Bandwidth allocation method and device for passive optical network, and storage medium

Abstract

The invention provides a bandwidth allocation method and device of a passive optical network and a storage medium, and relates to the field of optical communication. The method comprises the following steps: determining a traffic prediction model corresponding to the equipment type of each ONU based on the equipment type of each ONU in a plurality of optical network units; traffic prediction is carried out based on the historical traffic of each ONU and a traffic prediction model corresponding to the equipment type, and the predicted traffic of each ONU is obtained; determining a bandwidth allocation proportion of each ONU in the plurality of ONUs based on the predicted traffic of the plurality of ONUs, the real-time traffic of the plurality of ONUs and the service priorities of the plurality of ONUs; and allocating the bandwidth to the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation proportion of each ONU in the plurality of ONUs. According to the method and the device, differences of different ONUs in multiple aspects of equipment types and service priorities can be considered, the accuracy of bandwidth allocation is improved, the network delay is effectively reduced, and the resource utilization rate is improved.

Description

Technical Field

[0001] This application relates to the field of optical communication, and in particular to a bandwidth allocation method, apparatus and storage medium for a passive optical network. Background Technology

[0002] In industrial passive optical networks (PONs), the rational allocation of bandwidth resources is a key factor in ensuring uplink transmission performance and quality of service.

[0003] Traditional dynamic bandwidth allocation schemes typically use dynamic bandwidth allocation (DBA) technology to allocate bandwidth. However, this method does not allocate bandwidth accurately. Summary of the Invention

[0004] This application provides a bandwidth allocation method, apparatus, and storage medium for a passive optical network, which can improve the accuracy of bandwidth allocation.

[0005] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application provides a bandwidth allocation method for a passive optical network (ONN). The method includes: determining a traffic prediction model corresponding to the device type of each ONU based on the device type of each ONU among multiple optical network units (ONUs); performing traffic prediction based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type to obtain the predicted traffic of each ONU; determining the bandwidth allocation ratio of each ONU among multiple ONUs based on the predicted traffic of multiple ONUs, the real-time traffic of multiple ONUs, and the service priority of multiple ONUs; and allocating bandwidth to the corresponding ONUs based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among multiple ONUs, wherein the guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

[0006] In conjunction with the first aspect mentioned above, in one possible implementation, the bandwidth allocation ratio of each ONU among multiple ONUs is determined based on the predicted traffic, real-time traffic, and service priorities of multiple ONUs. This includes: for each ONU, determining the required bandwidth and differentiation coefficient of the ONU based on its predicted and real-time traffic, whereby the differentiation coefficient characterizes the degree of difference between the real-time and predicted traffic of the ONU; determining the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the ONU's service priority; and determining the bandwidth allocation ratio of the ONU based on the product of the ONU's required bandwidth, the ONU's differentiation coefficient, and the ONU's bandwidth allocation priority weight.

[0007] In conjunction with the first aspect above, in one possible implementation, the required bandwidth and differentiation coefficient of the ONU are determined based on the predicted traffic and real-time traffic of the ONU, including: determining the maximum value between the predicted traffic and real-time traffic of the ONU as the required bandwidth of the ONU; determining the ratio of the real-time traffic to the predicted traffic of the ONU as a first ratio; if the first ratio is less than a first ratio threshold, determining the first ratio threshold as the differentiation coefficient of the ONU; if the first ratio is greater than a second ratio threshold, determining the second ratio threshold as the differentiation coefficient of the ONU; and if the first ratio is greater than or equal to the first ratio threshold and less than or equal to the second ratio threshold, determining the first ratio as the differentiation coefficient of the ONU.

[0008] In conjunction with the first aspect mentioned above, in one possible implementation, the bandwidth allocation priority weight of the ONU is determined based on the guaranteed bandwidth and the service priority of the ONU, including: determining the total guaranteed bandwidth of multiple ONUs, and determining the difference between the total network bandwidth and the total guaranteed bandwidth as the first difference; determining the difference between the first difference and the reserved bandwidth as the bandwidth remaining amount; and determining the bandwidth allocation priority weight of the ONU corresponding to the service priority based on the ratio of the bandwidth remaining amount to the total network bandwidth.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the bandwidth allocation priority weights satisfy the following formula:

[0010] In the formula, w represents the bandwidth allocation priority weight. C represents the remaining bandwidth, and C represents the total network bandwidth.

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, the device type includes a first type, a second type, and a third type; based on the device type of each ONU in the multiple optical network units (ONUs), the traffic prediction model corresponding to the device type of each ONU is determined, including: when the device type of the ONU is the first type, determining the traffic prediction model corresponding to the device type as a neural network model combining a first long short-term memory network and an attention mechanism; when the device type of the ONU is the second type, determining the traffic prediction model corresponding to the device type as a neural network model combining a second long short-term memory network and an attention mechanism; when the device type of the ONU is the third type, determining the traffic prediction model corresponding to the device type as a neural network model combining a third long short-term memory network and an attention mechanism.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, bandwidth allocation is performed for the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among multiple ONUs. This includes: determining the total guaranteed bandwidth of the multiple ONUs; determining the difference between the total network bandwidth and the total guaranteed bandwidth as a first difference; determining the difference between the first difference and the reserved bandwidth as the bandwidth surplus; determining the bandwidth allocation amount for each ONU by multiplying the bandwidth surplus by the bandwidth allocation ratio and summing it with the guaranteed bandwidth; and allocating bandwidth to the corresponding ONU according to the bandwidth allocation amount for each ONU.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the bandwidth of each ONU satisfies the following formula:

[0014] In the formula, i represents the i-th ONU; This represents the bandwidth allocation for the i-th ONU; This indicates guaranteed bandwidth; This represents the bandwidth allocation ratio of the i-th ONU; j represents the j-th ONU; m represents the total number of ONUs; This indicates the amount of bandwidth remaining.

[0015] In conjunction with the first aspect mentioned above, in one possible implementation, the guaranteed bandwidth is determined as follows: the ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth is determined as the occupancy ratio; if the occupancy ratio is less than a first occupancy threshold, a first preset value is determined as the guaranteed bandwidth; if the occupancy ratio is greater than or equal to the first occupancy threshold and less than or equal to a second occupancy threshold, a second preset value is determined as the guaranteed bandwidth; if the occupancy ratio is greater than the second occupancy threshold, the guaranteed bandwidth corresponding to ONUs with a service priority of first priority is determined as a third preset value, and the guaranteed bandwidth corresponding to ONUs with a service priority other than first priority is determined as a second preset value; wherein, the first preset value is less than the second preset value, and the second preset value is less than the third preset value.

[0016] Secondly, this application provides a bandwidth allocation device for a passive optical network. The device includes: a first determining unit, configured to determine a traffic prediction model corresponding to the device type of each ONU based on the device type of each ONU among a plurality of optical network units (ONUs); a prediction unit, configured to perform traffic prediction based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, to obtain the predicted traffic of each ONU; a second determining unit, configured to determine the bandwidth allocation ratio of each ONU among the plurality of ONUs based on the predicted traffic of the plurality of ONUs, the real-time traffic of the plurality of ONUs, and the service priority of the plurality of ONUs; and an allocation unit, configured to allocate bandwidth to the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among the plurality of ONUs, wherein the guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

[0017] In conjunction with the second aspect above, in one possible implementation, the second determining unit is used to: for each ONU, determine the required bandwidth and differentiation coefficient of the ONU based on the predicted traffic and real-time traffic of the ONU, wherein the differentiation coefficient is used to characterize the degree of differentiation between the real-time traffic and the predicted traffic of the ONU; determine the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the service priority of the ONU; and determine the bandwidth allocation ratio of the ONU based on the product of the required bandwidth of the ONU, the differentiation coefficient of the ONU, and the bandwidth allocation priority weight of the ONU.

