Multi-passive optical network system based on digital subcarrier multiplexing and dynamic bandwidth allocation method thereof

Abstract

The invention discloses a multi-passive optical network system based on digital subcarrier multiplexing and a dynamic bandwidth allocation method thereof, and the scheme comprises the steps: dividing a single optical carrier into a plurality of independent and adjustable subcarrier channels through a digital subcarrier multiplexing technology on an electric domain, and dynamically adjusting the number of subcarriers allocated to each PON (Passive Optical Network), therefore, differentiated bandwidth allocation among different PONs is realized, and the bandwidth requirements of multiple PONs changing along with regions and time are met. The dynamic bandwidth allocation method comprises the following steps: S1, predicting a bandwidth demand for each ONU (Optical Network Unit) in multiple PONs (Passive Optical Network Units), calculating a demand satisfaction rate and determining a priority; s2, counting bandwidth requirements of different priorities of each PON; s3, calculating and dynamically distributing a certain number of subcarriers of each PON; s4, dynamically allocating bandwidth to each ONU in the PON according to the priority; s5, activating / closing neighbor subcarriers; and S6, carrying out global subcarrier reconfiguration.

Description

Technical Field

[0001] The present invention relates to the field of resource allocation of optical fiber access communication technology, specifically a multi-passive optical network (PON) system based on digital subcarrier multiplexing (DSCM) and a dynamic bandwidth allocation method thereof, which is suitable for efficient spectrum resource management in multi-service convergence scenarios. Background Art

[0002] With the convergence of multiple services, including 5G backhaul, enterprise access, and home broadband, the data traffic carried by optical networks is growing at an unprecedented rate. User demands for network resources are no longer limited to simple bandwidth expansion; they increasingly prioritize flexibility, timeliness, and adaptability. Traditional PON systems, which employ fixed wavelength configurations for each PON, are increasingly lacking in their ability to share wavelengths and bandwidth to serve diverse services. They struggle to cope with the diverse, dynamic, and sudden bandwidth demands across PONs caused by temporal and regional variations, leading to low resource utilization. To address this issue, multi-PON system technology is gaining attention. A multi-PON system involves the simultaneous deployment of multiple PON networks within the same fiber infrastructure. It typically uses multiple wavelengths for shared transmission across multiple PONs, increasing the overall capacity and resource efficiency of the PON network and meeting the diverse, high-bandwidth service demands of the future.

[0003] In existing PON technology, static wavelength and bandwidth allocation strategies between PONs cannot respond to traffic fluctuations in real time, resulting in resource waste or shortages. This is particularly true in medium-to-high load scenarios, where latency degrades significantly. This fails to meet the QoS requirements of high-priority services and the "on-demand allocation and real-time scheduling" demands of next-generation optical access networks. Therefore, more efficient and refined resource management methods are urgently needed. High-speed multi-PON technology based on electrical domain digital subcarrier multiplexing (DSCM) and coherent optical transceivers is one such technology. DSCM technology carries multiple digital subcarriers (SCs) on a single optical carrier, each capable of independently carrying data, enabling refined management and efficient utilization of spectrum resources. This flexible subcarrier configuration supports dynamic, on-demand bandwidth allocation, enabling the network to quickly respond to traffic fluctuations and avoiding the resource waste or shortages found in traditional network architectures. Spectrum management enables flexible and efficient bandwidth allocation, significantly improving channel utilization and reducing resource waste. Furthermore, by adopting DSCM technology, multi-PON systems only need to use a single wavelength. Compared with traditional solutions that use multiple wavelengths and laser arrays, they require fewer lasers and do not require additional wavelength switching and routing equipment (such as AWGs). This effectively reduces the complexity and overall cost of network equipment, further improving the economic efficiency and practicality of the network. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-PON (Passive Optical Network) system architecture and its dynamic bandwidth allocation method based on digital subcarrier multiplexing technology (DSCM) to achieve flexible control and precise allocation of network resources and comprehensively improve the adaptability of the network in multi-service convergence scenarios.

[0005] To achieve the above objectives, the present invention uses digital subcarrier multiplexing technology to enable each subcarrier channel to independently carry data. On this basis, subcarriers are dynamically allocated to each PON and bandwidth is dynamically allocated to each ONU within the PON, thereby maximizing network resource utilization and reducing latency. At the same time, when ONU bandwidth allocation is unsuccessful, adjacent subcarrier resources are preferentially activated / deactivated, reducing the spectrum reconstruction cost and latency caused by frequent subcarrier migration. Specifically, the following steps are included:

[0006] Step S1: predict bandwidth requirements for each ONU in the multi-PON, calculate the demand satisfaction rate and determine the priority;

[0007] Step S2: Count the bandwidth requirements of each PON;

[0008] Step S3: Calculate and dynamically allocate a certain number of subcarriers to each PON;

[0009] Step S4: Dynamically allocate bandwidth to each ONU in the PON based on priority;

[0010] Step S5: activating / deactivating neighbor subcarriers;

[0011] Step S6: Perform global subcarrier reconfiguration;

[0012] Furthermore, this allocation method uses two separate cycles:

[0013] Long cycle (10 minutes): During this cycle, the number of subcarriers is dynamically allocated to each PON based on service demand.

