Upgrade method for partial port wavelength conversion nodes in multi-band optical networks and optical communication devices
By upgrading nodes one by one in a multi-band optical network and evaluating the blocking rate, an upgrade sequence is generated. Combined with dynamic service simulation and cost constraints, the optimal number and location of wavelength conversion nodes for some ports are determined, solving the problem of balancing cost and performance in multi-band optical networks and achieving efficient network optimization and stability improvement.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU ETERN
- Filing Date
- 2025-10-21
- Publication Date
- 2026-07-31
AI Technical Summary
In existing multi-band optical networks, fully deploying traditional multi-band optical switching nodes results in high service blocking rates, while fully deploying partial port wavelength conversion multi-band optical switching nodes results in high deployment costs. Furthermore, when nodes need to be upgraded gradually, the question arises of how to determine the optimal number of partial port wavelength conversion nodes and the target deployment nodes to balance the overall network deployment cost and service blocking performance.
By upgrading nodes in a multi-band optical network one by one, evaluating service blocking rates, generating upgrade sequences, establishing a mapping relationship between the number of nodes and blocking rates through dynamic service simulation, and determining the optimal number and location of wavelength conversion nodes for some ports based on overall deployment cost constraints, the orderly upgrading of nodes is achieved.
It effectively balances the overall network deployment cost with the service congestion rate, avoids cost waste or performance deficiencies caused by blind upgrades, improves network performance stability, adapts to the actual scenario of gradual node upgrades, and provides efficient optimization support for multi-band optical networks.
Smart Images
Figure CN121691965B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communications, and in particular to a method for upgrading partial port wavelength conversion nodes in a multi-band optical network and an optical communication device. Background Technology
[0002] With the rapid development of bandwidth-intensive services such as video streaming and cloud computing, optical networks are facing unprecedented capacity demands. To meet these demands, multi-band transmission is considered an important network capacity expansion strategy due to its cost-effectiveness. However, to fully realize the potential of multi-band transmission, the core lies in deploying an adapted multi-band optical switching architecture and resolving its optimal deployment location within the network. Current research has explored various aspects of multi-band optical switching architectures.
[0003] A. Napoli et al. proposed, for example Figure 1 The traditional multi-band optical switching architecture shown (see paper: A. Napoli, N. Calabretta, JK Fischer et al., “Perspectives of multi-band optical communication systems,” in 2018 23rd Opto-Electronics and Communications Conference (OECC), IEEE, 2018, pp. 1-2. DOI: 10.1109 / OECC.2018.8730026) consists of core components including multi-band multiplexers, multi-band demultiplexers, amplifiers, and wavelength selective switches (WSSs) operating in different bands. This architecture relies on optical devices capable of supporting new bands and, limited by wavelength continuity requirements, cannot fully utilize the wavelength resources in optical fibers, resulting in low wavelength utilization and significant shortcomings in network performance.
[0004] To overcome the limitations imposed by wavelength continuity, H. Kawahara et al. proposed a wavelength-selective, band-switchable multi-band optical switching architecture (see paper: H. Kawahara, M. Nakagawa, T. Seki, and T. Miyamura, “Experimental demonstration of wavelength-selective band / direction-switchable multi-band OXC using an inter-band all-optical wavelength converter,” in 2020 European Conference on Optical Communications (ECOC), IEEE, 2020, pp. 1-4. DOI: 10.1109 / ECOC48923.2020.9333270). The key to this architecture lies in the use of an inter-band wavelength converter. However, this architecture requires high-port-count wavelength selective switches (WSSs), and the manufacturing of high-port-count WSSs currently faces technical challenges, directly limiting the scalability of the architecture and making it difficult to adapt to the needs of large-scale network deployments.
[0005] To address the shortcomings of the aforementioned architecture, previous proposals have included... Figure 2 The diagram shows a partial port wavelength conversion multi-band optical switching architecture. Compared to the architecture of H. Kawahara et al., this architecture significantly reduces the number of required WSS ports and wavelength converters while maintaining similar network performance, effectively controlling hardware costs. Its design includes two port types: one equipped with an inter-band wavelength converter, which improves switching flexibility and supports the creation of more optical channels, but reduces the generalized signal-to-noise ratio (GSNR) due to all-optical wavelength conversion; the other without a converter, which avoids GSNR degradation to support higher-capacity optical channels, but has lower switching flexibility due to wavelength continuity constraints. Therefore, while the optimal ratio of the two types of ports needs further optimization, the core problem with this architecture is that its cost is still higher than traditional multi-band optical switching architectures, failing to simultaneously meet the dual requirements of high flexibility and low cost.
[0006] F. Jin et al. further compared two schemes: deployment of all traditional nodes and deployment of all partial-port wavelength conversion nodes. The results verified the limitations of single-architecture deployment: when all traditional multi-band optical switching nodes are deployed in the network, although lower costs can be achieved, key indicators such as network blocking performance are poor due to wavelength continuity constraints; when all nodes in the network are deployed with partial-port wavelength conversion multi-band optical switching nodes, blocking performance can be significantly improved, but the overall cost increases significantly due to the need to equip them with inter-band wavelength converters.