[0018] In conjunction with the second aspect above, in one possible implementation, the second determining unit is configured to: determine the maximum value between the predicted traffic and the real-time traffic of the ONU as the required bandwidth of the ONU; determine the ratio of the real-time traffic to the predicted traffic of the ONU as a first ratio; if the first ratio is less than a first ratio threshold, determine the first ratio threshold as the differentiation coefficient of the ONU; if the first ratio is greater than a second ratio threshold, determine the second ratio threshold as the differentiation coefficient of the ONU; and if the first ratio is greater than or equal to the first ratio threshold and less than or equal to the second ratio threshold, determine the first ratio as the differentiation coefficient of the ONU.

[0019] In conjunction with the second aspect above, in one possible implementation, the second determining unit is used to: determine the total guaranteed bandwidth of multiple ONUs, and determine the difference between the total network bandwidth and the total guaranteed bandwidth as a first difference; determine the difference between the first difference and the reserved bandwidth as the bandwidth surplus; and determine the bandwidth allocation priority weight of the ONU corresponding to the service priority based on the ratio of the bandwidth surplus to the total network bandwidth.

[0020] In conjunction with the second aspect above, in one possible implementation, the bandwidth allocation priority weights satisfy the following formula:

[0021] In the formula, w represents the bandwidth allocation priority weight. C represents the remaining bandwidth, and C represents the total network bandwidth.

[0022] In conjunction with the second aspect above, in one possible implementation, the device type includes a first type, a second type, and a third type; the first determining unit is configured to: when the ONU's device type is the first type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a first long short-term memory network and an attention mechanism; when the ONU's device type is the second type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a second long short-term memory network and an attention mechanism; and when the ONU's device type is the third type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a third long short-term memory network and an attention mechanism.

[0023] In conjunction with the second aspect above, in one possible implementation, the allocation unit is used to: determine the total guaranteed bandwidth of multiple ONUs; determine the difference between the total network bandwidth and the total guaranteed bandwidth as a first difference; determine the difference between the first difference and the reserved bandwidth as the bandwidth surplus; determine the bandwidth allocation amount for each ONU by multiplying the bandwidth surplus by the bandwidth allocation ratio and the guaranteed bandwidth; and allocate bandwidth to the corresponding ONU according to the bandwidth allocation amount for each ONU.

[0024] In conjunction with the second aspect above, in one possible implementation, the bandwidth of each ONU satisfies the following formula:

[0025] In the formula, i represents the i-th ONU; This represents the bandwidth allocation for the i-th ONU; This indicates guaranteed bandwidth; This represents the bandwidth allocation ratio of the i-th ONU; j represents the j-th ONU; m represents the total number of ONUs; This indicates the amount of bandwidth remaining.

[0026] In conjunction with the second aspect above, in one possible implementation, the third determining unit is configured to: determine the ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth as the occupancy ratio; determine a first preset value as the guaranteed bandwidth when the occupancy ratio is less than a first occupancy threshold; determine a second preset value as the guaranteed bandwidth when the occupancy ratio is greater than or equal to the first occupancy threshold and less than or equal to a second occupancy threshold; and determine the guaranteed bandwidth corresponding to ONUs with a service priority of first priority as a third preset value and the guaranteed bandwidth corresponding to ONUs with a service priority other than first priority as a second preset value when the occupancy ratio is greater than the second preset value; wherein, the first preset value is less than the second preset value, and the second preset value is less than the third preset value.

[0027] Thirdly, this application provides an electronic device, including: a processor and a communication interface; the communication interface and the processor are coupled, and the processor is used to run computer programs or instructions to implement the bandwidth allocation method of the passive optical network as described in the first aspect and any possible implementation of the first aspect.

[0028] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform a bandwidth allocation method for a passive optical network as described in the first aspect and any possible implementation thereof.

[0029] Fifthly, this application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform a bandwidth allocation method for a passive optical network as described in the first aspect and any possible implementation thereof.

[0030] In a sixth aspect, this application provides a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run computer programs or instructions to implement the bandwidth allocation method for a passive optical network as described in the first aspect and any possible implementation thereof.

[0031] Specifically, the chip provided in this application also includes a memory for storing computer programs or instructions.

[0032] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the device, or it may be packaged separately from the processor of the device; this application does not impose any limitation on this.

[0033] In a seventh aspect, this application provides a bandwidth allocation system for a passive optical network, comprising: an OLT and an ONU, wherein the OLT is used to perform the bandwidth allocation method for a passive optical network as described in the first aspect and any possible implementation thereof.

[0034] The descriptions of aspects two through seven in this application can be referenced to the detailed description of aspect one; and the beneficial effects of the descriptions of aspects two through seven can be referenced to the analysis of the beneficial effects of aspect one, which will not be repeated here.

[0035] In this application, the names of the bandwidth allocation devices for the passive optical network do not limit the devices or functional modules themselves. In actual implementation, these devices or functional modules may appear under other names. As long as the functions of each device or functional module are similar to those in this application, they fall within the scope of the claims of this application and their equivalents.

[0036] These or other aspects of this application will become more readily apparent in the following description.

[0037] The bandwidth allocation method for passive optical networks (PONs) provided in this application addresses the different traffic characteristics corresponding to different ONU device types. By employing corresponding traffic prediction models based on the ONU device type, the accuracy of traffic prediction can be improved. By fully considering the real-time traffic, predicted traffic, and service priorities of each ONU, bandwidth is allocated according to the bandwidth allocation ratio and guaranteed bandwidth. This not only ensures the basic communication needs of the ONUs but also takes into account the differences in device type and service priority among different ONUs. It dynamically allocates bandwidth to ONUs with different device types and service priorities, improving the accuracy of bandwidth allocation, effectively reducing network latency, and increasing resource utilization. Attached Figure Description

[0038] Figure 1 A schematic diagram of the architecture of a bandwidth allocation system for a passive optical network provided in an embodiment of this application; Figure 2 A flowchart illustrating a bandwidth allocation method for a passive optical network provided in this application embodiment; Figure 3 A schematic diagram of the structure of a multi-branch LSTM-Attention model provided in an embodiment of this application; Figure 4 A flowchart illustrating another bandwidth allocation method for a passive optical network provided in this application embodiment; Figure 5 This is a schematic diagram of the real-time traffic of multiple ONUs provided in the embodiments of this application; Figure 6 A schematic diagram illustrating multiple predicted traffic and real-time traffic provided in the embodiments of this application; Figure 7 This application provides a schematic diagram comparing bandwidth allocation performance in an embodiment. Figure 8 A schematic diagram of the structure of a bandwidth allocation device for a passive optical network provided in an embodiment of this application; Figure 9 This is a schematic diagram of the hardware structure of a bandwidth allocation device for a passive optical network provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0041] The terms "first" and "second," etc., used in the specification and drawings of this application are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.

[0042] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0043] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0044] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0045] In industrial passive optical networks (PONs), the rational allocation of bandwidth resources is a key factor in ensuring uplink transmission performance and quality of service.

[0046] In industrial passive optical networks (PONs), industrial equipment (e.g., robots, surveillance cameras, sensors) has stringent requirements for low latency and high reliability, necessitating the priority transmission of critical data. This critical data may include control commands and sensor feedback. Simultaneously, network traffic exhibits complex patterns, including periodicity (e.g., production line cycle time), burstiness (e.g., fault alarms), and multi-device collaboration (e.g., distributed control systems). The network environment is also dynamically changing, requiring real-time responses to events such as device additions / removals, link failures, and load fluctuations.

[0047] Industrial PON systems employ a point-to-multipoint (P2MP) topology. In this architecture, multiple optical network units (ONUs) are prone to data collisions during uplink transmission, leading to packet loss and excessive latency. Industrial scenarios demand extremely high real-time performance, determinism, and reliability in data transmission. Traditional dynamic bandwidth allocation (DBA) technology, due to its inherent limitations, struggles to meet the scheduling requirements of industrial environments.