[0014] Short cycle (2 milliseconds): Within this cycle, bandwidth resources are quickly and dynamically allocated between ONUs within each PON. When the service change within a PON exceeds a predetermined limit, the number of subcarriers is triggered to be reallocated.

[0015] The structure of a multi-PON system based on digital subcarrier multiplexing divides the available bandwidth of the PON system into multiple subcarriers and flexibly allocates them among multiple PONs in a P2MP (point-to-multipoint) manner. Each subcarrier can be processed independently of other subcarriers, including modulation, management, aggregation, etc., and routed to different destinations. It can flexibly allocate subcarriers among different PONs according to different service scenarios, and flexibly allocate bandwidth among different ONUs (optical network units) in the same PON, improving the utilization rate of optical fiber resources and simplifying the planning and upgrade of the access network.

[0016] (1) ONU Bandwidth Demand Prediction, Dynamic Priority Judgment, and Subcarrier Scheduling (Step S1): For the i-th optical network unit (ONU), define the demand satisfaction rate DSR i , preset a threshold T (e.g., T = 0.8). If DSR i < T, mark it as a high-priority ONU, otherwise as a low-priority ONU. The definition of DSR i is as follows:

[0017]

[0018] Among them, represents the bandwidth actually allocated to the ONU i in the previous cycle, is the predicted demand of the ONU i in the previous cycle.

[0019] Load Prediction: Collect the bandwidth demand of the ONU in the previous several time periods, and predict the bandwidth demand of the ONU in the next time period through the moving average algorithm <(

[0020]

[0021] where W is the length of the moving average window, that is, select the data of the past W cycles for average calculation; d i (t−τ) is the actual bandwidth demand of the i-th ONU in the past t−τ cycle.

[0022] Subcarrier Scheduling: First allocate subcarriers to high-priority ONUs. When resources are insufficient, execute adjacent subcarrier release, service migration, or global reconfiguration strategies.

[0023] This step dynamically adjusts subcarrier resources, first ensuring that high-priority services can obtain the required bandwidth resources, thus preventing the quality of service of high-priority services from degrading due to insufficient bandwidth. At the same time, when resources are tight, subcarrier release or service migration is implemented, making the network resource allocation more flexible and efficient, avoiding the resource waste caused by traditional static configuration, and improving the overall spectrum utilization rate and network response speed.

[0024] (2) Counting the bandwidth requirements of each PON (step S2)

[0025] Count the total bandwidth requirements of high-priority ONUs of each PON separately

[0026]

[0027] and the total bandwidth requirements of low-priority ONUs

[0028]

[0029] They are used to determine the overall bandwidth requirements of each PON

[0030]

[0031] Where α is a weight coefficient used to control the impact of low-priority requests on subcarrier allocation. For example, if α = 0.6, a larger α value (e.g., 0.6 ≤ α ≤ 1) will reserve more subcarrier resources, thereby reducing the frequency of subsequent inter-PON subcarrier changes.

[0032] The advantage of this step is that it reduces the frequency of subsequent subcarrier adjustments and resource reconfigurations between PONs. Frequent subcarrier changes can lead to network service interruptions or unstable performance. However, reserving certain resources in advance with a larger weight coefficient can significantly reduce the frequency of system reconfiguration, thereby improving network stability and service continuity. This is particularly suitable for environments where frequent spectrum resource adjustments are costly, maintaining stable system operation and reducing maintenance costs.

[0033] (3) Calculate and dynamically allocate the number of subcarriers for each PON (step S3)

[0034] Dynamically adjust subcarrier allocation based on bandwidth requirements of each PON: Dynamically adjust the number of subcarriers of each PON based on bandwidth requirements of each PON. Assume that the system has N subcarriers, n PONs, and each PON has N subcarriers. j The calculation is as follows:

[0035]

[0036] in, PON j The total bandwidth requirement.

[0037] The number of subcarriers N of the last PON n The following formula will be used to adjust to ensure that all subcarrier resources are allocated:

[0038]

[0039] The specific advantages of this step are: First, it leverages the proportional relationship between bandwidth requirements to fairly and reasonably allocate resources to each PON, effectively meeting users' differentiated bandwidth needs and avoiding over- or under-allocation of resources. Second, the number of subcarriers for the last PON in the formula is determined by the difference in the subcarrier counts of all previous PONs. This ensures full utilization of subcarrier resources, maximizing overall spectrum resource utilization and management efficiency. This flexible configuration significantly enhances the network's adaptability to load changes, reduces the likelihood of frequent resource reconfiguration, and improves network stability and reliability.

