CN105490934A - Fair routing and spectrum assignment method based on fragment processing - Google Patents

Fair routing and spectrum assignment method based on fragment processing Download PDF

Info

Publication number
CN105490934A
CN105490934A CN201610049605.9A CN201610049605A CN105490934A CN 105490934 A CN105490934 A CN 105490934A CN 201610049605 A CN201610049605 A CN 201610049605A CN 105490934 A CN105490934 A CN 105490934A
Authority
CN
China
Prior art keywords
business
path
frequency spectrum
spectrum
fairness
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201610049605.9A
Other languages
Chinese (zh)
Other versions
CN105490934B (en
Inventor
熊余
范雪
刘书鸣
王汝言
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201610049605.9A priority Critical patent/CN105490934B/en
Publication of CN105490934A publication Critical patent/CN105490934A/en
Application granted granted Critical
Publication of CN105490934B publication Critical patent/CN105490934B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/12Shortest path evaluation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/22Alternate routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/70Admission control; Resource allocation
    • H04L47/80Actions related to the user profile or the type of traffic
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q11/00Selecting arrangements for multiplex systems
    • H04Q11/0001Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0005Switch and router aspects

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to a fair routing and spectrum assignment method based on fragment processing, and belongs to the technical field of routing and spectrum assignment of an elastic optical network. The method includes the following steps: virtualizing routing and spectrum and other network resources to form a resource pool, and solving small rectangular blocks that meet spectrum dual constraint conditions through a two-dimension rectangular Packing model; calculating the departure time of services by sensing the lasting time of the services, and assigning adjacent spectrum resources to the services that depart at the similar time so that spectrum fragments are less likely to generate; and constructing a fairness model of mixed grain services, and packing the same grain services into small rectangular blocks to carry out routing and spectrum assignment. The method meets the requirement of dual constraints of spectrum contiguity and continuity, avoids obstructions, and improves the overall performance of a network.