[0048] Traditional dynamic bandwidth allocation schemes rely on the Interleaved Polling with Adaptive Cycle Time (IPACT) protocol to allocate bandwidth through a polling process. All ONUs send the same data frames to the optical line terminal (OLT) on demand, requesting bandwidth. This results in a large number of Multi-Point Control Protocol (MPCP) data frames consuming bandwidth within the network. Furthermore, the IPACT-based bandwidth allocation scheme cannot utilize the periodic traffic characteristics of industrial PON networks, failing to pre-allocate bandwidth resources in a targeted manner. This leads to significant bandwidth waste, increased network latency, and negatively impacts industrial production.

[0049] Existing solutions primarily improve the architecture and bandwidth allocation methods of traditional time-division multiplexing passive optical networks (PONs) to provide differentiated allocation schemes for different types of user traffic, thereby increasing the bandwidth resource utilization of the entire optical access network. In this solution, by dividing the network into two logically isolated virtual OLTs and dividing the ONUs connected to the OLTs into two slices, the flexibility and scalability of the network are effectively improved. By collecting historical traffic data and inputting it into a joint neural network model for accurate traffic classification, differentiated resource allocation is achieved. Furthermore, by classifying traffic and allocating user data traffic to specific ONU slices, and employing different bandwidth allocation methods in different slices, the overall bandwidth resource utilization of the network is improved.

[0050] However, this approach also has some drawbacks.

[0051] 1. When using a combined neural network model of long short-term memory (LSTM) and gated recurrent unit (GRU) to predict the flow of the next cycle, the impact of burst flow in industrial PON is not fully considered.

[0052] Second, predicting network traffic based on the same LSTM and GRU joint neural network model does not take into account the service quality requirements of different devices.

[0053] Third, when allocating bandwidth based on the prediction results, the priority of industrial equipment is not considered in the periodic traffic slicing. This will result in the bandwidth demand of high-priority equipment not being fully guaranteed during peak traffic periods.

[0054] Therefore, DBA technology for industrial PON needs to have more refined network traffic monitoring capabilities, more efficient scheduling algorithms, and more flexible priority management mechanisms, so as to achieve on-demand bandwidth allocation and real-time guarantee in industrial environments.

[0055] The bandwidth allocation method for passive optical networks (PONs) provided in this application addresses the different traffic characteristics corresponding to different ONU device types. By employing corresponding traffic prediction models based on the ONU device type, the accuracy of traffic prediction can be improved. By fully considering the real-time traffic, predicted traffic, and service priorities of each ONU, bandwidth is allocated according to the bandwidth allocation ratio and guaranteed bandwidth. This not only ensures the basic communication needs of the ONUs but also takes into account the differences in device type and service priority among different ONUs. It dynamically allocates bandwidth to ONUs with different device types and service priorities, improving the accuracy of bandwidth allocation, effectively reducing network latency, and increasing resource utilization.

[0056] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0057] The technical solution of this application is applicable to optical network systems, such as gigabit-capable passive optical networks (GPON), NGPON, 50G PON, future optical network systems, or integrated systems of multiple optical network systems.

[0058] Figure 1 This is a schematic diagram of the architecture of a bandwidth allocation system for a passive optical network provided in an embodiment of this application. Figure 1 As shown, the architecture includes: ONU 101 and OLT 102.

[0059] One end of the ONU 101 connects upwards to the OLT 102, and the other end connects downwards to terminal devices, such as computers or landline telephones. The ONU 101 can connect to various types of digital subscriber line (DSL) or Ethernet access gateway devices, which then connect to network terminals. When used in conjunction with the OLT 102, the ONU 101 enables Ethernet Layer 2 and Layer 3 functionality, providing users with voice, data, and multimedia services.

[0060] In some embodiments, ONU 101 can send data to OLT 102.

[0061] In some embodiments, ONU 101 can receive data sent by OLT 102 and respond to commands issued by OLT 102 to make corresponding adjustments.

[0062] One end of the OLT 102 connects to the upper-layer network to complete the access of uplink signals. The upper-layer network can be an Internet Protocol (IP) backbone or a Public Switched Telephone Network (PSTN). The other end of the OLT 102 connects to the ONU 101 to complete the transmission of downlink signals. The OLT 102 can realize functions such as control, management and ranging of the ONU 101.

[0063] The OLT 102 can be deployed in locations such as laboratories, residential communities, streets, and central control stations. One OLT 102 can connect to multiple ONU 101 units. Figure 1Only one OLT 102 connected to one ONU 101 is shown in the diagram. It should be understood that the bandwidth allocation system of this passive optical network may also include a greater number of ONUs 101, and this application does not limit this.

[0064] In some embodiments, the OLT 102 may provide communication functionality to enable communication with the ONU 101. For example, the OLT 102 may obtain the real-time traffic and service priorities of the ONU 101 through communication with the ONU 101.

[0065] In some embodiments, the OLT 102 may provide processing functions. For example, the OLT 102 determines the traffic prediction model corresponding to the device type of each ONU among multiple optical network units (ONUs). For example, the OLT 102 performs traffic prediction based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, obtaining the predicted traffic for each ONU. For example, the OLT 102 determines the bandwidth allocation ratio for each ONU among multiple ONUs based on the predicted traffic of multiple ONUs, the real-time traffic of multiple ONUs, and the service priorities of multiple ONUs. For example, the OLT 102 allocates bandwidth to the corresponding ONUs based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among multiple ONUs.

[0066] Figure 2 This is a flowchart illustrating a bandwidth allocation method for a passive optical network provided in an embodiment of this application. Figure 2 As shown, the bandwidth allocation method includes the following steps: S201. Based on the device type of each ONU in multiple optical network units (ONUs), determine the traffic prediction model corresponding to the device type of each ONU.

[0067] For example, the types of devices that ONUs may use include robot controller terminals, surveillance cameras, sensors, and so on.

[0068] For example, the device type of an ONU can be a type categorized according to the traffic characteristics of different ONU devices. For instance, the device types of an ONU include a first type, a second type, and a third type. The first type is characterized by periodic traffic, low latency, and a high-frequency traffic cycle. The second type is characterized by continuous high traffic volume and bursts of traffic. The third type is characterized by low traffic volume, periodic traffic, and a low-frequency traffic cycle.

[0069] For example, different device types have different traffic characteristics, such as large differences in traffic periodicity, burstiness, and latency. If a unified traffic prediction model is used for traffic prediction, there will be problems with inaccurate traffic prediction, which will lead to inaccurate bandwidth allocation accuracy in the future.

[0070] For example, the data sent by the ONU to the OLT includes a device type identifier. This device type identifier is used to indicate the device type of the ONU.

[0071] S202. Based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, traffic prediction is performed to obtain the predicted traffic of each ONU.

[0072] For example, the historical traffic of ONUs of the first device type is used as input data and input into a neural network model that combines a first long short-term memory network with an attention mechanism to predict the traffic, thereby obtaining the predicted traffic of the corresponding ONU.

[0073] For example, the historical traffic of ONUs of the second type is used as input data and input into a neural network model that combines a second long short-term memory network with an attention mechanism to predict the traffic, thereby obtaining the predicted traffic of the corresponding ONU.

[0074] For example, the historical traffic of ONUs of type 3 is used as input data and input into a neural network model that combines a third long short-term memory network with an attention mechanism to predict the traffic, thereby obtaining the predicted traffic of the corresponding ONU.

[0075] S203. Based on the predicted traffic of multiple ONUs, the real-time traffic of multiple ONUs, and the service priority of multiple ONUs, determine the bandwidth allocation ratio of each ONU among the multiple ONUs.