[0040] (4) Dynamically allocate bandwidth to ONUs in each PON based on priority (step S4)

[0041] After differentiating high-priority and low-priority ONUs by demand fulfillment rate, bandwidth is allocated to high-priority ONUs first, and then to low-priority ONUs. When allocating bandwidth to ONUs, calculate the current ONU i Unmet needs r i (Initially set to d i ). In the subcarrier set, find the subcarrier k with remaining capacity and perform a bandwidth allocation δ=min(r i , remaining capacity). Update r i ←r i -δ and the available capacity of the subcarrier; if r i If the bandwidth allocated to the high-priority ONU is insufficient, bandwidth can be released from the neighboring low-utilization subcarriers to meet the demand.

[0042] (5) Adjacent subcarrier activation / deactivation (step S5):

[0043] When an ONU needs more bandwidth and exceeds the idle bandwidth of all subcarriers allocated to the ONU, it first releases low-utilization subcarriers from its neighboring ONUs and then allocates them to the high-demand ONU.

[0044] Subcarrier utilization and release strategy: Let u k is the utilization rate of subcarrier k, and its expression is as follows:

[0045]

[0046] where b i，k Assigned to ONU i The bandwidth on subcarrier k, C k is the maximum capacity of the subcarrier. kIf the value is lower than the reference value U, the subcarrier is considered to be partially or completely released to other ONUs in need. When the subcarrier is released, the original ONU services on the subcarrier will be migrated. If the adjacent subcarrier is successfully released, the process returns to step S4 to continue allocation. If the release fails, the subcarrier is shared across PONs.

[0047] ① Sharing of subcarrier resources between neighboring ONUs in the same PON

[0048] First, the algorithm checks subcarrier resources within a single PON. If the bandwidth demand allocated to a particular ONU cannot be directly met, the algorithm searches for subcarriers with lower utilization among neighboring ONUs within the PON, freeing up some or all of these resources and allocating them to ONUs with higher bandwidth requirements to meet the bandwidth needs of higher-priority ONUs.

[0049] ② Sharing of neighboring subcarrier resources across PONs

[0050] If the released resources within the same PON still cannot meet the ONU's bandwidth needs, a cross-PON subcarrier resource sharing mechanism is considered. The algorithm dynamically uses low-utilization subcarrier resources in adjacent PONs, temporarily releasing them and reallocating them to PONs with greater bandwidth requirements to optimize overall network resource utilization.

[0051] ③ONU service migration

[0052] The system searches for adjacent low-priority ONUs and checks whether their subcarriers can be migrated as a whole to other subcarriers or further away. This frees up the adjacent subcarrier resources occupied by the ONU for use by the current high-priority ONU. The subcarrier with the lowest total ONU real-time bandwidth demand or a subcarrier with zero ONUs is identified as an idle subcarrier, and the ONUs on the idle subcarrier are migrated to other designated subcarriers. During migration, ONUs with lower real-time bandwidth demands are prioritized until the bandwidth utilization of the target subcarrier is no less than 50% of the subcarrier capacity. When migrating ONUs, the target subcarrier should have sufficient remaining capacity to accommodate the real-time bandwidth demands of the migrating ONUs. Service migration ensures more sufficient and timely resource supply for high-priority ONUs, significantly improving network Quality of Service (QoS). Furthermore, by selecting subcarriers with the lowest bandwidth demands or completely idle subcarriers as migration sources, significant interference or interruption to other services is avoided, and the target subcarrier load is ensured to not exceed a reasonable threshold (50%). This not only improves the success rate of migration operations, but also reduces network performance fluctuations and latency issues caused by frequent migrations.

[0053] The advantage of the adjacent subcarrier activation and deactivation strategy is that it can aggregate subcarriers into continuous subcarrier frequency bands, which is conducive to using coherent detection at the receiving end to detect the information on the subcarrier frequency band at one time from the optical frequency.o +0.5×(f start +f end ), which passes and carries (f start , f end ) range of sub-carrier frequency band signal optical carrier f o Beat frequency, can be detected at one time (f start , f end ) range. By dynamically evaluating the utilization of subcarriers, subcarriers below a certain threshold can be released in a timely manner, allowing these resources to be reallocated to ONUs with higher demands, thereby quickly responding to changes in bandwidth demand and improving the flexibility and responsiveness of network resources. In addition, the use of neighbor ONU resource sharing can not only achieve local bandwidth optimization within a single PON, but also flexibly schedule idle resources across PONs, reducing the risk of service quality damage and service interruption caused by insufficient bandwidth, and reducing the cost and complexity of network resource reallocation. This step can also optimize overall network performance, reduce unnecessary service migration and subcarrier reconstruction operations, and improve network service continuity and user experience stability.