[0007] In summary, existing multi-band optical switching technologies have significant drawbacks: First, a single architecture deployment cannot balance cost and performance; traditional architectures are low-cost but have poor performance, while partial-port wavelength conversion architectures offer superior performance but are expensive. Second, some architectures suffer from technical bottlenecks, such as the reliance of wavelength-selectable and band-switchable architectures on high-port-count WSSs, which is difficult to overcome due to current manufacturing limitations. Third, the core issue of hybrid deployment remains unresolved: how to determine the optimal upgrade strategy for target nodes that need to be upgraded to partial-port wavelength conversion nodes and the baseline nodes that need to be retained as traditional nodes is still unclear, making it impossible to achieve a globally optimal balance between blocking performance and overall construction cost in multi-band optical networks. Summary of the Invention
[0008] Therefore, the technical problem to be solved by this invention is to address the issue of how to determine the optimal number of partial port wavelength conversion nodes and the target deployment nodes in a multi-band optical network, in order to balance the overall network deployment cost and service blocking performance, when the deployment of traditional multi-band optical switching nodes with full deployment has a high service blocking rate and the deployment of partial port wavelength conversion multi-band optical switching nodes with full deployment has a high deployment cost and the nodes need to be upgraded gradually. Thus, this invention provides a method for upgrading partial port wavelength conversion nodes in a multi-band optical network and an optical communication device.
[0009] Specifically, the method for upgrading partial port wavelength conversion nodes in the multi-band optical network includes the following steps:
[0010] S1: Construct an initial multi-band optical network, and upgrade the original multi-band optical switching nodes in the initial multi-band optical network one by one to partial port wavelength conversion multi-band optical switching nodes; after each node is upgraded, the service blocking rate of the current network is evaluated to obtain the network service blocking rate corresponding to each original multi-band optical switching node after individual upgrade.
[0011] S2: Based on the service blocking rate of each original multi-band optical switching node after individual upgrade, sort all the original multi-band optical switching nodes according to the service blocking rate to obtain the node upgrade sequence;
[0012] S3: Based on the node upgrade sequence, the original multi-band optical switching nodes are upgraded to partial port wavelength conversion multi-band optical switching nodes in sequence; after each upgrade, dynamic service simulation is performed on the current network to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes in the current network and the corresponding service blocking rate.
[0013] S4: Based on the mapping relationship and combined with the overall deployment cost constraint of the multi-band optical network, determine the optimal number of partial port wavelength conversion multi-band optical switching nodes and the corresponding target deployment nodes that can balance network blocking performance and overall deployment cost, and complete the node upgrade.
[0014] In one embodiment of the present invention, in S3, the method for obtaining the mapping relationship between the number of multi-band optical switching nodes with partial port wavelength conversion and the corresponding service blocking rate in the current network is as follows:
[0015] S31: After each node upgrade, when a new service request s arrives, obtain all virtual links between its source and destination nodes, and query whether there is a virtual link l among all virtual links with remaining capacity greater than or equal to the bandwidth of the new service request s.
[0016] If it exists, the new service request s is carried through the virtual link l, the remaining capacity of the virtual link l is updated, and then the next service processing flow is entered, and step S36 is executed;
[0017] If it does not exist, proceed to step S32;
[0018] S32: Calculate a preset number of shortest paths between the source node and the destination node for the new service request s, and determine whether there is a shortest path that satisfies the wavelength continuity condition:
[0019] If yes, proceed to step S34; if no, proceed to step S33.
[0020] S33: For all shortest paths, determine whether there exists a path that satisfies the wavelength conversion condition between bands:
[0021] If so, proceed to step S34;
[0022] If not, mark service s as a blocked service, record the number of blocking times, and execute step S36;
[0023] S34: Include all optical channels that meet the conditions in step S32 or step S33 into the candidate set G, and perform the following operations on each candidate optical channel in the candidate set G:
[0024] Calculate the optical path transmission quality evaluation index value for each candidate optical channel, sort all candidate optical channels according to the optical path transmission quality evaluation index value, allocate a bearer channel for service s according to the sorting result, and execute step S35;
[0025] If the candidate set G is empty, mark service s as a blocked service, record the number of blocking times, and execute step S36.
[0026] S35: Based on the optical path transmission quality evaluation index value of the optimal optical channel, select the highest-order modulation format that satisfies the transmission quality margin, and establish a new virtual link between the source node and the destination node based on the channel capacity of this modulation format. Utilizing links Support service s, update the remaining capacity of the link;
[0027] S36: Repeat steps S31 to S35 until all service requests are processed, calculate the service blocking rate in the current network, and count the actual number of partial port wavelength conversion multi-band optical switching nodes in the current network. Pair the number of nodes with the corresponding blocking rate. After all original multi-band optical switching nodes have been upgraded, integrate all data pairs to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes and the service blocking rate.