[0076] S204. Based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among multiple ONUs, allocate bandwidth to the corresponding ONU.

[0077] Among them, the guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

[0078] Without a guaranteed bandwidth allocation, during network congestion, some ONUs may monopolize bandwidth for extended periods, while others may not receive any allocated bandwidth, potentially leading to ONU "starvation." In this situation, the ONUs without allocated bandwidth cannot maintain normal service operation, resulting in service interruptions. To avoid this, a guaranteed bandwidth allocation is implemented, reserving a minimum amount of bandwidth for each ONU in advance. This ensures that even under network congestion, the ONU receives the minimum necessary bandwidth to maintain normal service operation.

[0079] The bandwidth allocation method for passive optical networks (PONs) provided in this application addresses the different traffic characteristics corresponding to different ONU device types. By employing corresponding traffic prediction models based on the ONU device type, the accuracy of traffic prediction can be improved. By fully considering the real-time traffic, predicted traffic, and service priorities of each ONU, bandwidth is allocated according to the bandwidth allocation ratio and guaranteed bandwidth. This not only ensures the basic communication needs of the ONUs but also takes into account the differences in device type and service priority among different ONUs. It dynamically allocates bandwidth to ONUs with different device types and service priorities, improving the accuracy of bandwidth allocation, effectively reducing network latency, and increasing resource utilization.

[0080] The following is about the above. Figure 2 The process of determining the bandwidth allocation ratio in step S203 will be explained. In one possible implementation, the process of determining the bandwidth allocation ratio in step S203 can be achieved through steps 11 to 13.

[0081] Step 11: For each ONU, determine the required bandwidth and differentiation coefficient of the ONU based on the predicted traffic and real-time traffic of the ONU.

[0082] The difference coefficient is used to characterize the degree of difference between the real-time traffic and the predicted traffic of the ONU.

[0083] Determining the bandwidth requirement of ONUs based on predicted and real-time traffic allows for more accurate estimation of ONU bandwidth allocation. It also corrects predicted traffic, optimizes resource allocation decisions, and improves resource utilization.

[0084] By determining the differentiation coefficient through predicted traffic and real-time traffic, the degree of difference between real-time traffic and predicted traffic can be intuitively displayed, making the subsequent determination of bandwidth allocation more accurate.

[0085] Step 12: Determine the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the service priority of the ONU.

[0086] For example, the data sent by the ONU to the OLT includes a service priority identifier. This service priority identifier is used to indicate the service priority of the ONU.

[0087] For example, the data sent by the ONU to the OLT includes a service type identifier. This service type identifier is used to identify the service. There is a mapping relationship between the service type identifier and the service priority, which allows the ONU's service priority to be determined.

[0088] For example, a higher service priority of an ONU indicates a greater importance and resource urgency for that ONU, signifying a higher value to the network for that user or the ONU, and thus requiring guaranteed bandwidth for that ONU. Determining bandwidth allocation priority weights based on the ONU's service priority ensures that subsequent bandwidth allocation to that ONU fully considers service priority, guarantees the QoS of that service, and prevents bandwidth from being monopolized by ONUs with lower service priority.

[0089] Step 13: Determine the bandwidth allocation ratio of the ONU based on the product of the ONU's required bandwidth, the ONU's differentiation coefficient, and the ONU's bandwidth allocation priority weight.

[0090] In this way, based on the bandwidth demand of the ONU, the differentiation coefficient, and the bandwidth allocation priority weight, the determined bandwidth allocation ratio can achieve precise and differentiated allocation of bandwidth, ensuring critical services while ensuring that each ONU obtains reasonable bandwidth, thereby improving resource utilization.

[0091] The process of determining the required bandwidth and the differentiation coefficient in step 11 above will be described below. As a possible embodiment of this application, the process of determining the required bandwidth and the differentiation coefficient in step 11 above can be implemented through steps 21 to 25.

[0092] Step 21: Determine the maximum value between the ONU's predicted traffic and real-time traffic as the ONU's required bandwidth.

[0093] For example, the bandwidth requirement of this ONU satisfies the following formula 1.

[0094] Formula 1 In the formula, i represents the i-th ONU; This represents the bandwidth requirement of the i-th ONU; This represents the predicted traffic for the i-th ONU; This represents the real-time traffic of the i-th ONU.

[0095] During use, there may be sudden traffic surges in the network. Determining the maximum value between the predicted traffic and the real-time traffic as the required bandwidth can not only avoid prediction errors in the predicted traffic, but also meet the user's sudden traffic surges, so that subsequent bandwidth allocation can meet the user's needs.

[0096] Step 22: Determine the ratio of the real-time traffic to the predicted traffic of the ONU as the first ratio.

[0097] For example, the first ratio satisfies the following formula 2.

[0098] Formula 2 In the formula, i represents the i-th ONU; This represents the first ratio of the i-th ONU; This represents the predicted traffic for the i-th ONU; This represents the real-time traffic of the i-th ONU; It represents a constant.

[0099] To avoid the problem of the denominator being 0 in Formula 2, a very small constant is set in Formula 2. .

[0100] When the first ratio is greater than 1, it indicates that the real-time traffic is greater than the predicted traffic. This suggests that the ONU may be experiencing a sudden surge in traffic, leading to an increase in real-time traffic. In this case, the bandwidth allocated to the ONU needs to be increased. This can be achieved by increasing the differentiation coefficient, thereby increasing bandwidth allocation to meet the ONU's surge demand. When the first ratio is less than or equal to 1, it indicates that the real-time traffic is less than or equal to the predicted traffic. In this situation, reducing the differentiation coefficient and thus decreasing bandwidth allocation can meet user needs and improve resource utilization.

[0101] Step 23: If the first ratio is less than the first ratio threshold, the first ratio threshold is determined as the differentiation coefficient of ONU.

[0102] To avoid the differential coefficient being too small, resulting in insufficient bandwidth allocated to the ONU and affecting normal business communication, a first ratio threshold is set to limit the lower limit of the differential coefficient.

[0103] For example, the first ratio threshold is 0.5.

[0104] Step 24: If the first ratio is greater than the second ratio threshold, the second ratio threshold is determined as the differentiation coefficient of ONU.

[0105] To avoid an excessively large difference coefficient, which would result in excessive bandwidth being allocated to this ONU and insufficient bandwidth being allocated to other ONUs, thus affecting the normal business communication of other ONUs, a second ratio threshold is set to limit the upper limit of the difference coefficient.

[0106] For example, the second ratio threshold is 5.

[0107] Step 25: If the first ratio is greater than or equal to the first ratio threshold and the first ratio is less than or equal to the second ratio threshold, the first ratio is determined as the differentiation coefficient of the ONU.

[0108] For example, with a first ratio threshold of 0.5 and a second ratio threshold of 5, the above-mentioned difference coefficients satisfy the following formula 3.

[0109] Formula 3 In the formula, i represents the i-th ONU; This represents the differentiation coefficient of the i-th ONU; This represents the predicted traffic for the i-th ONU; This represents the real-time traffic of the i-th ONU; It represents a constant.

[0110] In this way, by setting the differentiation coefficient differently, various situations are fully considered, enabling more accurate and reasonable bandwidth allocation in the future.

[0111] The process of determining the bandwidth allocation priority weight in step 12 above will be described below. As a possible embodiment of this application, the process of determining the bandwidth allocation priority weight in step 12 above can be implemented through steps 31 to 33.

[0112] Step 31: Determine the total guaranteed bandwidth of multiple ONUs, and determine the difference between the total network bandwidth and the total guaranteed bandwidth as the first difference.

[0113] For example, the total network bandwidth is 50Gbps.

[0114] Step 32: Determine the remaining bandwidth by comparing the first difference with the reserved bandwidth.