[0054] (6) Global subcarrier reconfiguration (step S6): The system finds the least utilized subcarrier from the global subcarriers and reconfigures it or preempts the bandwidth of low-priority ONUs to reduce the overhead caused by frequent global changes. After completing this reconfiguration, it attempts to allocate bandwidth to the current ONU. If the allocation is still not satisfied, it waits for the next cycle to allocate bandwidth. If the global reconfiguration still fails, it indicates that the network resources have reached a bottleneck. Further allocations in this cycle are abandoned and scheduling is resumed in the next cycle.

[0055] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof are implemented.

[0056] A computer-readable storage medium stores computer instructions, which, when executed by a processor, implement the multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof.

[0057] Compared with the existing technology, the advantages of the present invention are as follows: (1) Strong network expansion capability. It can achieve an increase in the number of PONs and optical network units (ONUs) by adding subcarriers. When the number of PONs or users increases, network operators do not need to make large-scale and complex modifications to the network architecture. By flexibly adjusting subcarrier resources, more PONs or users can be connected. Compared with the traditional passive optical network, which requires adding optical transceivers and wavelength routers to add PONs, and the problem of optical fiber resource shortage and limited splitting ratio when adding ONUs, the solution based on digital subcarriers provides a convenient way to expand network capacity; (2) ONUs assigned different subcarriers do not need to be synchronized on the uplink, which simplifies the system design and operation mechanism. In traditional passive optical networks, all ONUs need to achieve time synchronization and ranging with the OLT to ensure that the signals sent by different ONUs do not conflict on the uplink. The implementation process is cumbersome and easily interfered by environmental factors. In a passive optical network based on digital subcarrier multiplexing, when each ONU uses an independent subcarrier to transmit data, the signal conflict problem is fundamentally avoided without the need for complex synchronization operations. This reduces the complexity of the system, equipment costs and maintenance difficulty; (3) The subcarriers allocated to the PON (or ONU) can be adaptively modulated based on the distance between the PON (or ONU) and the OLT. When the distance is close, high-order modulation is used to improve the spectrum efficiency of each subcarrier, and when the distance is far, low-order modulation is used to improve transmission performance. This is impossible to achieve in traditional PON networks; (4) The proposed dynamic bandwidth allocation scheme prioritizes resource allocation for high-demand nodes through demand prediction and dynamic priority determination mechanisms, effectively solving the problem of insufficient priority guarantee in traditional static allocation. In medium and high load scenarios, it can significantly improve subcarrier utilization. At the same time, in high load scenarios, DSCM-DA can significantly reduce network latency compared to static allocation algorithms and has strong adaptability to load changes; (5) Compared with the solution of realizing multi-PON system with multiple wavelengths and laser arrays, the present invention adopts DSCM technology, so the multi-PON system only needs to use a single wavelength, requires a small number of lasers and optical receivers, and does not require additional wavelength switching and routing equipment. In summary, the present invention has advantages in PON and ONU expansion, simplified uplink synchronization, spectrum utilization, transmission performance, prioritized service quality, and device cost. It has great application potential and development prospects in building efficient, reliable, and scalable future access networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a diagram of the multi-PON system structure based on digital subcarrier multiplexing.

[0059] Figure 2 This is a flow chart of the bandwidth allocation algorithm for multi-PON systems based on digital subcarrier multiplexing technology.

[0060] Figure 3 is the average subcarrier utilization diagram under different subcarrier numbers,

[0061] Figure 4 This is a comparison chart of average subcarrier utilization under different strategies.

[0062] Figure 5 The trend chart of ONU service migration load and fixed distribution load changes under different normalized loads is shown.

[0063] Figure 6 is the average network delay diagram under different subcarrier numbers,

[0064] Figure 7 This is a comparison chart of average network delay under different strategies.

[0065] Figure 8 The following is a comparison chart of satisfaction factors in different business scenarios. DETAILED DESCRIPTION

[0066] The following is a further detailed description of an embodiment of the present invention in conjunction with the accompanying drawings, but this embodiment is not intended to limit the present invention. All similar structures and similar variations of the present invention should be included in the protection scope of the present invention.

[0067] Example: Multi-PON system based on DSCM technology (such as Figure 1 As shown in the figure, P2MP pluggable transceivers are deployed at the OLT, which divides the total 400G bandwidth into 16 25Gb / s subcarriers. Each subcarrier can independently transmit user data. According to the bandwidth requirements of each PON, one subcarrier (25Gbps) is allocated to PON1, and three (75Gbps), four (100Gbps) and eight (200Gbps) adjacent subcarriers are allocated to PON2, PON3 and PON4 respectively.