[0028] In one embodiment of the present invention, in S32, the wavelength continuity condition is as follows:
[0029] Among the three shortest paths calculated for the service, there must be an available wavelength in the available optical fiber of each link along any path that can run through the entire path, and the transmitter of the source node and the receiver of the destination node of the path must be in an idle state and match the band and wavelength corresponding to the service.
[0030] In one embodiment of the present invention, in S33, the wavelength conversion condition between the bands is as follows:
[0031] Condition 1: There are unused fiber optic resources in each link along the path;
[0032] Condition 2: For wavelength resources on the optical fiber, if no inter-band wavelength conversion is required along the entire path, there is an idle wavelength that can run through all links of the path; if the path passes through a multi-band optical switching node with partial port wavelength conversion and requires inter-band wavelength conversion, the inter-band wavelength converter integrated in the node is in an idle state and can convert the idle wavelength of the current link into the idle wavelength of the band required by the subsequent link.
[0033] Condition 3: The transmitter at the source node that matches the transmission requirements of the service and the receiver at the destination node that matches the reception requirements of the service are both in an unoccupied state.
[0034] In one embodiment of the present invention, in S34, the method for calculating the optical path transmission quality evaluation index value for each candidate optical channel is as follows:
[0035] The generalized signal-to-noise ratio (SNR) of each candidate optical channel is calculated, which includes the calculation of the impact of spontaneous emission noise, nonlinear interference noise, stimulated Raman scattering noise, and wavelength conversion on the signal transmission performance of the optical channel.
[0036] In one embodiment of the present invention, in step S1, the method of evaluating the service blocking rate of the current network after each node upgrade is as follows: a dynamic service simulation evaluation method is adopted, in which the service arrival between each pair of nodes in the multi-band optical network follows a Poisson distribution, and the duration of each service follows an exponential distribution.
[0037] In one embodiment of the present invention, in S1, all nodes in the initial multi-band optical network consist of a multi-band multiplexer, a multi-band demultiplexer, an amplifier, and wavelength selection switches operating in different bands, and do not have wavelength conversion function.
[0038] In one embodiment of the present invention, in S1, the partial port wavelength conversion multi-band optical switching node includes two types of ports: a first type of port equipped with an inter-band wavelength converter and a second type of port not equipped with an inter-band wavelength converter.
[0039] In one embodiment of the present invention, in S4, the overall network deployment cost includes the device costs of multi-band multiplexers, multi-band demultiplexers, optical amplifiers of different bands, inter-band wavelength converters, and wavelength selection switches in the multi-band optical switching node.
[0040] This invention provides an optical communication device for implementing a method for upgrading partial port wavelength conversion nodes in a multi-band optical network, comprising: a control unit, a node unit, a service processing unit, and a simulation evaluation unit;
[0041] During the initialization phase, the control unit first configures the node units, initializing all nodes as traditional multi-band optical switching nodes. The control unit then issues upgrade commands to the original multi-band optical switching nodes in the node units one by one, switching them to partial port wavelength conversion nodes. After each switch, the simulation evaluation unit is triggered to calculate the network service blocking rate by collecting network service data in real time, and the results are fed back to the control unit for storage.
[0042] After all nodes have been individually upgraded and evaluated, the control unit sorts the network service blocking rates and generates a node upgrade sequence. According to the node upgrade sequence, batch upgrade instructions are issued to the node units one by one. After each upgrade is completed, the service processing unit automatically starts processing dynamic services. The simulation evaluation unit synchronously records the number of wavelength conversion nodes at some ports and the corresponding service blocking rates.
[0043] The control unit analyzes the data to determine the optimal number and location of upgrade nodes, issues the final configuration command to the node unit, completes the optimal upgrade of the device nodes, and achieves a balance between blocking performance and cost.
[0044] Compared with the prior art, the above-described technical solution of the present invention has the following advantages:
[0045] This invention generates a precise upgrade sequence by upgrading nodes one by one and evaluating their blocking rates. It then establishes a mapping relationship between the number of nodes and the blocking rate through dynamic service simulation. Finally, it determines the optimal deployment based on cost constraints. This approach accurately identifies key upgrade nodes, avoiding cost waste or performance deficiencies caused by blind upgrades. It effectively balances overall network deployment costs and service blocking performance, addressing the pain point of high blocking rates in traditional nodes while avoiding the drawback of excessively high costs associated with all port nodes. Furthermore, the dynamic simulation incorporates a generalized signal-to-noise ratio model, fully considering the impact of spontaneous radiated noise, nonlinear interference, and wavelength conversion on transmission quality, ensuring accurate blocking rate assessment and improving network performance stability. The upgrade process is orderly and controllable, adapting to real-world scenarios of gradual node upgrades, making it highly practical and providing reliable support for efficient optimization of multi-band optical networks. Attached Figure Description
[0046] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.
[0047] Figure 1 This is a schematic diagram of a traditional multi-band optical switching device.