Description

A kind of fair route based on fragment process and frequency spectrum distributing method
Technical field
The invention belongs to elastic optical network route and frequency spectrum distributing technique field, relate to a kind of fair route based on fragment process and frequency spectrum distributing method.
Background technology
Along with HD video, the emerging in multitude of the intensive business of cloud computing equiband, the information transfer facilities based on optical-fiber network subjects the increasing pressure in high bandwidth.Due to existing wavelength division multiplexed optical network (WavelengthDivisionMultiplexing, WDM) stationary conduit interval is adopted, bandwidth resources by photosphere are coarse, be limited in a wavelength magnitude (being generally 50GHz) evenly, rigidly, cause that the flexibility of WDM optical-fiber network is low, bandwidth waste is serious, be difficult to the application supporting super 100G business, can not adapt to following Large Copacity, high bandwidth, extendible flexible transmission needs.For this reason, the elastic optical network (ElasticOpticalNetworks, EONs) based on orthogonal frequency division multiplexi (OrthogonalFrequencyDivisionMultiplexing, OFDM) was suggested in 2009 years.EON uses OFDM technology the allocated bandwidth granularity of photosphere to be reduced to the sub-wavelength magnitude (6.25GHz or 12.5GHz) of channel subcarrier, namely the elementary cell of photosphere Signal transmissions is no longer the wavelength of coarseness, but " frequently gap " (FrequencySlot, FS) of more fine granulation.When the combination grain business that bandwidth demand is different occurs, as long as distribute requisite number object FS for it.Thus effectively achieve the virtual of photosphere resource and configure as required, substantially increase network bandwidth utilization factor and flexible rate, so that it can support that the business of 1Tb/s is transmitted.
Route and spectrum allocation may (RoutingandSpectrumAssignment, RSA) are the most basic problems realizing elastic optical network high bandwidth transmission.Elastic optical network distinctive resource fine granularity turns to allocated bandwidth and brings new challenge, comprises the double constraints how meeting frequency spectrum adjacency and spectral continuity, how to reduce frequency spectrum fragment on the impact of resource utilization and how to guarantee fairness.Document [JinnoM, KozickiB, TakaraH, etal.Distance-adaptivespectrumresourceallocationinspectr um-slicedelasticopticalpathnetwork [J] .IEEECommunicationsMagazine, 2010,48 (8): 138-145.] from the angle meeting RSA double constraints, propose a kind of distance adaptive spectrum resource allocation methods to guarantee the frequency spectrum minimum number used, spectrum efficiency is higher, but the problem also bringing complexity too high.Document [ChristodoulopoulosK, TomkosI, VarvarigosEA.ElasticbandwidthallocationinflexibleOFDM-ba sedopticalnetworks [J] .JournalofLightwaveTechnology, 2011,29 (9): 1354-1366.] route and frequency spectrum distributing method are decomposed into route, modulation and spectrum allocation may two benches to process respectively, and adopt simulated annealing method to business sorting consistence, reach the object reducing computation complexity, improve the availability of frequency spectrum, but do not consider the impact of frequency spectrum fragment.For this reason, document [CastroA, VelascoL, RuizM, etal.Dynamicroutingandspectrum (re) allocationinfutureflexgridopticalnetworks [J] .ComputerNetworks, 2012,56 (12): 2869-2883.] to propose frequency spectrum reconfiguration method be service dynamic provides required frequency gap, thus reduces the generation of frequency spectrum fragment, reduces the blocking rate of network.Although above document proposes different solutions for elastic optical network route from spectrum allocation may problem, all do not consider the fairness problem of different grain size business RSA.Elastic optical network resource make full use of with flexible dispatching to sacrifice fairness for cost, the refinement of frequency spectrum granularity makes the granularity of business just may not meet frequency spectrum adjacency and successional constraint slightly greatly, thus blocks.And the WDM optical-fiber network that this situation is taking wavelength as basic allocation unit is just so not outstanding.Visible, the refinement of bandwidth unit causes the blocking rate of combination grain business to there is larger difference, and this brings coarsegrain business in elastic optical network and has a strong impact on, and causes the deterioration of network in general performance.
For this reason, document [FujiiS, HirotaY, MurakamiK, etal.Pathdivisionmethodforfairnessindynamicelasticoptica lpathnetworks [C] // 2012InternationalConferenceonPhotonicsinSwitching (PS2012), Ajaccio, France:IEEEpress, 2012:1-3.] utilize the relation of path distance and modulation, long jumping business is divided into several short jumping, and the method for hop-by-hop pathfinding avoids single business to block, reach the effect reducing network blocking probability.But this design is only for single business, lack the process to gap between combination grain business.Document [MaS, WangY, GuoB, etal.Afairness-awaredynamicspectrumallocationschemeinela sticopticalnetworks [C] //OptoElectronicsandCommunicationsConferenceandPhotonicsin Switching (OECC/PS2013), KyotoJapan:IEEEpress, 2013:1-2.] first network intermediate frequency spectrum is evenly divided into several region, the region differing maximum by blocking rate subsequently processes, thus improves fairness.But do not consider the availability of frequency spectrum.Document [Ma Songwei, Guo Bingli, Chen Xin, Deng. based on fairness aware dynamic spectrum allocation schemes (English) [J] of effectiveness in elastic optical network. photon journal, 2014,43 (7): 1-5.] utilize utility function to reduce blocking rate gap between different grain size business, thus the contradiction between balance fairness and blocking rate.But the process lacked frequency spectrum fragment, affects network performance.Document [BonaniLH, dosSantosTozettiA, CallegatiF, etal.Routingissuesonspectrumsharingandpartitioningforfle xibleopticalnetworks [C] // 201416thInternationalConferenceonTransparentOpticalNetwo rks (ICTON2014), Graz, Austria:IEEEpress, 2014:1-4.] network is equally assigned into the identical region of n size, and the business of same particle sizes is transmitted at each spectral regions, thus reduce the probability of frequency spectrum fragment formation, but because the frequency gap number of different grain size service needed is different, for different grain size traffic assignments size same asset may bring inequitable problem.Document [CallegatiF, BonaniL, LezamaF, etal.TrunkReservationforFairUtilizationinFlexibleOptical Networks [J] .IEEECommunicationLetters, 2014,18 (5): 1-4.] main line future mechanism (TrunkReservation, TR) is incorporated into elastic optical network to process fairness problem, if namely business granularity exceeds thresholding, then block this business.Can realize the fairness between different grain size business well by setting thresholding, but fairness and blocking rate are runed counter to, thus method guarantees fairness and makes blocking rate higher.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of fair route based on fragment process and frequency spectrum distributing method, the method, can with the overall situation of current network for foundation by the consideration business duration, when the business of considering as a whole reaches and business leave the service condition of rear resources in network.