[0068] The dynamic bandwidth allocation algorithm for multiple PON systems using DSCM technology is shown in the following flowchart: Figure 2 As shown, it is divided into long-period multi-PON subcarrier scheduling and short-period ONU bandwidth allocation. Long-period multi-PON subcarrier scheduling performs the following operations:

[0069] Operation 1: Count the total bandwidth requirements of ONUs of different priorities on each PON

[0070] Count the total bandwidth requirements of high-priority ONUs of each PON separately

[0071]

[0072] and the total bandwidth requirements of low-priority ONUs

[0073]

[0074] They are used to determine the overall bandwidth requirements of each PON

[0075]

[0076] Where α is a weight coefficient used to control the impact of low-priority requests on subcarrier allocation. For example, if α = 0.6, a larger α value (0.6 ≤ α ≤ 1) will reserve more subcarrier resources, thereby reducing the frequency of subsequent inter-PON subcarrier changes.

[0077] Operation 2: Calculate and dynamically adjust the number of subcarriers for each PON

[0078] According to the bandwidth requirements of each PON, the number of subcarriers of each PON is dynamically adjusted. Assume that the system has N subcarriers and n PONs. PON j The total bandwidth requirement of PON j The number of subcarriers obtained N j The calculation is as follows:

[0079]

[0080] The number of subcarriers N of the last PON n The following formula will be used to adjust to ensure that all subcarrier resources are allocated:

[0081]

[0082] In short-cycle ONU bandwidth allocation, each ONU i , do the following:

[0083] Operation 1: Collect historical demand data and predict ONU bandwidth demand for the next cycle;

[0084] Collect ONU i The bandwidth demand of the previous several periods is predicted by the moving average algorithm. i Bandwidth requirements for the next time period

[0085]

[0086] Where W is the length of the moving average window, that is, the data of the past W periods are selected for averaging calculation; d i (t-τ) is the actual bandwidth requirement of the ONU in the past t-τ period.

[0087] Operation 2: Calculate ONU i Bandwidth demand satisfaction rate (DSR) i , and determine priorities;

[0088] Let the demand of the ONU i be d i , and calculate the demand satisfaction rate of this ONU according to the allocated bandwidth in the previous period and the predicted demand . If the DSR < T, mark the ONU i as a high-priority ONU, otherwise as a low-priority ONU. i

[0089] Operation 3: Allocate bandwidth to the ONUs in each PON according to the priority;

[0090] Allocate bandwidth within each PON according to the ONU priority. First, satisfy the predicted bandwidth demand of the high-priority ONUs. If the bandwidth resources are insufficient for allocation, release the subcarriers with low utilization rate of the neighboring ONUs in the same PON (migrate their services), and then perform the bandwidth allocation for this ONU; if the bandwidth resources are still insufficient after releasing subcarriers in the same PON, then consider the subcarrier allocation between different PONs.

[0091] Operation 4: Release subcarriers with low utilization rate of the neighboring ONUs in the same PON;

[0092] Check the subcarrier resource situation within the PON. When the bandwidth demand of a certain ONU cannot be directly satisfied, search for subcarriers with lower utilization rate among the neighboring ONUs in this PON, release some or all of the resources and allocate them to the ONUs with higher demand to meet the bandwidth demand of the high-priority ONUs.

[0093] Operation 5: Share the subcarrier resources of neighboring PONs;

[0094] If the ONU bandwidth demand still cannot be satisfied after releasing resources within the same PON, then further consider the subcarrier resource sharing mechanism across PONs. The algorithm dynamically uses the subcarrier resources with low utilization rate in neighboring PONs, temporarily releases and reallocates them to the PONs with greater bandwidth demand to optimize the utilization efficiency of the overall network resources.

[0095] Operation 6: Global subcarrier reconfiguration;

[0096] If the bandwidth demand of the current ONU still cannot be satisfied through the above steps, the algorithm will perform a wider range of subcarrier resource reconstruction. The specific measure is to search for subcarrier resources globally, preferentially release the subcarriers with the lowest utilization rate or preempt the resources serving low-priority ONUs to meet the bandwidth demand of high-priority or high-demand ONUs.