[0048] Figure 2 This is a schematic diagram of a multi-band optical switching device with partial port wavelength conversion;
[0049] Figure 3 This is a flowchart illustrating a method for upgrading partial port wavelength conversion nodes in a multi-band optical network, provided in an embodiment of the present invention.
[0050] Figure 4 It is a COST239 standard network containing 11 nodes and 26 links;
[0051] Figure 5 This is a flowchart illustrating the process of dynamically simulating the current network using a single-hop traffic diversion algorithm after each upgrade, thereby obtaining the mapping relationship between the number of multi-band optical switching nodes with wavelength conversion at some ports in the current network and the corresponding service blocking rate.
[0052] Figure 6 The results are experimental findings from a simulation evaluation of the total network deployment cost and service blocking probability corresponding to changes in the number of multi-band optical switching nodes with wavelength conversion at some ports, under a dynamic service scenario where the service load between node pairs is 295 Erlang.
[0053] Figure 7These are the blocking probabilities of three different network configurations with combinations of multi-band optical switching nodes under different service loads.
[0054] Figure 8 This is a schematic diagram of the structure of an optical communication device provided in an embodiment of the present invention.
[0055] Explanation of reference numerals in the accompanying drawings: 100, control unit; 200, node unit; 300, service processing unit; 400, simulation evaluation unit. Detailed Implementation
[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.
[0057] Reference Figure 3 As shown, this invention provides a method for upgrading partial port wavelength conversion nodes in a multi-band optical network, specifically including:
[0058] S1: Construct an initial multi-band optical network, and upgrade the original multi-band optical switching nodes in the initial multi-band optical network one by one to partial port wavelength conversion multi-band optical switching nodes; after each node is upgraded, the service blocking rate of the current network is evaluated to obtain the network service blocking rate corresponding to each original multi-band optical switching node after individual upgrade.
[0059] S2: Based on the service blocking rate of each original multi-band optical switching node after individual upgrade, sort all the original multi-band optical switching nodes according to the service blocking rate to obtain the node upgrade sequence;
[0060] S3: Based on the node upgrade sequence, the original multi-band optical switching nodes are upgraded to partial port wavelength conversion multi-band optical switching nodes in sequence; after each upgrade, dynamic service simulation is performed on the current network to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes in the current network and the corresponding service blocking rate.
[0061] S4: Based on the mapping relationship and combined with the overall deployment cost constraint of the multi-band optical network, determine the optimal number of partial port wavelength conversion multi-band optical switching nodes and the corresponding target deployment nodes that can balance network blocking performance and overall deployment cost, and complete the node upgrade.
[0062] This embodiment is based on the actual deployment requirements of multi-band optical networks, and is based on an 11-node, 26-link COST239 standard network (such as...). Figure 4This document (shown) elaborates on the upgrade method for partial port wavelength conversion nodes in a multi-band optical network and the implementation process of the optical communication device. During implementation, four pairs of optical fibers are deployed between adjacent node pairs, covering the C+L+S three bands (each band has a bandwidth of 4.8 THz, divided into 32 channels under a 150 GHz WDM grid, with a channel transmission rate of 128 GBaud symbol rate). All device costs are normalized to the cost of the C-band erbium-doped fiber amplifier (EDFA), and the system reserves a 1.0 dB optical path transmission quality (QoT) margin to ensure the repeatability and practicality of the technical solution.
[0063] Further, in step S1, all nodes in the initial multi-band optical network consist of multi-band multiplexers, multi-band demultiplexers, amplifiers, and wavelength selection switches operating in different bands, and do not have wavelength conversion functionality. Specifically, all original multi-band optical switching nodes in the initial multi-band optical network are traditional multi-band optical switching nodes, with the following structure: Figure 1 As shown, the core components include: a multi-band multiplexer (to combine multi-band signals), a multi-band demultiplexer (to separate multi-band signals), an optical amplifier (to compensate for signal transmission loss), and a wavelength selective switch (WSS) operating in different C / L / S bands. This type of node lacks wavelength conversion capabilities, and service transmission is limited by wavelength continuity constraints, making it impossible to fully utilize fiber optic wavelength resources.
[0064] The original multi-band optical switching nodes in the initial multi-band optical network are upgraded one by one to partial port wavelength conversion multi-band optical switching nodes, including two types of ports: a first type of port equipped with an inter-band wavelength converter and a second type of port without an inter-band wavelength converter.
[0065] The first type of port can convert the wavelength of the current band to the corresponding wavelength of another band according to a specific pump wavelength, which can improve switching flexibility and support the creation of more optical channels, but the all-optical wavelength conversion will lead to the degradation of the generalized signal-to-noise ratio (GSNR) of the optical signal; the second type of port can avoid GSNR degradation to support higher capacity optical channels, but it is constrained by wavelength continuity and has lower switching flexibility.
[0066] In step S2, each node in the initial multi-band optical network is upgraded one by one. After the upgrade, other nodes remain traditional nodes to ensure that a single evaluation only reflects the performance impact of upgrading a single node. After each upgrade, dynamic service simulation is used to evaluate the current network's service blocking rate. The simulation parameters are set as follows:
[0067] Service arrival model: Service arrival between each pair of nodes follows a Poisson distribution with an arrival rate of λ.