[0115] By planning and reserving bandwidth in advance, we can flexibly respond to sudden or peak demand for bandwidth resources in the future, thereby improving the network's responsiveness and resilience. Furthermore, setting reserved bandwidth ensures that the total allocated bandwidth does not exceed the network's total bandwidth capacity.

[0116] For example, the reserved bandwidth can be a fixed value, or it can be an adjusted value based on the required bandwidth. For instance, the reserved bandwidth could be 0.05Gbps, 0.1Gbps, etc. As another example, when the ratio of the total bandwidth required by all ONUs to the total network bandwidth exceeds a preset ratio, the reserved bandwidth is increased accordingly. For example, the preset ratio is 90%.

[0117] For example, the remaining bandwidth satisfies the following formula 4.

[0118] Formula 4 In the formula, C represents the remaining bandwidth; G represents the total network bandwidth; and C represents the reserved bandwidth. This indicates the guaranteed bandwidth; i represents the i-th ONU; m represents the total number of ONUs.

[0119] Step 33: Based on the ratio of remaining bandwidth to total network bandwidth, determine the bandwidth allocation priority weight of the ONU corresponding to the service priority.

[0120] In some embodiments, the bandwidth allocation priority weights satisfy the following formula 5.

[0121] Formula 5 In the formula, w represents the bandwidth allocation priority weight. C represents the remaining bandwidth, and C represents the total network bandwidth.

[0122] Formula 5 above clarifies the calculation method for bandwidth allocation priority weights for different service priorities. By setting differentiated bandwidth allocation priority weights for ONUs with different service priorities, the normal operation of high-priority ONUs can be ensured, and the fairness of dynamic bandwidth allocation can be improved.

[0123] In this way, by determining the bandwidth allocation priority weight based on the remaining bandwidth of ONUs with different service priorities, it is possible to ensure that ONUs with high service priorities are allocated more bandwidth, and to prioritize the service priorities of ONUs.

[0124] The following is about the above. Figure 2 The process of determining the traffic prediction model in step S201 will be described below. As a possible embodiment of this application, the ONU device type includes a first type, a second type, and a third type. In this case, the process of determining the traffic prediction model in step S201 can be implemented through steps 41 to 43.

[0125] Step 41: When the ONU device type is the first type, determine the traffic prediction model corresponding to the device type as a neural network model that combines the first long short-term memory network with the attention mechanism.

[0126] Long short-term memory (LSTM) networks can handle long-term dependencies in time series data, while the attention mechanism allows the model to focus on traffic characteristics at key time points when processing sequential data. Combining LSTM with the attention mechanism can improve performance in time series prediction.

[0127] For example, when the ONU's device type is type 1, the corresponding traffic characteristics are periodicity, low latency, and a high-frequency period. Using different models for different ONU device types ensures accurate traffic prediction for each type. A neural network model combining a first long short-term memory network and an attention mechanism can accurately predict the traffic of type 1 ONUs based on their traffic characteristics, thus obtaining the traffic data for those ONUs.

[0128] For example, historical traffic from ONUs of type 1 is collected and divided into training, validation, and test sets at ratios of 60%, 20%, and 20%, respectively. The training set data is input into a neural network model combining a first long short-term memory network and an attention mechanism for training, yielding the output. The output is compared with the real-time traffic at that moment, and the difference is quantified using root mean squared error (RMSE) and mean absolute error (MAE). The neural network model combining the first short-term memory network and the attention mechanism is then adjusted based on the quantification results.

[0129] Step 42: When the ONU device type is the second type, determine the traffic prediction model corresponding to the device type as a neural network model that combines the second long short-term memory network with the attention mechanism.

[0130] For example, when the ONU's device type is the second type, the corresponding traffic characteristics are continuous high traffic and bursty traffic. A neural network model combining a second long short-term memory network and an attention mechanism can accurately predict the traffic of the second type of ONU based on its traffic characteristics, thereby obtaining the traffic flow of the second type of ONU.

[0131] For example, historical traffic from ONUs of type 2 is collected and divided into training, validation, and test sets at 60%, 20%, and 20% respectively. The training set data is input into a neural network model combining a second long short-term memory network and an attention mechanism for training, yielding the output. The output is compared with the real-time traffic at that moment, and the difference is quantified using RMSE and MAE. The neural network model combining the second short-term memory network and the attention mechanism is then adjusted based on the quantization results.

[0132] Step 43: When the ONU device type is the third type, determine that the traffic prediction model corresponding to the device type is a neural network model that combines the third long short-term memory network with the attention mechanism.

[0133] For example, when the ONU's device type is the third type, the corresponding traffic characteristics are low traffic, periodic traffic, and a low-frequency periodicity. A neural network model combining a third long short-term memory network with an attention mechanism can accurately predict the traffic of the third type of ONU based on its traffic characteristics, thereby obtaining the traffic data of the third type of ONU.

[0134] For example, historical traffic from ONUs of type 3 is collected and divided into training, validation, and test sets at ratios of 60%, 20%, and 20%, respectively. The training set data is input into a neural network model combining a third-generation long short-term memory network and an attention mechanism for training, yielding the output. The output is compared with the real-time traffic at that moment, and the difference is quantified using RMSE and MAE. The neural network model combining the third-generation short-term memory network and the attention mechanism is then adjusted based on the quantization results.

[0135] Furthermore, the three neural network models that combine long short-term memory networks with attention mechanisms can be designed with different layer depths and number of units based on the traffic characteristics corresponding to different ONU device types, in order to handle different heterogeneous traffic.

[0136] For example, in actual use, the neural network model combining the first long short-term memory network with the attention mechanism, the neural network model combining the second long short-term memory network with the attention mechanism, and the neural network model combining the third long short-term memory network with the attention mechanism can be spliced ​​together to generate a multi-branch LSTM-Attention model. Figure 3 This is a schematic diagram of the structure of a multi-branch LSTM-Attention model provided in an embodiment of this application.

[0137] like Figure 3 As shown, the multi-branch LSTM-Attention model includes an input layer, a neural network model combining a first long short-term memory network and an attention mechanism, a neural network model combining a second long short-term memory network and an attention mechanism, a neural network model combining a third long short-term memory network and an attention mechanism, a splicing layer, a fully connected layer, and an output layer.

[0138] The input layer is used to input historical traffic. The first neural network model combining a Long Short-Term Memory (LSTM) network and an attention mechanism includes LSTM layers, dropout layers, attention layers, pooling layers, and fully connected layers. The second and third neural network models combining LSTM networks and an attention mechanism also include LSTM layers, dropout layers, attention layers, pooling layers, and fully connected layers. These three models, through concatenation layers and fully connected layers, ultimately output the predicted traffic. The output layer is used to output the predicted traffic.

[0139] Thus, a neural network model combining different long short-term memory networks with attention mechanisms enables traffic prediction for ONUs of different device types. Furthermore, the attention mechanism captures bursty characteristics, achieving comprehensive traffic prediction. Predictions are made separately for historical traffic of ONUs of different device types, fully considering the traffic characteristics of different device types, resulting in more accurate predictions and providing the possibility for customized services.

[0140] The following is about the above. Figure 2 The process of bandwidth allocation in step S204 will be described below. As a possible embodiment of this application, the process of bandwidth allocation in step S204 can be implemented by steps 51 to 54.

[0141] Step 51: Determine the total guaranteed bandwidth of multiple ONUs, and determine the difference between the total network bandwidth and the total guaranteed bandwidth as the first difference.

[0142] Step 52: Determine the remaining bandwidth by comparing the first difference with the reserved bandwidth.

[0143] For example, the remaining bandwidth satisfies the following formula 6.

[0144] Formula 6 In the formula, C represents the remaining bandwidth; G represents the total network bandwidth; and C represents the reserved bandwidth. This indicates the guaranteed bandwidth; i represents the i-th ONU; m represents the total number of ONUs.

