Frequency spectrum allocation in a dynamic elastic optical network

By categorizing users and partitioning frequency spectrum into user-specific and shared zones, the method addresses the inefficiencies in dynamic elastic optical networks, enhancing spectrum allocation efficiency and reducing fragmentation.

WO2026009018A1PCT designated stage Publication Date: 2026-01-08UNIV DE VALPARAISO +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/IB2024/056512
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-07-03
Publication Date
2026-01-08

AI Technical Summary

Technical Problem

Existing dynamic elastic optical networks face challenges in efficiently allocating frequency spectrum while adhering to continuity and contiguity constraints, leading to spectrum fragmentation and suboptimal bandwidth utilization.

Method used

A method and system that categorize users based on bandwidth requirements and partition frequency spectrum into zones, using a 'jumping First Fit' approach for user-specific zones and 'traditional First Fit' for shared zones to ensure contiguity and continuity, with a shared zone available for all users.

Benefits of technology

This approach reduces frequency spectrum fragmentation and improves data transmission efficiency by adaptively allocating frequency spectrum, considering different traffic demands and Quality of Service (QoS) requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2024056512_08012026_PF_FP_ABST
    Figure IB2024056512_08012026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to a method and system for allocating a frequency spectrum (100) to users of a dynamic elastic optical network, taking into account contiguity and continuity constraints. The method comprises: storing, in a database (110), data of the optical network topology, such as interconnected nodes, links and capacities of each link; storing, in the database, user information, such as source and destination nodes, required traffic load and paths; classifying users into categories based on their bandwidth requirements in frequency slot units (FSU); partitioning the frequency spectrum of each link into areas (410, 420) associated with each user category; and allocating contiguous FSUs (411) or adjacent contiguous FSUs (412) in the area (410) corresponding to the user category.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] METHOD AND SYSTEM FOR ALLOCATING FREQUENCY SPECTRUM TO USERS OF A DYNAMIC ELASTIC OPTICAL NETWORK, TAKING INTO ACCOUNT CONTIGUITY AND CONTINUITY RESTRICTIONS

[0002] The present invention relates to dynamic elastic optical networks with wavelength continuity and contiguity constraints. In particular, the present invention consists of a method and a system for assigning frequency spectra to elastic optical networks, which improves the efficiency of frequency spectrum allocation to the connections.

[0003] BACKGROUND

[0004] In recent years, a new approach to network design, called elastic optical networks (EONs), has been proposed to improve the efficiency of wavelength division multiplexing (WDM) optical communications. In EONs, the frequency grid is divided into narrower frequency slot units (FSUs) than in WDM networks, allowing for better bandwidth allocation based on the needs of each connection. This narrow banding improves spectrum management compared to the fixed spectrum grid previously used in the state of the art.

[0005] One of the challenges in EONs is the efficient dynamic allocation of FSUs, which is subject to continuity and contiguity constraints. Continuity implies using the same FSUs throughout the connection path, while contiguity requires that the allocated FSUs be adjacent in the frequency spectrum. The presence of unused FSUs among the allocated ones is known as spectrum fragmentation, a problem that must be avoided to efficiently utilize the available bandwidth.

[0006] In elastic optical networks, the traffic pattern can be static or dynamic. In static operation, all required connections are known in advance, and resources are permanently allocated at the start of network operation. In contrast, in dynamic optical networks, resources are dynamically allocated and released, allowing the same resources to be used by different connections at different times. The present invention focuses on dynamic elastic optical networks.

[0007] In the state of the art, various types of technologies have been developed that aim to solve the problems indicated above. One of the relevant solutions is disclosed in document CN108616783B, which describes a method for activating a frequency spectrum arrangement based on the detection of the frequency spectrum in an elastic optical network.The method comprises the steps of: calculating a frequency spectrum availability and service demand before allocation; comparing the frequency spectrum availability before allocation with the service demand, and when the frequency spectrum availability before allocation is greater than or equal to the service demand, it is established that frequency spectrum allocation is not needed; when the frequency spectrum availability before allocation is less than the service requirement, calculating the ordered frequency spectrum availability; and comparing the ordered frequency spectrum availability with the service demand, and activating a frequency spectrum allocation mechanism when the ordered frequency spectrum availability is greater than or equal to the service demand.When the availability of the frequency spectrum after classification is less than the service requirement, do not perform the frequency spectrum classification.