[0068] Business duration: The duration of each business follows an exponential distribution with a mean of 1 / μ = 1.0;
[0069] Flow load definition: Flow load ρ = λ / μ = λ (unit: Erlang);
[0070] Service characteristics: The bandwidth of a single service is fixed at 200Gb / s, and service splitting is not allowed (i.e., a single service is carried through only one optical channel).
[0071] Simulation sample size: 10 in total 6 Each business request ensures the accuracy of the blocking rate statistics.
[0072] Through the above simulation, the network service blocking rate corresponding to each traditional node after individual upgrade is recorded. For example, the blocking rate of node A after individual upgrade is 1.2%, and the blocking rate of node B after individual upgrade is 1.5%, etc.
[0073] Based on the correspondence between individual node upgrades and blocking rates obtained in step S1, all 11 traditional nodes are sorted in ascending order of service blocking rate to generate a node upgrade sequence. The sorting logic is as follows: nodes with lower blocking rates after individual upgrades have a more significant effect on improving network performance and should be prioritized for upgrades.
[0074] In step S3, based on the node upgrade sequence generated in step S2, nodes are upgraded in batches (one node is upgraded each time, and the cumulative number of upgrades increases from 1 to 11). After each upgrade is completed, a single-hop traffic diversion algorithm is used to perform dynamic service simulation on the current network to obtain the mapping relationship between the number of multi-band optical switching nodes with wavelength conversion at some ports in the current network and the corresponding service blocking rate. The specific process is as follows: Figure 5 As shown:
[0075] S31: After each node upgrade, when a new service request s arrives, obtain all virtual links between its source and destination nodes, and query whether there is a virtual link l among all virtual links with remaining capacity greater than or equal to the bandwidth (200Gb / s) of the new service request s.
[0076] If it exists, the new service request s is carried through the virtual link l, the remaining capacity of the virtual link l is updated, and then the next service processing flow is entered, and step S36 is executed;
[0077] If it does not exist, proceed to step S32;
[0078] S32: Calculate a preset number (3) of shortest paths between the source node and the destination node for the new service request s, and determine whether there is a shortest path that satisfies the wavelength continuity condition:
[0079] If yes, proceed to step S34; if no, proceed to step S33.
[0080] S33: For all shortest paths, determine whether there exists a path that satisfies the wavelength conversion condition between bands:
[0081] If so, proceed to step S34;
[0082] If not, mark service s as a blocked service, record the number of blocking times, and execute step S36;
[0083] S34: Include all optical channels that meet the conditions in step S32 or step S33 into the candidate set G, and perform the following operations on each candidate optical channel in the candidate set G:
[0084] Calculate the optical path transmission quality evaluation index value for each candidate optical channel, sort all candidate optical channels according to the optical path transmission quality evaluation index value, select the candidate optical channel with the largest optical path transmission quality evaluation index value as the bearer channel for service s based on the sorting result, and execute step S35.
[0085] The method for calculating the optical path transmission quality evaluation index value for each candidate optical channel is as follows:
[0086] Calculate the generalized signal-to-noise ratio (GSNR) for each candidate optical channel, which includes calculating the spontaneous emission noise (ASE), nonlinear interference noise (NLI), stimulated Raman scattering noise (SRS), and the impact of wavelength conversion on the signal transmission performance of the optical channel.
[0087] If the candidate set G is empty, mark service s as a blocked service, record the number of blocking times, and execute step S36.
[0088] S35: Based on the optical path transmission quality evaluation index value of the optimal optical channel, select the highest-order modulation format that satisfies a 1.0dB transmission quality (QoT) margin. This embodiment supports four modulation formats: DP-QPSK, DP-8QAM, DP-16QAM, and DP-32QAM, with corresponding channel capacities of 400Gb / s, 600Gb / s, 800Gb / s, and 1000Gb / s, respectively. Based on the channel capacity of the selected modulation format, a new virtual link is established between the source node and the destination node of service s. Utilizing links Carry service s, update the remaining link capacity. For example, when using DP-8QAM format, the link capacity is 600Gb / s, and after carrying 200Gb / s of service, 400Gb / s remains; then execute step S36.
[0089] S36: Repeat steps S31~S35 until 10. 6 After all business requests have been processed, according to the formula... Calculate the current service blocking rate in the network and count the actual number of partial port wavelength conversion multi-band optical switching nodes in the current network. Pair the number of nodes with the corresponding blocking rate. After all the original multi-band optical switching nodes have been upgraded, integrate all the data pairs to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes and the service blocking rate.
[0090] Furthermore, in this embodiment, in S32, the wavelength continuity condition is as follows:
[0091] Among the three shortest paths calculated for the service, there must be an available wavelength in the available optical fiber of each link along any path that can run through the entire path, and the transmitter of the source node and the receiver of the destination node of the path must be in an idle state and match the band and wavelength corresponding to the service.