[0145] Step 53: Multiply the remaining bandwidth by the bandwidth allocation ratio and sum the product with the guaranteed bandwidth to determine the bandwidth allocation for each ONU.

[0146] In some embodiments, the bandwidth of each ONU satisfies the following formula 7.

[0147] Formula 7 In the formula, i represents the i-th ONU; This represents the bandwidth allocation for the i-th ONU; This indicates guaranteed bandwidth; This represents the bandwidth allocation ratio of the i-th ONU; j represents the j-th ONU; m represents the total number of ONUs; This indicates the amount of bandwidth remaining.

[0148] in, ; This represents the bandwidth requirement of the j-th ONU; This represents the difference coefficient of the j-th ONU; This represents the bandwidth allocation priority weight of the j-th ONU.

[0149] Thus, Formula 7 above clearly defines the bandwidth allocation for each ONU. Essentially, for each ONU, the bandwidth allocation consists of two parts: a fixed value, i.e., guaranteed bandwidth, and a floating value, i.e., the remaining bandwidth allocated according to the bandwidth allocation ratio. This not only guarantees the basic services of the ONU but also allows for allocation based on the different service needs of the ONU, improving the accuracy of bandwidth allocation.

[0150] Step 54: Allocate bandwidth to the corresponding ONU according to the bandwidth allocation amount of each ONU.

[0151] The bandwidth allocation for each ONU is formed into a bandwidth mapping domain (BW-map) and then broadcast.

[0152] In this way, by ensuring bandwidth and bandwidth allocation ratio to generate the final determined bandwidth allocation, compared with the static setting of bandwidth allocation, it can fully meet the differentiated needs of ONUs with different service priorities, and can meet the current needs of different ONUs, and can allocate bandwidth to ONUs more accurately.

[0153] In some embodiments, the guaranteed bandwidth is determined by steps 61 to 64.

[0154] Step 61: Determine the usage ratio as the ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth.

[0155] For example, the occupancy ratio satisfies the following formula 8.

[0156] Formula 8 In the formula, v represents the occupancy ratio; i represents the i-th ONU; and m represents the total number of ONUs. C represents the bandwidth requirement of the i-th ONU; C represents the total network bandwidth.

[0157] Step 62: If the occupancy rate is less than the first occupancy threshold, the first preset value is determined as the guaranteed bandwidth.

[0158] For example, the first occupancy threshold is 40%.

[0159] For example, the first preset value is 0.05Gbps.

[0160] If the bandwidth occupancy rate is less than the first occupancy threshold, it indicates that the overall network load is low. In this case, reducing the guaranteed bandwidth is equivalent to increasing the available bandwidth, meaning that for each ONU, the bandwidth occupancy is increased by allocating the available bandwidth according to the bandwidth allocation ratio. This allows the bandwidth allocation for each ONU to better meet its service requirements.

[0161] Step 63: When the occupancy rate is greater than or equal to the first occupancy threshold and less than or equal to the second occupancy threshold, the second preset value is determined as the guaranteed bandwidth.

[0162] For example, the second occupancy threshold is 90%.

[0163] For example, the second preset value is 0.1Gbps.

[0164] When the occupancy rate is greater than or equal to the first occupancy threshold and less than or equal to the second occupancy threshold, it indicates that the network is under normal load. The bandwidth value is then clearly guaranteed by setting the second preset value.

[0165] Step 64: When the occupancy rate is greater than the second occupancy threshold, the guaranteed bandwidth corresponding to the ONU with the first priority is determined to be the third preset value, and the guaranteed bandwidth corresponding to the ONU with the non-first priority is determined to be the second preset value.

[0166] Among them, the first preset value is less than the second preset value, and the second preset value is less than the third preset value.

[0167] For example, the third preset value is 0.2Gbps.

[0168] For example, the first preset value, the second preset value, and the third preset value may be multiples of each other. For instance, the second preset value may be twice the first preset value, and the third preset value may be twice the second preset value.

[0169] When the occupancy rate exceeds the second occupancy threshold, it indicates that the network is under high load. Introducing a congestion control mechanism at this point, which assigns different guaranteed bandwidths to ONUs with different service priorities, can reduce the amount of bandwidth remaining. While ensuring normal ONU communication, it can also increase the bandwidth allocation value for ONUs with higher service priorities.

[0170] Figure 4 A flowchart illustrating another bandwidth allocation method for a passive optical network provided in this application embodiment. The method includes the following steps: S401. Based on the device type of each ONU among multiple ONUs, determine the traffic prediction model corresponding to the device type of each ONU.

[0171] S402. Based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, traffic prediction is performed to obtain the predicted traffic of each ONU.

[0172] S403. For each ONU, the maximum value between the ONU's predicted traffic and real-time traffic is determined as the ONU's required bandwidth.

[0173] S404. Determine the usage ratio as the ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth.

[0174] S405. Determine the guaranteed bandwidth based on the occupancy rate.

[0175] S406. For each ONU, determine the differentiation coefficient of the ONU based on the predicted traffic and real-time traffic of the ONU; determine the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the service priority of the ONU.

[0176] S407. Determine the bandwidth allocation ratio of the ONU based on the product of the ONU's required bandwidth, the ONU's differentiation coefficient, and the ONU's bandwidth allocation priority weight.

[0177] S408. Based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among multiple ONUs, bandwidth is allocated to the corresponding ONU.

[0178] As a concrete example, to verify the feasibility of the above method, a simulation experiment was conducted. Simulation parameters such as the number of ONUs, DBA cycle, and total network bandwidth were initialized. An industrial PON with a total network bandwidth of 50Gbps was used as an example. This industrial PON includes 32 ONUs, and the fiber optic distance from the ONUs to the OLT is 20 kilometers.

[0179] ONUs 1 to 8 connect to robot controller terminals, totaling 16 terminals. ONUs 9 to 24 connect to monitoring camera terminals, totaling 24 cameras. ONUs 25 to 32 connect to temperature and humidity sensor terminals, totaling 120 terminals. Each terminal generates random data packets (64-1518 bytes) following a Pareto distribution. The dynamic bandwidth allocation period is set to 2 milliseconds, and the simulation time is 120 seconds.

[0180] Among them, the robot controller terminal is of type 1 and has the highest service priority; the monitoring camera terminal is of type 2 and has the highest service priority; the temperature and humidity sensor terminal is of type 3 and has the highest service priority. Specifically, ONUs 1 to 8 are of type 1 and have the highest service priority; ONUs 9 to 24 are of type 2 and have the highest service priority; and ONUs 25 to 32 are of type 3 and have the highest service priority.

[0181] The real-time traffic of ONU 1, ONU 9, and ONU 25 is monitored. To more closely resemble the real-world scenario, the periodic traffic of the terminals on the production line is retained during the simulation, while random burst traffic is introduced. The traffic from all ONUs is aggregated and then input into a multi-branch LSTM-Attention neural network for processing.

[0182] Figure 5 This is a schematic diagram illustrating the real-time traffic of multiple ONUs provided in an embodiment of this application. For example... Figure 5 The figures show the real-time traffic of ONU 1, ONU 9, and ONU 25, respectively.

[0183] Historical traffic data from 32 ONUs was collected as input data. Traffic prediction was performed using three neural network models combining a first long short-term memory network with an attention mechanism, a second long short-term memory network with an attention mechanism, and a third long short-term memory network with an attention mechanism. The model learning rate was 0.005. Figure 6 These are schematic diagrams illustrating multiple predicted traffic and real-time traffic patterns provided in embodiments of this application. Figure 6 As shown, ONU 1 has an RMSE of 0.0647 Gbps and a MAE of 0.03%. The lower MAE value indicates higher prediction accuracy and precise capture of burst flow peaks. ONU 9 has an RMSE of 0.4845 Gbps and a MAE of 0.36%, with negligible errors. ONU 25 has an RMSE of 0.0545 Mbps, with extremely small and negligible MAE errors. In the simulation, the burst flow from the temperature and humidity sensor was set as a short-duration pulse. Such signals are difficult to capture, limiting the model's predictive ability. Considering the small flow rate and low priority of the temperature and humidity sensor, its prediction error is negligible.