[0097] ​In order to verify the effectiveness of the above-mentioned DSCM technology PON system and its bandwidth allocation algorithm, simulation analysis was carried out based on key performance indicators such as average subcarrier utilization, latency, and ONU migration load, and compared with the traditional static allocation algorithm. In order to better simulate the network's bursty service needs, the bandwidth demand value of each ONU in each cycle is randomly generated using a Poisson distribution. Among them, the average value of the random number avg of the Poisson distribution is 1.5, and the range of random numbers is limited to 0-8 to control the amplitude of demand fluctuations. The randomly generated integer values ​​are normalized and mapped to the actual bandwidth demand range, thereby ensuring that the generated bandwidth demand has both burst randomness and remains within the reasonable carrying range of network resources. The specific simulation parameter settings are as follows:

[0098] Table 1 Simulation parameter settings

[0099]

[0100] like Figure 3 The figure shows the average subcarrier utilization for 4, 8, and 16 subcarriers. For 4 subcarriers, utilization rises rapidly as load increases, eventually reaching saturation. In contrast, utilization for 8 and 16 subcarriers exhibits a more gradual upward curve. When load is low, the system can more effectively allocate bandwidth to more subcarriers, ensuring that each subcarrier's resources are used effectively. As load increases, more subcarriers are activated, further improving overall resource utilization. Increasing the number of subcarriers enables efficient resource allocation even when load is low, avoiding excessive bandwidth concentration and enhancing system flexibility and scalability.

[0101] When the number of subcarriers is 16, the average subcarrier utilization under different strategies is as follows: Figure 4As shown, the DSCM-DA (Digital Subcarrier Multiplexing-Dynamic Allocation) curve rapidly reaches a high level and approaches saturation at low loads; in contrast, static subcarrier allocation improves more slowly, gradually approaching saturation at higher loads. Under all normalized load conditions, the average subcarrier utilization of the DSCM-DA scheme is significantly higher than that of the static subcarrier allocation scheme. At moderate loads (e.g., 0.4-0.8), DSCM-DA demonstrates a significant advantage over static subcarrier allocation, demonstrating its greater adaptability to load fluctuations and more efficient resource utilization. This is primarily due to the DSCM-DA algorithm's ability to dynamically adjust subcarrier allocation. When traffic bursts occur, it prioritizes searching for or reclaiming low-priority bandwidth resources in adjacent subcarriers and fully utilizes the remaining bandwidth for allocation, ultimately achieving relatively ideal uplink bandwidth utilization.

[0102] Define the sum of the data volume of all migrated ONUs in the tth period as the service migration load M(t) of the period, then:

[0103]

[0104] Among them, M i (t) represents the service data volume of the i-th ONU migration in the t-th period.

[0105] The average service migration load M of all cycles during the simulation is expressed as follows: where T is the total number of scheduling cycles.

[0106]

[0107] The changing trends of ONU service migration load and fixed distribution load under different load conditions are as follows: Figure 5 As shown in the figure, as the network load gradually increases, network resource utilization tends to be saturated. Some ONUs cannot meet their needs within the fixed allocated resources, triggering service migration to better utilize adjacent subcarrier resources, thereby improving the overall bandwidth utilization efficiency.

[0108] like Figure 6As shown, under the three different subcarrier configurations, average latency increases gradually with increasing normalized load. When the load is low, the latency differences between the three subcarrier configurations are small, demonstrating good system responsiveness and resource allocation efficiency under low-load conditions. As the load increases, the latency of the 16-subcarrier system increases more gradually, indicating that the system can effectively share the load with more subcarrier resources, thereby reducing system congestion. However, the latency of the 4-subcarrier system increases more rapidly, indicating a more significant latency issue. This is because when the number of subcarriers is small, system resources are relatively scarce. When the load increases, subcarrier resources are easily preempted, leading to bandwidth allocation bottlenecks and significantly increasing latency. Overall, increasing the number of subcarriers can help alleviate latency issues under high load conditions, but adding too many subcarriers can also lead to underutilization. Therefore, a proper balance between resource allocation and latency control is required in different situations.

[0109] The average network delay performance of DSCM-DA and traditional static allocation algorithm under different network loads is as follows: Figure 7 As shown in the figure, as network load gradually increases, the network latency of the traditional static allocation algorithm climbs rapidly under higher loads, showing a significant trend of latency degradation. In contrast, the DSCM-DA proposed in this invention, due to its use of digital subcarrier multiplexing technology, can flexibly and dynamically adjust subcarrier allocation based on actual network needs. This can cope with changes in bandwidth requirements under high load conditions and better maintain the network's latency performance advantages even when the network load increases.

[0110] In order to more comprehensively reflect the rationality of the overall bandwidth allocation of the system and the user experience, this bandwidth allocation algorithm introduces the average bandwidth demand satisfaction A avg Evaluate bandwidth resource allocation. The average bandwidth demand satisfaction is defined as the weighted average of each ONU's bandwidth demand satisfaction:

[0111]

[0112] Where n is the total number of ONUs, ω i Indicates the weight corresponding to the i-th ONU, reflecting its service priority or importance. And it meets the weight normalization conditions:

[0113]

[0114] η i It represents the bandwidth requirement satisfaction of the i-th ONU, that is, the ratio between the actual allocated bandwidth and the requested bandwidth:

[0115]

[0116] When the average bandwidth demand satisfaction is close to 1, it means that the overall bandwidth resource allocation of the system is ideal.