[0008] Another relevant development is disclosed in US patent 2017 / 0054506 A1, which describes an optical node device, an optical network controller, and a method for controlling the optical network. The optical network controller includes a frequency region establishment means that establishes a plurality of frequency regions within a frequency band used in an optical network based on a dense wavelength division multiplexing system, using a flexible frequency network. The controller further comprises an optical path establishment means for establishing paths that have a common attribute in at least one of the plurality of frequency regions, and a frequency region control means for changing a frequency region's frequency width and instructing the frequency region configuration means to reconfigure a plurality of frequency regions.

[0009] While the previously cited prior art documents aim to increase the efficiency of frequency spectrum allocation by establishing modifiable bandwidth regions, a need remains, and there is still room for further efficiency improvements by seeking ways to achieve intelligent and efficient resource allocation according to user demand. The present invention seeks to provide such improvements in frequency spectrum allocation efficiency, taking into account contiguity and continuity constraints.

[0010] BRIEF DESCRIPTION OF THE INVENTION To overcome the aforementioned deficiencies, a method for allocating frequency spectrum to users of a dynamic elastic optical network is presented, taking into account contiguity and continuity constraints. The method comprises the following steps: obtaining data associated with an optical network topology, including information on interconnected nodes, links, and the capacities of each link; obtaining information on optical network users, related to the source and destination nodes, the traffic load demanded by each user, and the corresponding routes for each user; and classifying users according to categories, where the categories are established based on their bandwidth requirements, measured in the number of FSUs.Partitioning the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and allocating a shared zone of the frequency spectrum so that it can be used by any user regardless of their category; and allocating a set of FSUs upon a transmission request from a user, which includes checking the availability of an initial number of contiguous FSUs within the zone associated with the user category, on each of the links that make up the user's route, where said initial number of FSUs is equal to the number of FSUs required by the user, and, in the event that said initial number of contiguous FSUs are not available, checking the availability of a new number of contiguous FSUs, on each of the links that make up the user's route, where the new number of contiguous FSUs are adjacent to the initial number of FSUs.

[0011] In this way, using the methodology described, user categorization assigns a group of users requiring the same bandwidth to a specific category, measured in FSUs, while frequency spectrum partitioning establishes zones, each with a specific number of FSUs, to accommodate the requirements of users in a given category. Furthermore, the method includes maintaining a region of the frequency spectrum as a shared zone, meaning a number of FSUs available for use by any user regardless of their category.

[0012] In the method presented, during the spectrum allocation process, the query of available FSUs on each link is carried out using a procedure that can be understood as "Jumping First Fit." This involves determining that when a certain number of contiguous FSUs in the area designated exclusively for the user category making the transmission request are not available on any of the links in that user's route, the availability of a new number of contiguous FSUs is queried on each of the links in the user's route. This second group of FSUs is adjacent to the first, so the queried FSUs are offset by a number equal to the number of FSUs required. Thus, the proposed methodology efficiently allocates frequency spectrum while meeting the contiguity and continuity requirements.

[0013] Thus, the method presented here is a variant of the method known as "First Fit," with the difference that in the First Fit method, when a spectrum allocation request is rejected, the query is performed by shifting one FSU relative to the FSUs consulted. In contrast, in the method of the present invention, the query is performed for a set of contiguous FSUs, according to a defined quantity for each user category, so that when there is no availability for the first set, the query shifts to a set of FSUs adjacent to the previous set.According to this logic of operation, the traditional First Fit method can be understood as a “sliding First Fit” method since it slides one FSU at a time in each query, while the proposed methodology can be understood as a “jumping First Fit” because when it is determined that the FSUS of a certain set are not available, the query moves or “jumps” to a new set of FSUs, adjacent to the previous set.

[0014] The shared area is intended to handle queries where no available FSU sets were found in the respective area associated with the category for a given user, so that query is then routed to the shared area. In preferred embodiments of the invention, the methodology comprises the use of a traditional First Fit method in the shared area.

[0015] The present invention also includes a system for allocating frequency spectrum to users of a dynamic elastic optical network, taking into account contiguity and continuity constraints. The system comprises: a database that stores information related to a provider network, including a set of nodes and links; information on network users, including the source and destination nodes for each user and their required bandwidth, measured in FSUs; and information on the capacity of each network link, also measured in FSUs; and processing means that communicate operationally with the database to perform user categorization, link zoning, and spectrum allocation processes, which includes: classifying users according to categories based on their bandwidth requirements, measured in the number of FSUs for each user;Partitioning the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and allocating a shared zone of the frequency spectrum so that it can be used by any user regardless of their category; allocating a set of FSUs to a user upon a transmission request, which includes checking the availability of an initial number of contiguous FSUs within the zone associated with the user's category, on each of the links that make up the user's route, where this initial number corresponds to the number of FSUs required by the user, and, in the event that this initial number of contiguous FSUs is not available, checking the availability of a new number of contiguous FSUs, on each of the links of the user's route, which are adjacent to the initial number.