[0092] Furthermore, in this embodiment, in S33, the wavelength conversion condition between bands is as follows:
[0093] Condition 1: There are unused fiber optic resources in each link along the path;
[0094] Condition 2: For wavelength resources on the optical fiber, if no inter-band wavelength conversion is required along the entire path, there is an idle wavelength that can run through all links of the path; if the path passes through a multi-band optical switching node with partial port wavelength conversion and requires inter-band wavelength conversion, the inter-band wavelength converter integrated in the node is in an idle state and can convert the idle wavelength of the current link into the idle wavelength of the band required by the subsequent link.
[0095] Condition 3: The transmitter at the source node that matches the transmission requirements of the service and the receiver at the destination node that matches the reception requirements of the service are both in an unoccupied state.
[0096] Specifically, in S4, the overall network deployment cost includes the device costs of multi-band multiplexers, multi-band demultiplexers, optical amplifiers for different bands, inter-band wavelength converters, and wavelength selective switches (WSS) in the multi-band optical switching nodes (see Table 1 for specific equipment costs). The specific calculation basis is as follows:
[0097] Device cost sources: See the paper JF Ó. Ramos, L. Cancela, and J. Rebola, “Influence of the ROADM architecture on the cost-per-bit in C+L+S multi-band optical networks,” in 2023 23rd International Conference on Transparent Optical Networks (ICTON), IEEE, 2023, pp. 1-4, and normalize to the cost of C-band erbium-doped fiber amplifiers (EDFAs).
[0098] Band cost coefficient: L-band device cost is 20% higher than C-band cost, S-band device cost coefficient α=1.2, α ranges from 1.2 to 1.5;
[0099] Wavelength converter cost: The cost coefficient for an all-optical wavelength converter (AO-WC) is β=2, and the value of β ranges from 0.5 to 2;
[0100] Node configuration: Colorless and directionless (CD) uplink / downlink configuration is adopted, and the ratio of the number of uplink / downlink ports to the number of line-side ports is 0.5.
[0101] Table 1 Relative Equipment Cost Table
[0102]
[0103] Based on an 11-node, 26-link COST239 test network (with 4 pairs of optical fibers deployed between adjacent nodes, covering C+L+S bands), a simulation evaluation was conducted on the total network deployment cost and service congestion probability corresponding to changes in the number of multi-band optical switching nodes with wavelength conversion at some ports under a dynamic service scenario of 295 Erlang between node pairs. The results are as follows. Figure 6 As shown, the specific analysis is as follows:
[0104] As the number of partial-port wavelength conversion nodes increases, the total network cost shows a monotonically increasing trend (due to the higher cost of AO-WC and high-port WSS). This trend is mainly because partial-port wavelength conversion multi-band optical switching nodes use high-cost wavelength converters and high-port-count wavelength selection switches, which are much more expensive than traditional multi-band optical switching nodes.
[0105] The service congestion rate exhibits a trend of first decreasing and then increasing: the congestion probability reaches its lowest point when the number of partial port wavelength conversion nodes reaches four. Beyond this number, further increases in the number of partial port wavelength conversion nodes lead to an increase in the congestion probability. The reasons are as follows: When the number of partial port wavelength conversion nodes is small, an optical path passes through only a few wavelength converters, thus the wavelength converters have a relatively small impact on the generalized signal-to-noise ratio (SNR) of the optical path. Furthermore, ports equipped with wavelength converters can improve network resource utilization. However, when there are a large number of partial port wavelength conversion nodes in the network, more ports need to perform inter-band conversion. As the number of converters increases, the generalized SNR of the S-band and L-band deteriorates significantly, leading to a decrease in network congestion performance and consequently an increase in the congestion rate.
[0106] Therefore, in this embodiment, the optimal number of partial port wavelength conversion nodes is 4, corresponding to the first 4 nodes in the upgrade sequence (nodes C, D, E, and F). This scheme can balance blocking performance and network cost. Based on this result, three network configuration schemes were set up for comparison and verification, all based on a COST239 network and the same dynamic service parameters (load ρ=295 Erlang):
[0107] Option 1 consists of all traditional nodes (11 traditional nodes); Option 2 consists of all partial port nodes (11 partial port wavelength conversion nodes); Option 3 is the solution of this invention (4 partial port wavelength conversion nodes + 7 traditional nodes).
[0108] The blocking probabilities of three different network configurations with various multi-band optical switching node combinations under different service loads were further compared, and the results are as follows: Figure 7 As shown in the figure, the blocking rate of all configurations increases with the increase of service load between node pairs. Compared with the scheme where all nodes are traditional nodes or all are partial-port wavelength conversion nodes, the proposed node upgrade strategy achieves the lowest blocking rate. The reason for this improvement is that, unlike the scheme with all partial-port wavelength conversion nodes, this strategy only upgrades key nodes to partial-port wavelength conversion nodes, thereby avoiding unnecessary wavelength conversion and improving the generalized signal-to-noise ratio of S-band and L-band; and compared with the scheme with all traditional nodes, the introduction of wavelength conversion function enables more efficient utilization of wavelength resources and can establish more optical channels.