[0184] The performance of the bandwidth allocation method for passive optical networks provided in this application embodiment is compared with that of the traditional IPACT method. Figure 7 This is a schematic diagram illustrating a bandwidth allocation performance comparison provided in an embodiment of this application. Figure 7 As shown, when the normalized network load is between 0.1 and 0.4, i.e., under low network load conditions, the average latency of the passive optical network (PON) bandwidth allocation method is slightly higher, due to the increased computational overhead from traffic prediction. When the normalized network load is between 0.5 and 0.7, i.e., under medium network load conditions, the advantage of the PON bandwidth allocation method begins to emerge by improving prediction accuracy through a traffic prediction model. When the normalized network load is between 0.8 and 1.0, i.e., under high network load conditions, the PON bandwidth allocation method exhibits a significant advantage, effectively suppressing the average latency through the synergistic effect of traffic prediction and congestion control. Simulation results show that under all load conditions, the PON bandwidth allocation method can stably control the average latency to around 1 millisecond, and it demonstrates a significant advantage when the load exceeds 0.6.

[0185] This application embodiment can divide the bandwidth allocation device of a passive optical network into functional modules or functional units according to the above method example. For example, each function can be divided into a separate functional module or functional unit, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or in software as a functional module or functional unit. The module or unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0186] Figure 8 This is a schematic diagram of the structure of a bandwidth allocation device 80 for a passive optical network provided in an embodiment of this application. The bandwidth allocation device 80 for the passive optical network includes: The first determining unit 801 is used to determine the traffic prediction model corresponding to the device type of each ONU based on the device type of each ONU in the multiple optical network units (ONUs); the prediction unit 802 is used to perform traffic prediction based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, and obtain the predicted traffic of each ONU; the second determining unit 803 is used to determine the bandwidth allocation ratio of each ONU in the multiple ONUs based on the predicted traffic of the multiple ONUs, the real-time traffic of the multiple ONUs, and the service priority of the multiple ONUs; the allocation unit 804 is used to allocate bandwidth to the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU in the multiple ONUs, where the guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

[0187] In one possible implementation, the second determining unit 803 is configured to: for each ONU, determine the required bandwidth and differentiation coefficient of the ONU based on the predicted traffic and real-time traffic of the ONU, wherein the differentiation coefficient is used to characterize the degree of differentiation between the real-time traffic and the predicted traffic of the ONU; determine the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the service priority of the ONU; and determine the bandwidth allocation ratio of the ONU based on the product of the required bandwidth of the ONU, the differentiation coefficient of the ONU and the bandwidth allocation priority weight of the ONU.

[0188] In one possible implementation, the second determining unit 803 is configured to: determine the maximum value between the predicted traffic and the real-time traffic of the ONU as the required bandwidth of the ONU; determine the ratio of the real-time traffic to the predicted traffic of the ONU as a first ratio; if the first ratio is less than a first ratio threshold, determine the first ratio threshold as the differentiation coefficient of the ONU; if the first ratio is greater than a second ratio threshold, determine the second ratio threshold as the differentiation coefficient of the ONU; and if the first ratio is greater than or equal to the first ratio threshold and less than or equal to the second ratio threshold, determine the first ratio as the differentiation coefficient of the ONU.

[0189] In one possible implementation, the second determining unit 803 is used to: determine the total guaranteed bandwidth of multiple ONUs, and determine the difference between the total network bandwidth and the total guaranteed bandwidth as a first difference; determine the difference between the first difference and the reserved bandwidth as the bandwidth surplus; and determine the bandwidth allocation priority weight of the ONU corresponding to the service priority based on the ratio of the bandwidth surplus to the total network bandwidth.

[0190] In one possible implementation, the bandwidth allocation priority weights satisfy the following formula:

[0191] In the formula, w represents the bandwidth allocation priority weight. C represents the remaining bandwidth, and C represents the total network bandwidth.

[0192] In one possible implementation, the device type includes a first type, a second type, and a third type; the first determining unit 801 is configured to: when the ONU's device type is the first type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a first long short-term memory network and an attention mechanism; when the ONU's device type is the second type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a second long short-term memory network and an attention mechanism; and when the ONU's device type is the third type, determine that the traffic prediction model corresponding to the device type is a neural network model combining a third long short-term memory network and an attention mechanism.

[0193] In one possible implementation, the allocation unit 804 is used to: determine the total guaranteed bandwidth of multiple ONUs; determine the difference between the total network bandwidth and the total guaranteed bandwidth as a first difference; determine the difference between the first difference and the reserved bandwidth as the bandwidth surplus; determine the bandwidth allocation amount for each ONU by multiplying the bandwidth surplus amount by the bandwidth allocation ratio and the sum of the guaranteed bandwidth; and allocate bandwidth to the corresponding ONU according to the bandwidth allocation amount for each ONU.

[0194] In one possible implementation, the bandwidth of each ONU satisfies the following formula:

[0195] In the formula, i represents the i-th ONU; This represents the bandwidth allocation for the i-th ONU; This indicates guaranteed bandwidth; This represents the bandwidth allocation ratio of the i-th ONU; j represents the j-th ONU; m represents the total number of ONUs; This indicates the amount of bandwidth remaining.

[0196] In one possible implementation, the third determining unit is configured to: determine the ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth as the occupancy ratio; determine a first preset value as the guaranteed bandwidth when the occupancy ratio is less than a first occupancy threshold; determine a second preset value as the guaranteed bandwidth when the occupancy ratio is greater than or equal to the first occupancy threshold and less than or equal to a second occupancy threshold; and determine the guaranteed bandwidth corresponding to ONUs with a service priority of first priority as a third preset value and the guaranteed bandwidth corresponding to ONUs with a service priority other than first priority as a second preset value when the occupancy ratio is greater than the second preset value; wherein the first preset value is less than the second preset value, and the second preset value is less than the third preset value.

[0197] Through the above description of the embodiments, those skilled in the art will clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0198] It should be noted that the various embodiments of this application can be referenced or learned from each other. For example, the same or similar steps, method embodiments, system embodiments and device embodiments can be referenced from each other without limitation.

[0199] When implemented in hardware, the various modules in the bandwidth allocation device of a passive optical network can be integrated into, for example... Figure 9 The bandwidth allocation device of the passive optical network shown is implemented in hardware. Specifically, as... Figure 9 As shown, the basic hardware structure of the bandwidth allocation device for passive optical networks is introduced.

[0200] Figure 9 This is a schematic diagram of the hardware structure of a bandwidth allocation device for a passive optical network provided in an embodiment of this application. Figure 9 As shown, the bandwidth allocation device of the passive optical network includes at least one processor 901, a communication line 902, and at least one communication interface 904, and may also include a memory 903. The processor 901, memory 903, and communication interface 904 can be connected via the communication line 902.

[0201] The processor 901 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0202] Communication line 902 may include a path for transmitting information between the aforementioned components.

[0203] The communication interface 904 is used to communicate with other devices or communication networks. It can use any transceiver-like device, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0204] The memory 903 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of including or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0205] In one possible design, the memory 903 can exist independently of the processor 901, meaning the memory 903 can be an external memory of the processor 901. In this case, the memory 903 can be connected to the processor 901 via a communication line 902 to store execution instructions or application code, and its execution is controlled by the processor 901 to implement the bandwidth allocation method for the passive optical network provided in the following embodiments of this application. In another possible design, the memory 903 can also be integrated with the processor 901, meaning the memory 903 can be an internal memory of the processor 901. For example, the memory 903 can be a cache, used to temporarily store some data and instruction information.