[0016] In summary, the method and system proposed in the present invention introduce a novel hybrid spectrum allocation algorithm designed for elastic dynamic optical networks (EONs), combining the concept of "jumping First Fit" in the frequency spectrum areas associated with the different categories and the use of "traditional or sliding First Fit" in the shared area. This provides adaptable and efficient data transmission, mitigating frequency spectrum fragmentation, considering different classes of traffic demands, and taking into account Quality of Service (QoS) requirements.

[0017] BRIEF DESCRIPTION OF THE FIGURES

[0018] Figure 1 shows a schematic representation of the assignment of user categories to respective frequency spectrum zones of each network link.

[0019] Figure 2 shows a diagram of a first exemplary configuration of the frequency spectrum allocation system of the present invention.

[0020] Figure 3 shows a schematic representation of an exemplary configuration of an optical network composed of a set of nodes and links.

[0021] Figure 4 shows a diagram of a second exemplary configuration of the frequency spectrum allocation system of the present invention.

[0022] Figures 5a to 5d show the different mesh topologies evaluated in the application example.

[0023] Figure 6 shows an exemplary configuration of the subdivision of a network link, according to the present invention.

[0024] Figures 7 to 18 show the comparative results obtained through the application example, comparing the probability of user blocking obtained through the traditional First Fit method and the method of the present invention.

[0025] DETAILED DESCRIPTION OF THE INVENTION

[0026] According to the exemplary configurations shown in the accompanying figures, the present invention comprises a method for allocating frequency spectrum to users of a dynamic elastic optical network, taking into account contiguity and continuity constraints. The method comprises the steps of: obtaining data associated with an optical network topology, including information on interconnected nodes, links, and the capacities of each link; obtaining information on users of the optical network, related to the source and destination nodes, the traffic load demanded by each user, and the corresponding routes for each user; and classifying users according to categories, where the categories are established according to their bandwidth requirements, measured in the number of FSUs.Partitioning the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and allocating a shared zone of the frequency spectrum so that it can be used by any user regardless of their category; and allocating a set of FSUs upon a transmission request from a user, which includes checking the availability of an initial number of contiguous FSUs within the zone associated with the user category, on each of the links that make up the user's route, where said initial number of FSUs is equal to the number of FSUs required by the user, and, in the event that said initial number of contiguous FSUs are not available, checking the availability of a new number of contiguous FSUs, on each of the links that make up the user's route, where the new number of contiguous FSUs are adjacent to the initial number of FSUs.

[0027] According to the example configuration in Figure 1, the step of classifying users into categories involves assigning a set of users requiring the same bandwidth to a specific category (c / ), where the bandwidth requirement is measured in FSUs. This establishes K categories (ci, C2,... Ck) that group all users of the optical network, with each category grouping users with the same FSU requirement. Furthermore, partitioning the frequency spectrum establishes an equivalent number of zones (F) in each link (E) of the optical network, where each zone is associated with a respective user category (F). c / ), and where each zone has a specific number of FSUs to accommodate the requirements of users in the corresponding category. Preferably, the number of FSUs in a given zone is a multiple of the number of FSUs required by users in that category.

[0028] The method also includes assigning a region of the frequency spectrum in the form of a shared zone (F c That is, an arbitrary number of FSUs are allocated for use by any user regardless of their category. In this way, the entire frequency spectrum of each link in the network is assigned to some type of category, either to one of the K zones associated with the categories or to the shared zone. Furthermore, each FSU in the network links is assigned to one and only one of the defined zones.

[0029] Preferably, the step of obtaining a set of FSUs for each user includes a spectrum allocation stage. That is, each time a user requests to transmit their data, the method determines whether the connection is accepted or rejected and, if accepted, determines which FSUs are allocated to handle that request on each of the links belonging to the user's route. Furthermore, each time a user disconnects because they have finished transmitting their data, the method includes the step of making available again the resources (FSUs) that were temporarily allocated to that user.

[0030] Preferably, the step of partitioning the frequency spectrum of each network link includes determining, for each network link, an identifier for each FSU of the link, including further establishing an identifier for the first FSU and the last FSU of each zone, an identifier for the first FSU and the last FSU of the shared zone, and a status for each FSU, either occupied or available.