[0109] This embodiment fully presents the implementation process of a partial port wavelength conversion node upgrade method in a multi-band optical network through specific network configuration, step execution, device implementation, and performance verification. Those skilled in the art can reproduce this scheme in multi-band optical networks with other topologies based on the parameters and process of this embodiment.
[0110] like Figure 8As shown, the present invention also provides an optical communication device to realize the upgrade method of some port wavelength conversion nodes in the multi-band optical network, including: control unit 100, node unit 200, service processing unit 300 and simulation evaluation unit 400. The hardware of each unit adopts FPGA+ARM architecture and the software is developed based on Linux system.
[0111] During the initialization phase, the control unit 100 first configures the node units 200, initializing all nodes as traditional multi-band optical switching nodes. These nodes consist of multi-band multiplexers, multi-band demultiplexers, erbium-doped fiber amplifiers, and conventional wavelength selection switches, and adopt a colorless and directionless (CD) uplink / downlink configuration. Simultaneously, the device parameters of the node units 200 are configured, such as the operating wavelength range of the WSS (C-band 1530-1565nm, L-band 1565-1625nm, S-band 1460-1530nm) and the gain of the EDFA (to ensure balanced signal power across bands), thus completing the functional verification of the initial network topology.
[0112] During the individual upgrade phase, the control unit 100 sends upgrade commands to the original multi-band optical switching nodes in the node unit 200 one by one, switching them to partial port wavelength conversion nodes. After each switch, the simulation evaluation unit 400 is triggered to receive the mapping data between a single node and the network blocking rate fed back by the simulation evaluation unit 400, and generates a node upgrade sequence in ascending order of blocking rate.
[0113] Subsequently, the control unit 100 issues batch upgrade instructions sequentially according to the node upgrade sequence, receives feedback data from the simulation evaluation unit 400, determines the optimal number and location of nodes by combining the cost calculation model, and finally issues configuration instructions to lock the node status.
[0114] The node unit 200 receives instructions from the control unit 100 to realize hardware switching between traditional nodes and partial port wavelength conversion nodes. Traditional nodes only perform multiplexing, demultiplexing and WSS routing of multi-band signals; partial port wavelength conversion nodes can also realize inter-band wavelength conversion through the AO-WC of the first type of port.
[0115] The service processing unit 300 maintains a list of virtual links between source and destination nodes and updates the remaining link capacity in real time; it uses the single-hop traffic diversion algorithm (S31-S36) of this embodiment to process dynamic service requests, including shortest path calculation, wavelength continuity / conversion condition judgment, modulation format selection, etc.; it counts the number of blocked services and feeds it back to the simulation evaluation unit 400.
[0116] The simulation evaluation unit 400 collects network service data in real time, including service arrival time, duration, bearer link, and congestion status; calculates the current network service congestion rate based on the collected data and feeds it back to the control unit; it has a built-in generalized signal-to-noise ratio model to calculate the GSNR value of candidate optical channels in real time, providing a basis for modulation format selection for the service processing unit 300; and calculates the overall network deployment cost based on the number of some port wavelength conversion nodes in the node unit 200, combined with the device cost coefficients (α, β), and feeds it back to the control unit 100 for optimal decision-making, achieving a balance between congestion performance and cost.
[0117] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A method for upgrading partial port wavelength conversion nodes in a multi-band optical network, characterized in that, include: S1: Construct an initial multi-band optical network, and upgrade each of the original multi-band optical switching nodes in the initial multi-band optical network to a partially port wavelength conversion multi-band optical switching node; after each node is upgraded, evaluate the current network service blocking rate to obtain the network service blocking rate corresponding to each original multi-band optical switching node after individual upgrade. S2: Based on the service blocking rate of each original multi-band optical switching node after individual upgrade, sort all the original multi-band optical switching nodes according to the service blocking rate to obtain the node upgrade sequence; S3: Based on the node upgrade sequence, the original multi-band optical switching nodes are sequentially upgraded to partial port wavelength conversion multi-band optical switching nodes; after each upgrade, dynamic service simulation is performed on the current network to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes and the corresponding service blocking rate in the current network, including: S31: After each node upgrade, when a new service request s arrives, obtain all virtual links between its source and destination nodes, and query whether there is a virtual link l among all virtual links with remaining capacity greater than or equal to the bandwidth of the new service request s. If it exists, the new service request s is carried through the virtual link l, the remaining capacity of the virtual link l is updated, and then the next service processing flow is entered, and step S36 is executed; If it does not exist, proceed to step S32; S32: Calculate a preset number of shortest paths between the source node and the destination node for the new service request s, and determine whether there is a shortest path that satisfies the wavelength continuity condition: If yes, proceed to step S34; if no, proceed to step S33. S33: For all shortest paths, determine whether there exists a path that satisfies the wavelength conversion condition between bands: If so, proceed to step S34; If not, mark service s as a blocked service, record the number of blocking times, and execute step S36; S34: Include all optical channels that meet the conditions in step S32 or step S33 into the candidate set G, and perform the following operations on each candidate optical channel in the candidate set G: Calculate the optical path transmission quality evaluation index value for each candidate optical channel, sort all candidate optical channels according to the optical path transmission quality evaluation index value, allocate a bearer channel for service s according to the sorting result, and execute step S35; If the candidate set G is empty, mark service s as a blocked service, record the number of blocking times, and execute step S36. S35: Based on the optical path transmission quality evaluation index value of the optimal optical channel, select the highest-order modulation format that satisfies the transmission quality margin, and establish a new virtual link between the source node and the destination node based on the channel capacity of this modulation format. Utilizing links Support service s, update the remaining capacity of the link; S36: Repeat steps S31~S35 until all service requests are processed, calculate the service blocking rate in the current network, and count the actual number of partial port wavelength conversion multi-band optical switching nodes in the current network. Combine the actual number of nodes with the corresponding blocking rate to form a data pair. After all original multi-band optical switching nodes have been upgraded, integrate all data pairs to obtain the mapping relationship between the number of partial port wavelength conversion multi-band optical switching nodes and the service blocking rate. S4: Based on the mapping relationship and combined with the overall deployment cost constraint of the multi-band optical network, determine the optimal number of partial port wavelength conversion multi-band optical switching nodes and the corresponding target deployment nodes that can balance network blocking performance and overall deployment cost, and complete the node upgrade.
2. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S32, the wavelength continuity condition is as follows: Among the three shortest paths calculated for the service, there must be an available wavelength in the available optical fiber of each link along any path that can run through the entire path, and the transmitter of the source node and the receiver of the destination node of the path must be in an idle state and match the band and wavelength corresponding to the service.
3. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S33, the wavelength conversion conditions between the bands are as follows: Condition 1: There are unused fiber optic resources in each link along the path; Condition 2: For wavelength resources on the optical fiber, if no inter-band wavelength conversion is required along the entire path, there is an idle wavelength that can run through all links of the path; if the path passes through a multi-band optical switching node with partial port wavelength conversion and requires inter-band wavelength conversion, the inter-band wavelength converter integrated in the node is in an idle state and can convert the idle wavelength of the current link into the idle wavelength of the band required by the subsequent link. Condition 3: The transmitter at the source node that matches the transmission requirements of the service and the receiver at the destination node that matches the reception requirements of the service are both in an unoccupied state.
4. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S34, the method for calculating the optical path transmission quality evaluation index value for each candidate optical channel is as follows: The generalized signal-to-noise ratio (SNR) of each candidate optical channel is calculated, which includes the calculation of the impact of spontaneous emission noise, nonlinear interference noise, stimulated Raman scattering noise, and wavelength conversion on the signal transmission performance of the optical channel.
5. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In step S1, the method for evaluating the service blocking rate of the current network after each node upgrade is as follows: a dynamic service simulation evaluation method is adopted. In the dynamic service simulation, the service arrival between each pair of nodes in the multi-band optical network follows a Poisson distribution, and the duration of each service follows an exponential distribution.
6. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S1, all nodes in the initial multi-band optical network consist of a multi-band multiplexer, a multi-band demultiplexer, an amplifier, and wavelength selection switches operating in different bands, and do not have wavelength conversion function.
7. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S1, the partial port wavelength conversion multi-band optical switching node includes two types of ports: a first type of port equipped with an inter-band wavelength converter and a second type of port not equipped with an inter-band wavelength converter.
8. The method for upgrading partial port wavelength conversion nodes in a multi-band optical network according to claim 1, characterized in that, In S4, the overall network deployment cost includes the device costs of multi-band multiplexers, multi-band demultiplexers, optical amplifiers for different bands, inter-band wavelength converters, and wavelength selection switches in the multi-band optical switching nodes.
9. An optical communication device, characterized in that, The method for upgrading partial port wavelength conversion nodes in a multi-band optical network as described in any one of claims 1 to 8 includes: a control unit, a node unit, a service processing unit, and a simulation evaluation unit; During the initialization phase, the control unit first configures the node units, initializing all nodes as traditional multi-band optical switching nodes. The control unit then issues upgrade commands to the original multi-band optical switching nodes in the node units one by one, switching them to partial port wavelength conversion nodes. After each switch, the simulation evaluation unit is triggered to calculate the network service blocking rate by collecting network service data in real time, and the results are fed back to the control unit for storage. After all nodes have been individually upgraded and evaluated, the control unit sorts the network service blocking rates and generates a node upgrade sequence. According to the node upgrade sequence, batch upgrade instructions are issued to the node units one by one. After each upgrade is completed, the service processing unit automatically starts processing dynamic services. The simulation evaluation unit synchronously records the number of wavelength conversion nodes at some ports and the corresponding service blocking rates. The control unit analyzes the data to determine the optimal number and location of upgrade nodes, issues the final configuration command to the node unit, completes the optimal upgrade of the device nodes, and achieves a balance between blocking performance and cost.