[0206] As one possible implementation, processor 901 may include one or more CPUs, for example Figure 9 CPU0 and CPU1 in the example. As another possible implementation, the bandwidth allocation device of a passive optical network may include multiple processors, such as... Figure 9 The processors 901 and 907 are included. As another possible implementation, the bandwidth allocation device for the passive optical network may also include an output device 905 and an input device 906.

[0207] This application provides a computer program product containing instructions that, when run on a computer, cause the computer to execute the bandwidth allocation method for a passive optical network described in the above method embodiments.

[0208] This application also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the bandwidth allocation method for a passive optical network in the method flow shown in the above method embodiments.

[0209] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires; a portable computer disk drive; a hard disk drive; random access memory (RAM); read-only memory (ROM); erasable programmable read-only memory (EPROM); a register; a hard disk drive; an optical fiber; a compact disc read-only memory (CD-ROM); an optical storage device; a magnetic storage device; or any suitable combination thereof; or any other form of computer-readable storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In the embodiments of this application, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0210] Since the bandwidth allocation device, computer-readable storage medium, and computer program product of the passive optical network in the embodiments of this application can be applied to the above method, the technical effects that can be obtained can also be referred to the above method embodiments. The embodiments of this application will not be repeated here.

[0211] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0212] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0213] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0214] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A bandwidth allocation method for a passive optical network, characterized in that, The method includes: Based on the device type of each ONU in multiple optical network units (ONUs), determine the traffic prediction model corresponding to the device type of each ONU. Traffic prediction is performed based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type to obtain the predicted traffic of each ONU; Based on the predicted traffic of the multiple ONUs, the real-time traffic of the multiple ONUs, and the service priority of the multiple ONUs, the bandwidth allocation ratio of each ONU among the multiple ONUs is determined; Based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among the multiple ONUs, bandwidth is allocated to the corresponding ONU. The guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

2. The method according to claim 1, characterized in that, The determination of the bandwidth allocation ratio for each ONU based on the predicted traffic, real-time traffic, and service priorities of the multiple ONUs includes: For each ONU, based on the predicted traffic and real-time traffic of the ONU, the required bandwidth and differentiation coefficient of the ONU are determined, and the differentiation coefficient is used to characterize the degree of difference between the real-time traffic and the predicted traffic of the ONU. Based on the guaranteed bandwidth and the service priority of the ONU, the bandwidth allocation priority weight of the ONU is determined; The bandwidth allocation ratio of the ONU is determined by multiplying the required bandwidth of the ONU, the differentiation coefficient of the ONU, and the bandwidth allocation priority weight of the ONU.

3. The method according to claim 2, characterized in that, The determination of the required bandwidth and differentiation coefficient of the ONU based on the predicted and real-time traffic of the ONU includes: The maximum value between the predicted traffic and the real-time traffic of the ONU is determined as the required bandwidth of the ONU; The ratio of the real-time traffic to the predicted traffic of the ONU is determined as the first ratio. If the first ratio is less than the first ratio threshold, the first ratio threshold is determined as the differentiation coefficient of the ONU; If the first ratio is greater than the second ratio threshold, the second ratio threshold is determined as the differentiation coefficient of the ONU; If the first ratio is greater than or equal to the first ratio threshold and the first ratio is less than or equal to the second ratio threshold, the first ratio is determined as the differentiation coefficient of the ONU.

4. The method according to claim 2, characterized in that, The determination of the bandwidth allocation priority weight of the ONU based on the guaranteed bandwidth and the service priority of the ONU includes: Determine the total guaranteed bandwidth of the multiple ONUs, and define the difference between the total network bandwidth and the total guaranteed bandwidth as the first difference. The difference between the first difference and the reserved bandwidth is determined as the remaining bandwidth. Based on the ratio of the remaining bandwidth to the total network bandwidth, the bandwidth allocation priority weight of the ONU corresponding to the service priority is determined.

5. The method according to claim 4, characterized in that, The bandwidth allocation priority weights satisfy the following formula: In the formula, w represents the bandwidth allocation priority weight. C represents the remaining bandwidth, and C represents the total network bandwidth.

6. The method according to claim 1, characterized in that, The device types of the ONU include the first type, the second type, and the third type; The traffic prediction model for each ONU's device type, based on the device type of each ONU among multiple optical network units (ONUs), includes: When the device type of the ONU is the first type, the traffic prediction model corresponding to the device type is determined to be a neural network model that combines a first long short-term memory network with an attention mechanism; When the device type of the ONU is the second type, the traffic prediction model corresponding to the device type is determined to be a neural network model that combines a second long short-term memory network with an attention mechanism; When the device type of the ONU is the third type, the traffic prediction model corresponding to the device type is determined to be a neural network model that combines a third long short-term memory network with an attention mechanism.

7. The method according to claim 1, characterized in that, The process of allocating bandwidth to the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among the multiple ONUs includes: Determine the total guaranteed bandwidth of the multiple ONUs, and define the difference between the total network bandwidth and the total guaranteed bandwidth as the first difference. The difference between the first difference and the reserved bandwidth is determined as the remaining bandwidth. The bandwidth allocation for each ONU is determined by summing the product of the remaining bandwidth and the bandwidth allocation ratio with the guaranteed bandwidth. Bandwidth is allocated to the corresponding ONU according to the bandwidth allocation amount for each ONU.

8. The method according to claim 7, characterized in that, The bandwidth of each ONU satisfies the following formula: In the formula, i represents the i-th ONU; This represents the bandwidth allocation for the i-th ONU; This indicates the guaranteed bandwidth; This represents the bandwidth allocation ratio of the i-th ONU; j represents the j-th ONU; m represents the total number of ONUs; This indicates the remaining bandwidth.

9. The method according to claim 2, characterized in that, The guaranteed bandwidth is determined in the following way: The ratio of the sum of the bandwidth requirements of all ONUs to the total network bandwidth is determined as the usage ratio; If the occupancy rate is less than the first occupancy threshold, the first preset value is determined as the guaranteed bandwidth; If the occupancy ratio is greater than or equal to the first occupancy threshold and the occupancy ratio is less than or equal to the second occupancy threshold, the second preset value is determined as the guaranteed bandwidth. When the occupancy rate is greater than the second occupancy threshold, the guaranteed bandwidth corresponding to the ONU with the first priority is determined to be the third preset value, and the guaranteed bandwidth corresponding to the ONU with the non-first priority is determined to be the second preset value. Wherein, the first preset value is less than the second preset value, and the second preset value is less than the third preset value.

10. A bandwidth allocation device for a passive optical network, characterized in that, The device includes: The first determining unit is used to determine the traffic prediction model corresponding to the device type of each ONU based on the device type of each ONU in the multiple optical network units (ONUs); The prediction unit is used to predict the traffic based on the historical traffic of each ONU and the traffic prediction model corresponding to the device type, so as to obtain the predicted traffic of each ONU. The second determining unit is used to determine the bandwidth allocation ratio of each ONU among the multiple ONUs based on the predicted traffic of the multiple ONUs, the real-time traffic of the multiple ONUs, and the service priority of the multiple ONUs. The allocation unit is used to allocate bandwidth to the corresponding ONU based on the guaranteed bandwidth and the bandwidth allocation ratio of each ONU among the plurality of ONUs. The guaranteed bandwidth is used to characterize the minimum bandwidth allocated to ensure the normal operation of the ONU.

11. An electronic device, characterized in that, include: A processor and a communication interface; the communication interface is coupled to the processor, the processor being configured to run computer programs or instructions to implement the method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed by a computer, perform the method as described in any one of claims 1-9.

13. A computer program product, characterized in that, The computer program product includes computer instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1-9.

Images