[0031] Additionally, in preferred configurations of the invention, the step of obtaining a set of FSUs for each user includes: a) receiving a request from a user of a certain category (o / ), who has a requirement for n FSUs; b) determining, for each of the links in the user's path, an identifier

[0032] (Di) of the first FSU of the zone associated with the category (o / ), and evaluate if a first number of FSUs contiguous to the FSU D¡ is available, where the number of contiguous FSUs corresponds to the user's requirement of n FSUs (ci), that is, the FSUs Di, D¡+1, D¡+n¡-ic) in the case that in each link of the user's route the first number of FSUs contiguous to the identifier Di is available, assign the identified FSUs of each link to meet the user's demand, and store in a database information relating to the fact that the corresponding n¡ FSUs assigned are occupied; d) in the event that the FSUs consulted are not available in at least one link, consult the availability of the new quantity of contiguous FSUs, which includes determining an identifier of the FSUs (0 / + n¡) up to (Di + 2n¡ -1), so that the new quantity of contiguous FSUs is adjacent to the first quantity;e) In the event that it is again identified that in at least one link the new FSUs consulted are not available, the method is configured to successively repeat step (d) to consult the availability of a new number of adjacent FSUs within the FSUs assigned to the user's category.;

[0033] In preferred embodiments of the invention, in cases where successive queries are performed on a given subset regarding the availability of a certain number of FSUs, and it is determined that there are no contiguous FSUs available that can meet the requirement, the method then includes the additional step of querying the availability of that number of FSUs in the shared area. More preferably, the query for available FSUs in the shared area is carried out using a traditional First Fit method.

[0034] Furthermore, if the connection request cannot be handled in the common area either, the method includes the step of definitively rejecting the user's request.

[0035] It should be noted that the preceding description assumes that the route associated with a user is unique and fixed. However, the invention is not restricted to this condition, as it can operate even if more than one fixed route is possible. In this case, for each query, the method involves evaluating each possible route in parallel or serially and, if more than one route exists, selecting one according to an efficiency criterion. Additionally, the described method is compatible with any dynamic routing method; that is, when a request is made, the first route is found, and subsequently, the method proposed here is used to assign the requested FSUs. With reference to Figure 2, the present invention also provides a system for allocating frequency spectrum to users of a dynamic elastic optical network.which takes into account contiguity and continuity constraints. The system comprises: a database 110 that stores information related to a provider network 111, which includes a set of nodes and links; information on network users 112, which includes the source and destination nodes for each user and their required bandwidth, measured in FSUs; information on the capacity of each link in the network 113, measured in FSUs; information on frequency spectrum partitioning 114 and the routes associated with each user; and processing means 120 that communicate operationally with the database to carry out processes of user categorization, link partitioning, and FSU allocation, which includes: classifying users into categories according to their bandwidth requirements,measured in number of FSUs; partitioning the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and assigning a common zone of the frequency spectrum so that it can be used by any user regardless of their category; assigning a set of FSUs to a user upon a transmission request, which includes checking the availability of an initial number of contiguous FSUs within the zone associated with the user's category, on each of the links that make up the user's route, where this initial number corresponds to the number of FSUs required by the user, and, in the event that this initial number of contiguous FSUs is not available, checking the availability of a new number of contiguous FSUs, on each of the links of the user's route, which are adjacent to the initial number.

[0036] Preferably, after the allocation of resources in FUSs to a user, the processes of user categorization, link partitioning and FSU allocation include storing in the database information related to the partitioning of the frequency spectrum, which includes the subsets into which each link is divided, an identifier of a starting FSU of each subset and an identifier of an ending FSU of each subset, and an identifier of a starting FSU and an identifier of an ending FSU of a common zone.

[0037] More particularly, according to the exemplary configuration of Figure 2, the system 100 can preferably be run in an environment comprising: an input interface 130, the processing means 120 already described, an output interface 140, and the database 110, which includes information related to a provider's network topology 111, information on network users 112, information on the routes and capacity of each network link 113, and the partitioning of each network link 114. In some configurations of the invention, the information relating to the routes assigned to each user 115 can be stored in a separate database, as shown in Figure 2, or it can be stored in the same database 110.

[0038] Preferably, the frequency spectrum allocation system 100 may include the use of one or more servers, or a computing device capable of communicating with other devices, such as a desktop or laptop computer, outside the provider's network. Furthermore, it preferably includes the use of a communication subsystem (not shown in Figure 2) for communication between its various elements.