[0117] The simulation results of satisfaction factors under different business scenarios are as follows: Figure 8 As shown, in 5G backhaul scenarios, the average bandwidth demand satisfaction of the traditional static allocation strategy is only around 0.62, demonstrating its inability to handle highly dynamic and fluctuating service demands. In contrast, the DSCM-DA strategy maintains an average bandwidth demand satisfaction of around 0.98. This demonstrates that in scenarios with high network performance and real-time requirements, the dynamic prediction and flexible scheduling capabilities of DSCM-DA significantly improve network resource allocation efficiency, ensuring stable and reliable bandwidth for high-priority services. In enterprise user scenarios, given the high bandwidth demands of enterprise users, static bandwidth allocation methods struggle to detect fluctuations in high-demand services such as video conferencing and cloud office services, making it difficult to dynamically and flexibly allocate resources. This results in an average bandwidth demand satisfaction of only 0.7. The DSCM-DA strategy, on the other hand, performs exceptionally well, maintaining an average bandwidth demand satisfaction rating above 0.9. This is primarily because the dynamic bandwidth allocation mechanism can promptly predict and respond to fluctuations in enterprise users' bandwidth demands, enabling rapid resource responsiveness and effectively improving enterprise user satisfaction with network service quality. In residential user scenarios, the traditional static subcarrier allocation strategy achieved an average bandwidth demand satisfaction of approximately 0.75, demonstrating good service performance. The DSCM-DA strategy, on the other hand, achieved an average bandwidth demand satisfaction of 0.92, significantly improving the network experience for residential users. This demonstrates that the DSCM-DA strategy can effectively utilize idle resources in relatively stable service scenarios, further improving the user experience.

[0118] In summary, the example of the present invention adopts a multi-passive optical network system based on digital subcarrier multiplexing and its dynamic bandwidth allocation method, which solves the problems of low spectrum utilization and insufficient flexibility of traditional optical networks by subdividing the spectrum of a single optical carrier into multiple independently adjustable digital subcarriers. The supporting algorithm realizes efficient resource scheduling by prioritizing demand satisfaction rate, load prediction and sharing neighboring subcarrier resources within the same PON and across PONs in the adjacent subcarrier release strategy, thereby improving spectrum utilization, reducing latency and optimizing costs in multi-service convergence scenarios. Simulation verification shows that the scheme is significantly superior to traditional static allocation schemes in key indicators such as subcarrier utilization, average latency, and service satisfaction, providing an efficient and flexible solution for the next generation of optical access networks.

[0119] It is understood from common technical knowledge that the present invention may be implemented by other embodiments that do not depart from its spirit or essential features. Therefore, the embodiments disclosed above are, in all respects, merely illustrative and not exclusive. All modifications within the scope of the present invention or equivalent to the scope of the present invention are intended to be encompassed by the present invention.

Claims

1. A multi-passive optical network system based on digital subcarrier multiplexing and a dynamic bandwidth allocation method thereof, characterized in that: The following steps are involved: Step S1: predict bandwidth requirements for each ONU in the multi-PON, calculate the demand satisfaction rate and determine the priority; Step S2: Count the bandwidth requirements of each PON; Step S3: Calculate and dynamically allocate a certain number of subcarriers to each PON; Step S4: Dynamically allocate bandwidth to each ONU in the PON based on priority; Step S5: activating / deactivating neighbor subcarriers; Step S6: Perform global subcarrier reconfiguration.

2. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1 is characterized in that: Step S1 performs ONU priority determination, load prediction, and subcarrier scheduling, as follows: Demand Satisfaction Rate Calculation and Dynamic Priority Determination: Based on the ratio of the bandwidth actually allocated to the ONU to the predicted demand, the demand satisfaction rate (DSR) of each ONU is calculated. For the i-th optical network unit (ONU), if the DSR i <T (T is the threshold), indicating that the bandwidth demand satisfaction rate of ONU in the previous cycle is low, then the ONU i Mark as high priority; otherwise mark as low priority, DSR i is defined as follows: in Indicates ONU i The bandwidth actually allocated in the previous period, For ONU i In the forecast demand of the previous cycle, Load forecasting: Collect the demand of ONU in the previous several periods and predict the bandwidth demand of ONU in the next period through the moving average algorithm. as follows, Where W is the length of the moving average window, that is, the data of the past W periods are selected for averaging calculation; d i (t-τ) is the actual bandwidth requirement of the i-th ONU in the past t-τ period, is the predicted bandwidth requirement of the i-th ONU in the next cycle; Subcarrier scheduling: Prioritizes subcarrier allocation for high-priority ONUs. When resources are insufficient, adjacent subcarrier release, service migration, or global reconfiguration strategies are implemented.

3. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1, characterized in that: Step S2 counts the bandwidth requirements of each ONU by priority and weights them to determine the bandwidth requirements of each PON, as follows: Count the total bandwidth requirements of each PON's high-priority ONU: Total bandwidth requirement of low-priority ONUs: Used to determine the overall bandwidth requirements for each PON: Where α is the weight coefficient, which is used to control the impact of low-priority requirements on subcarrier allocation. When a larger α value is set (0.6≤α≤1), more subcarrier resources will be reserved, thereby reducing the frequency of subsequent inter-PON subcarrier changes.

4. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1, characterized in that: Step S3 dynamically adjusts subcarrier allocation according to the bandwidth requirements of each PON, as follows: Assume that the system has N subcarriers and n PONs, then PON j The number of subcarriers obtained N j as follows: in, PON j The total bandwidth requirement, The number of subcarriers N of the last PON n The following formula will be used to adjust to ensure that all subcarrier resources are allocated:

5. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1, characterized in that: The algorithm for dynamically allocating bandwidth to each ONU in the PON based on priority in step S4 is as follows: After distinguishing high-priority and low-priority ONUs by demand satisfaction rate, bandwidth is allocated to high-priority ONUs first, and then to low-priority ONUs. When allocating bandwidth to ONUs, the current ONU is calculated. i Unmet needs r i (Initially set to d i ), find the subcarrier k with remaining capacity in the subcarrier set, and perform a bandwidth allocation δ=min(r i , remaining capacity), update r i ←r i -δ and the available capacity of the subcarrier; if r i If it is still greater than 0, the next subcarrier will be allocated. When the bandwidth allocated to the high-priority ONU is insufficient, bandwidth can be released from the neighboring low-utilization subcarrier to meet the demand.

6. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1, characterized in that: Step S5 implements the adjacent subcarrier activation / deactivation strategy, as follows: Calculate the subcarrier utilization and release the subcarriers whose utilization is lower than the reference threshold. Let u k is the utilization rate of subcarrier k, then where b i，k It is allocated to the ONU on subcarrier k i Bandwidth, C k is the maximum capacity of the subcarrier, if u k If the subcarrier is lower than the reference threshold U, it is considered that the subcarrier can be partially or completely released to other ONUs with greater bandwidth requirements. When the subcarrier is released, the original ONU services on the subcarrier will be migrated. (1) Neighboring ONU subcarrier resource sharing within the same PON First, the subcarrier resource situation is checked within a single PON. When the bandwidth demand allocated to a certain ONU cannot be directly met, the algorithm will search for subcarriers with lower utilization among neighboring ONUs within the PON, release some or all of the resources, and allocate them to ONUs with higher demand to meet the bandwidth needs of high-priority ONUs. (2) Cross-PON neighbor subcarrier resource sharing If the internal resources of the same PON still cannot meet the ONU bandwidth requirements after being released, the cross-PON subcarrier resource sharing mechanism is further considered. The algorithm dynamically uses the subcarrier resources with low utilization in the adjacent PON, temporarily releases them and reallocates them to the PON with greater bandwidth requirements, so as to optimize the utilization efficiency of the overall network resources. If the adjacent subcarriers are released successfully, the process returns to step S4 to continue allocating bandwidth to the ONUs to which bandwidth has not been allocated successfully.

7. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 6, characterized in that: Subcarrier activation / deactivation strategy, the ONU service migration strategy is as follows: Find adjacent low-priority ONUs and check whether their subcarriers can be migrated as a whole to other subcarriers or farther locations. Release the adjacent subcarrier resources occupied by the ONU for use by the current high-priority ONU. Determine the subcarrier with the lowest total ONU real-time bandwidth demand or the subcarrier with zero ONU number as an idle subcarrier, and migrate the ONUs on the idle subcarrier to other designated subcarriers. During migration, give priority to ONUs with smaller real-time bandwidth demands until the bandwidth utilization of the target subcarrier is no less than 50% of the subcarrier capacity. When migrating ONUs, the target subcarrier should have sufficient remaining capacity to carry the real-time bandwidth demand of the migrated ONU.

8. The multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof according to claim 1, characterized in that: Step S6 implements global subcarrier reconfiguration, specifically as follows: Find the subcarrier with the lowest utilization rate from the global subcarriers and reconfigure it or preempt the bandwidth of the low-priority ONU. After completing the reconfiguration, try to allocate bandwidth to the current ONU. If it still fails, it means that network resources are indeed in short supply. At this time, you can wait for the next cycle to allocate bandwidth. If the global reconfiguration still fails, it means that the network resources have reached a bottleneck. In this case, give up further allocation in this cycle and wait for the next cycle to schedule again.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof are implemented as described in any one of claims 1 to 8.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the multi-passive optical network system based on digital subcarrier multiplexing and the dynamic bandwidth allocation method thereof are implemented as described in any one of claims 1 to 8.

Images