresponding to the planned bandwidth have been found. A planned bandwidth sequence is generated based on the planned bandwidth of each network node, and a planned performance sequence is generated based on the planned performance of each network node.

8. An FTTR network resource planning system, characterized in that, include: The target terminal determination module is used to monitor the current location information of each user terminal in the FTTR network at a preset frequency, and determine the target terminal based on the current location information. The target terminal is a user terminal whose distance between the current location and the previous location exceeds a preset distance. The target node determination module is used to determine a signal strength sequence based on the current location information and to determine a target node based on the signal strength sequence, wherein the signal strength sequence includes the signal strength from the target terminal to each network node, and each network node includes at least a master optical modem and a slave optical modem; The target network speed sequence determination module is used to acquire the current network speed and behavior data of each user terminal, and determine the target network speed sequence based on the current network speed and the behavior data, wherein the target network speed sequence includes the target network speed that is adapted to the behavior of each user terminal; The planning module is used to set bandwidth constraints and performance constraints, call the network speed prediction model, and determine the planned bandwidth sequence and the planned performance sequence corresponding to the planned bandwidth sequence according to the gradient descent algorithm, so that the difference between the predicted network speed and the target network speed of each network node is within a preset range. The planned bandwidth sequence includes the planned bandwidth of each network node, and the planned performance sequence includes the planned performance of each network node corresponding to the planned bandwidth. The network adjustment module is used to convert the Internet access node of the target terminal into the target node, and adjust the bandwidth of each network node to the corresponding planned bandwidth according to the planned bandwidth sequence.

9. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and execute the method of any one of claims 1 to 7.

Images