[0039] According to Figure 3, the provider's network 200 is an optical network, composed of a set of nodes 210 interconnected by a set of fiber or multi-fiber optical links 220. The network nodes 210 can be of any type of dynamic optical network.

[0040] Returning to Figure 2, in the frequency spectrum allocation system 100, when a user requests a connection, the system accepts the request through the input interface 130. The system then queries database 110 to request: network topology data, user data, link routes and capacity, and link partitioning information. With this information, the processing equipment 120 performs the processes of user categorization and frequency spectrum partitioning for each link, according to the previously described methodology, and is configured to check, for each link in the user's route, whether the FSUs of the category requested by the user are marked as available in database 110.If successful, the FSUs are assigned to each link and the information is stored in database 110, indicating the user and the respective FSUs assigned on each network link. This includes storing the start and end identifiers of the assigned FSUs. Similarly, the FSUs assigned to each link in the route are marked as occupied in the database. If the assignment fails on any of the network links, the system rechecks the FSUs in the user's assigned zone on each link in their route, using the information from the database.

[0041] Once the frequency assignment for each user has been made, this information is stored in the database of routes assigned to each user 115, and this information can be delivered to any other system or device using the output interface 140.

[0042] According to Figure 4, in alternative configurations of the invention, the frequency spectrum allocation system 100 can be subdivided into multiple subsystems, including a network topology information subsystem 320, a user information subsystem 330, a link routing and capacity information subsystem 340, a link partitioning information subsystem 350, and a frequency allocation subsystem 370. In this configuration, each subsystem can comprise a separate server, or a network device inside or outside the provider's network, including an input / output interface and its own database (not shown in the figures).

[0043] Thus, in the configuration shown in Figure 4, the frequency spectrum allocation system 100 comprises a network from provider 310, whose structure and link capacities are stored in the network topology information subsystem 320. The user information subsystem 330 contains the set of users, including information about the source and destination node pairs of provider 310's network, and the traffic load in FSUs demanded by each user. Furthermore, the routing and link capacity information subsystem 340 is responsible for determining routes for the network users.Each of these subsystems communicates operationally with the provider's network 310, as shown in Figure 4, and in turn communicates with a communications subsystem 360, which allows the reception of connection requests from users and allows communication between the different subsystems to carry out the steps of the method of the present invention.

[0044] Once the steps of the frequency spectrum allocation method have been executed, the link partitioning information subsystem 350 stores the partitioning or zoning information according to the different defined zones and the common zone, including the identifier of each FSU, the identifiers of the starting and ending FSUs for each zone, and the identifiers of the starting and ending FSUs for the common zone. To perform the frequency allocation, the frequency allocation subsystem 370 uses the provider's network routing information 310, stored in the routing information subsystem 340, the user information and their demands stored in the user information subsystem 330, and the information from the link zoning information subsystem 350, and can send or receive information from other systems via the communication subsystem 160.

[0045] Preferably, in this configuration each subsystem comprises: an input interface that receives a data request from the 360 ​​communication subsystem, a processor that processes said data request in order to retrieve the requested data from its respective database, and an output interface for sending the requested information. The database of each subsystem can preferably store the respective data in the form of a table.

[0046] The 360 ​​communication system can be any network system that allows two or more devices to be connected, such as a cellular network, a public land mobile network (PLMN), a second generation (2G) network, a third generation (3G) network, a fourth generation (4G) network, a long-term evolution (LTE) network, a fifth generation (5G) network, a code division multiple access (CDMA) network, a global system for mobile communications networks (GSM), a general package of radio services (GPRS), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), an ad hoc network, an intranet, the Internet, a fiber optic-based network, a satellite network, a television network, or a mixture of one or more of these systems.

[0047] APPLICATION EXAMPLE

[0048] The method and system of the present invention were modeled using a custom simulator developed for this purpose, employing four network topologies, as shown in Figure 5a-d. To evaluate the performance of the proposed method, a random sequence of multiple transmission requests from all users with a homogeneous traffic load (p = 0.3, op = 0.5) was used. Furthermore, for the purposes of this document and without loss of generality, we assume that all links in the network have the same capacity, i.e., K(i, j) = K for all (i, j) ≤ L.

[0049] Figure 6 shows an exemplary configuration of the subdivision of a 400 link of the network, where users are classified into two exemplary categories, according to their FSU requirements, particularly, in this example, class 2 users who have a demand of 2 FSUs and class 3 users who have a demand of 3 FSUs are defined. In this way, the network link is subdivided into two subsets 410 and 420, where the first subset 410 is intended for class 3 users and comprises 6 FSUs in total, thus having two groups of 3 contiguous FSUs 411 and 412, for class 3 users. The second subset 420 is intended for class 2 users and comprises the same 6 FSUs, but grouped into three groups of two contiguous FSUs, 421, 422 and 423. The remaining available FSUs of the link are allocated to the common area 430, which operates using a traditional First Fit method, with each FSU being considered individually 431.

[0050] Thus, experiments were conducted with 2 and 3 user classes, with bandwidth requirements of W = {3, 4} and W = {2, 3, 5} FSU, respectively. The experiments determined that the class to which a specific user (s, d) belongs depends on the length of their path, according to three possible distributions:

[0051] • Ordered users: users are ordered from shortest to longest path. For the 2-class scenario with W = {3, 4}, the first half of the users are in class 3 and the second half are in class 4. For the 3-class scenario with W = {2, 3, 5}, the first third of the list is in class 2, the second third is in class 3, and the last third is in class 5.

[0052] • Inverted users: users are ordered from shortest to longest path. For the 2-class scenario with W = {3, 4}, the first half of the users are in class 4 and the second half are in class 3. For the 3-class scenario with W = {2, 3, 5}, the first third of the list is in class 5, the second third is in class 3, and the last third is in class 2.

[0053] • Uniform users: users are ordered from shortest to longest path. For the 2-class scenario with W = {3, 4}, users in odd positions are in class 3 and those in even positions are in class 4. For the 3-class scenario with W = {2, 3, 5}, the first user in the list is in class 2, the second is in class 3, the third is in class 5, the fourth is in class 2, the fifth is in class 3, the sixth is in class 5, and so on, assigning classes to all users in the list.

[0054] Figure 5 shows the evaluated mesh topologies, with Figure 6a showing an 11-node, 50-link EuroCore topology, Figure 6b showing a 21-node, 78-link UKNet topology, Figure 6c showing a 20-node, 78-link elastic optical network topology, and Figure 6d showing a 43-node, 180-link EuroLarge topology. The method of the present invention was thus evaluated and compared with the traditional First-Fit method, which is the most widely used method in the literature and generally exhibits better overall performance. Since the proposed method includes parameters specifying the number of blocks reserved for each user class, these parameters were set to the configuration that showed the best blocking probability performance. Another spectrum allocation method, Prime-Partition SA, where all FSUs are reserved, was also evaluated.The Prime-Partition method also has parameters that specify the number of blocks reserved for each class, and for comparison, the configuration of these parameters that showed the best performance in terms of blocking probability was selected.

[0055] The comparative results are shown in Figures 7 to 18, where the results obtained using the traditional First Fit method are shown in white columns, and the results obtained using the method of the present invention are shown in gray columns.

[0056] Figure 7 shows the probability of blocking users with a traffic load p = 0.3 and require W = {3; 4} FSUs with a uniform bandwidth requirement distribution.

[0057] Figure 8 shows the probability of blocking users with a traffic load p = 0.3 and require W = {3; 4} FSUs with an ordered bandwidth requirement distribution.

[0058] Figure 9 shows the probability of blocking users with a traffic load p = 0.3 and require W = {3; 4} FSUs with inverted bandwidth requirement distribution.

[0059] Figure 10 shows the probability of blocking users with a traffic load p = 0.5 and require W = {3; 4} FSUs with a uniform bandwidth requirement distribution.

[0060] Figure 11 shows the probability of blocking users with a traffic load p = 0.5 and require W = {3; 4} FSUs with an ordered bandwidth requirement distribution.

[0061] Figure 12 shows the probability of blocking users with a traffic load p = 0.5 and requiring W = {3; 4} FSUs with an inverse bandwidth requirement distribution. Figure 13 shows the probability of blocking users with a traffic load p = 0.3 and requiring W = {2; 3; 5} FSUs with a uniform bandwidth requirement distribution.

[0062] Figure 14 shows the probability of blocking users with a traffic load p = 0.3 and require W = {2; 3; 5} FSUs with an ordered bandwidth requirement distribution.

[0063] Figure 15 shows the probability of blocking users with a traffic load p = 0.3 and require W = {2; 3; 5} FSUs with inverted bandwidth requirement distribution.

[0064] Figure 16 shows the probability of blocking users with a traffic load p = 0.5 and require W = {2; 3; 5} FSUs with a uniform bandwidth requirement distribution.

[0065] Figure 17 shows the probability of blocking users with a traffic load p = 0.5 and require W = {2; 3; 5} FSUs with an ordered bandwidth requirement distribution.

[0066] Figure 18 shows the probability of blocking users with a traffic load p = 0.5 and require W = {2; 3; 5} FSUs with inverted bandwidth requirement distribution.

[0067] Finally, it should be noted that the invention has been described primarily with reference to some preferred embodiments, exemplified in the accompanying figures. However, a person skilled in the art will readily recognize that other embodiments or modifications are equally possible within the spirit of the invention. Consequently, the foregoing detailed description should be understood in a broad sense, without limiting the spirit and scope of the invention.

Claims

CLAIMS 1. A method for allocating frequency spectrum to users of a dynamic elastic optical network, taking into account contiguity and continuity constraints, CHARACTERIZED in that it comprises the steps of: - to have data associated with an optical network topology including information on interconnected nodes, links and the capabilities of each link; - obtain information from users of the optical network, related to the source and destination nodes, the traffic load demanded by each user and the routes corresponding to each user; - classify users according to categories, where the categories are established according to their bandwidth requirements, measured in number of FSUs; - partition the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and allocate a shared zone of the frequency spectrum so that it can be used by any user regardless of their category; and - allocate a set of FSUs upon a transmission request from a user, which includes checking the availability of an initial number of contiguous FSUs within the zone corresponding to the user's category, on each of the links that make up the user's route, where said initial number of FSUs is equal to the number of FSUs required by the user, and, in the event that this initial number of contiguous FSUs is not available, check the availability of a new number of contiguous FSUs, on each of the links that make up the user's route, where the new number of contiguous FSUs are adjacent to the initial number of FSUs.

2. A frequency spectrum allocation method according to claim 1, CHARACTERIZED in that the step of classifying users according to categories includes assigning a set of users requiring the same bandwidth to a given category, such that each category groups users who have the same FSU requirement.

3. A frequency spectrum allocation method according to claim 2, CHARACTERIZED in that the partitioning of the frequency spectrum establishes a number of zones in each link of the optical network, where each zone is associated with a respective user category, and where each zone has a certain number of FSUs in order to provide space for the requirements of the users of the corresponding category.

4. A frequency spectrum allocation method according to claim 3, CHARACTERIZED in that the number of FSUs comprising each zone is equal to a multiple of the number of FSUs required by users of the corresponding category.

5. A frequency spectrum allocation method according to claim 2, CHARACTERIZED in that the shared zone comprises an arbitrary number of FSUs to be used by any user regardless of their category, such that the entire frequency spectrum of each of the network links is allocated to some type of category, either to one of the zones associated with the categories or to the shared zone.

6. A frequency spectrum allocation method according to claim 1, CHARACTERIZED in that the step of partitioning the frequency spectrum of each network link includes determining, for each network link, an identifier for each FSU of the link, further including establishing an identifier of the first FSU and the last FSU of each zone, an identifier of the first FSU and the last FSU of the shared zone, and a status of each FSU, either occupied or available.

7. A frequency spectrum allocation method according to claim 6, CHARACTERIZED in that the step of obtaining a set of FSUs for each user includes: a) receiving a request from a user of a given category (c / ), who has a requirement for n FSUs; b) determining, for each of the links in the user's path, an identifier (Di) of the first FSU in the zone associated with the category (o / ), and evaluating whether a first number of FSUs contiguous to FSU D / is available, where the number of contiguous FSUs corresponds to the user's requirement of n FSUs (o / ); c) in the case that the first number of contiguous FSUs to the identifier Di is available on each link of the user's route, assign the identified FSUs of each link to meet the user's demand, and store in a database information relating to the fact that the corresponding n FSUs assigned are occupied; d) in the case that the consulted FSUs are not available on at least one link, consult the availability of the new number of contiguous FSUs, which includes determining an identifier of the FSUs (Di + n¡) up to (Di + 2r>i-1), so that the new number of contiguous FSUs is adjacent to the first number;e) In the event that it is again identified that in at least one link the new FSUs consulted are not available, the method is configured to successively repeat step (d) to consult the availability of a new number of adjacent FSUs within the FSUs assigned to the user's category.; 8. A frequency spectrum allocation method according to claim 1, CHARACTERIZED in that if successive queries are made in a given subset regarding the availability of a certain number of FSUs, and it is determined that there are no contiguous FSUs available that can meet the requirement, the method then includes the additional step of querying the availability of said number of FSUs in the shared area and, if the connection request cannot be met in the shared area either, the method comprises the step of definitively rejecting the user's request.

9. A frequency spectrum allocation method according to claim 8, CHARACTERIZED in that the query of available FSUs in the shared area is carried out using a traditional First Fit method.

10. A frequency spectrum allocation method according to claim 1, CHARACTERIZED in that each time a user disconnects, because they have finished transmitting their corresponding data, the method comprises the step of making available again the FSUs that were temporarily assigned to said user.

11. A frequency spectrum allocation system for users of a dynamic elastic optical network, taking into account contiguity and continuity constraints, CHARACTERIZED in that it comprises: a database that stores information related to a provider network, including a set of nodes and links; information on network users, including the source and destination nodes of each user and required bandwidth, measured in FSUs; information on the capacity of each network link, measured in FSUs; information on frequency spectrum partitioning and the paths associated with each user; and processing means that communicate operationally with the database to carry out processes of user categorization, link partitioning, and FSU allocation, which includes: classifying users into categories, according to their bandwidth requirements, measured in the number of FSUs;Partitioning the frequency spectrum of each network link by subdividing the total number of FSUs into zones associated with each user category, and allocating a common zone of the frequency spectrum so that it can be used by any user regardless of their category; allocating a set of FSUs to a user upon a transmission request, which includes checking the availability of an initial number of contiguous FSUs within the zone associated with the user's category, on each of the links that make up the user's route, where this initial number corresponds to the number of FSUs required by the user, and, in the event that this initial number of contiguous FSUs is not available, checking the availability of a new number of contiguous FSUs, on each of the links of the user's route, which are adjacent to the initial number.

12. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the processes of user categorization, link partitioning, and FSU allocation further include storing in the database information related to frequency spectrum partitioning, which includes the subsets into which each link is divided, an identifier of a starting FSU for each subset and an identifier of an ending FSU for each subset, and an identifier of a starting FSU and an identifier of an ending FSU for a common zone.

13. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the system can be run in an environment comprising: an input interface, processing means, an output interface, and a database.

14. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the frequency spectrum allocation system includes the use of one or more servers, or a computer device with the ability to communicate with other devices, such as a desktop or laptop computer, outside the provider's network, and includes the use of a communication subsystem for communication between its various elements.

15. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the provider's network is an optical network, composed of a set of nodes interconnected by a set of optical fiber or multi-fiber links, wherein the network nodes can be of any type of dynamic optical network.

16. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the frequency spectrum allocation system is configured to use the input interface to receive and accept a connection request from a user, query the database to request network topology data, user data, link routes and capacity, and link partitioning information, use an output interface to deliver system information to any other system or device, and use the processing means to: carry out the processes of categorizing users and partitioning the frequency spectrum of each link, and check for each link in the user's route whether the FSUs of the category requested by the user are marked as available in the database;assign, if positive, said FSUs for each link and store the information in the database, indicating the user and the respective FSUs assigned in each link of the network; mark in the database; as occupied the FSUs assigned to each of the links in the route; and in the event that the assignment fails in any of the links of the network, check the FSUs of the zone that corresponds to the user in each of the links of their route, using the information from the database.

17. A frequency spectrum allocation system according to claim 11, CHARACTERIZED in that the frequency spectrum allocation system comprises multiple subsystems, including a network topology information subsystem, a user information subsystem, a link routing and capacity information subsystem, a link partitioning information subsystem, and a frequency allocation subsystem, wherein each subsystem may comprise a distinct server, or a network device within or outside the provider's network, including an input and output interface and its own database.

18. A frequency spectrum allocation system according to claim 17, CHARACTERIZED in that the network topology information subsystem stores information about a provider's network; the user information subsystem contains information about the user set, including information about the source and destination node pair of the provider's network, and the traffic load on FSUs demanded by each user; the routing and link capacity information subsystem is configured for determining routes for network users; the communications subsystem allows the reception of connection requests from users and allows communication between the different subsystems; the link partitioning information subsystem stores information about partitioning or zoning according to the different determined zones and the common zone;and the frequency allocation subsystem communicates operationally with the various subsystems to carry out the frequency allocation.

Citation Information

Patent Citations

  • Allocation of spectral capacity in a wavelength-division multiplexing optical network

    CN103797738A

  • Network management device and network management method

    JP2016134850A

  • Spectral slot assignment and placement of wavelength shifters in flexible grid optical networks

    US10348439B1

  • Optical transmission systems, devices, and methods

    US7130540B2

  • Exchange device

    US9178646B2