For achieving the above object, the invention provides following technical scheme:
Based on fair route and the frequency spectrum distributing method of fragment process, in the method, by virtual for the Internet resources such as route and frequency spectrum be " resource pool ", by " the little rectangular block " that utilize two-dimensional rectangle Packing model solution to meet frequency spectrum double constraints condition; By the perception service duration, the time departure of computing service, is the frequency spectrum resource that the close traffic assignments of time departure is contiguous, reduces the generation of frequency spectrum fragment; Build the fairness model of combination grain business, same particle sizes business is packaged into " little rectangular block " to carry out route and spectrum allocation may.
Further, in the method, describedly calculate business time departure by the business duration and be: t leaving=t arrival+ t holding, wherein t leaving, t arrivaland t holdingbe respectively the time departure of business, reach time and duration;
The distribution service that multiple time departure is identical may be there is in network, for determining path optimizing, being optimized routing method (OptimalPathSelectedPolicy, OPSP), specifically comprising the following steps:
Step 1: find available frequency gap near the immediate distribution service of time departure, if existed, is then defined as path optimizing;
Step 2: if the path that step 1 obtains is not unique, then after selecting business to leave away, gap is maximum frequently continuously in release;
Step 3: if path optimizing is still not unique, selects high the carrying out of frequency spectrum continuity degree to distribute; Frequency spectrum continuity degree is defined as: Maxmize (2*x+y), t>=t leave, wherein, x, y are followed successively by business and leave the continuation degree produced at transverse axis and the longitudinal axis.
Further, the method of described route and spectrum allocation may is: first according to combination grain fairness model, different grain size business is packaged as transmission block, then K bar alternative path is obtained by KSP method, again by the perception of business duration, to be optimized path according to OPSP, if now path optimizing is unique, to select the position in the lower left corner as far as possible according to two-dimensional rectangle Packing model basic principle, carry out compact layout, specifically comprise the following steps:
Step 1: business is packaged into transmission block;
Step 2: to each transmission block, determines k bar optional path set by k shortest-path method (k-shortest-path, KSP); For business each in block, according to OPSP step one selecting paths, if this path is unique, then determine that it is path optimizing; If there are many such paths, then jump to step 3; If there is a business path optimizing cannot be found, then jump to step 5;
Step 3: select path optimizing according to OPSP step 2; If there are many path optimizings, then jump to step 4; If such path cannot be found, then jump to step 5;
Step 4: for determining path optimizing, is screened by OPSP step 3; If the selection result is unique, then determine that it is path optimizing; If still cannot determine, then according to two-dimensional rectangle Packing model basic principle, select lower left corner frequency spectrum as far as possible; Otherwise, jump to step 5;
Step 5: block all business in this transmission block, method terminates.
Beneficial effect of the present invention is: the method for the invention for " resource pool ", solves by utilizing two-dimensional rectangle Packing theory " the little rectangular block " that meet frequency spectrum double constraints condition by virtual for the Internet resources such as route and frequency spectrum; By the perception service duration, be the frequency spectrum resource that the close traffic assignments of time departure is contiguous, reduce the generation of frequency spectrum fragment; Build the fairness model of combination grain business, same particle sizes business is packaged into " little rectangular block " to carry out route and spectrum allocation may.
Accompanying drawing explanation
In order to make object of the present invention, technical scheme and beneficial effect clearly, the invention provides following accompanying drawing and being described:
Fig. 1 is two-dimensional rectangle Packing schematic diagram;
Fig. 2 is continuation degree schematic diagram;
Fig. 3 is spectrum allocation may schematic diagram;
Fig. 4 is the schematic flow sheet of the method for the invention.
Embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
Fig. 4 is the schematic flow sheet of the method for the invention, in the method, by virtual for the Internet resources such as route and frequency spectrum be " resource pool ", by " the little rectangular block " that utilize two-dimensional rectangle Packing model solution to meet frequency spectrum double constraints condition; By the perception service duration, the time departure of computing service, is the frequency spectrum resource that the close traffic assignments of time departure is contiguous, reduces the generation of frequency spectrum fragment; Build the fairness model of combination grain business, same particle sizes business is packaged into " little rectangular block " to carry out route and spectrum allocation may.
Specifically:
Because the present invention is intended to fairness, the reduction network spectrum fragment of Strengthens network Resourse Distribute.Related notion and measure as follows:
One, fairness
Fairness is used to the evenness problem that handles network resources is distributed, and is applicable to judge whether resource distributes to business in a kind of average mode.The distribution according to need of elastic optical network exacerbates the difference of combination grain service blocking rate, and then traffic affecting fairness.Defining fairness is in the present invention judge that whether resource is fair to the obstruction of different grain size business.
For carrying out quantitative evaluation to fairness, the concept of fairness index (FairnessIndex, FI) is proposed.On this basis, for the fairness quantification problem of combination grain business, combination grain business fairness index is proposed.For convenience of weighing, business identical for granularity is divided into same class, and the fairness of different classes of business judges as follows:
Definition 1: combination grain business fairness index, the i.e. tolerance of fairness between different grain size class of service, as shown in. represent m class service blocking rate.N' represents the class of service number existed in network.FI multi-granularityfor combination grain business fairness index.
Wherein, the blocking rate of m class business can be by try to achieve. for the m class business number blocked, and for all m class business numbers.
FI ∈ [0,1] represents that fairness has boundedness, and its value, more close to 1, represents that fairness is better.Otherwise, represent that fairness needs to improve.If all kinds of service blocking rate is identical, namely meet fairness is best, and now FI value is 1.If all kinds of service blocking rate is inconsistent, fairness can worsen along with the aggravation of inconsistent degree.
Theorem 1: all kinds of service blocking rate is more close, fairness is better, and when all kinds of service blocking rate is identical, fairness is best.Below to issuing a certificate.
FI m u l t i - g r a n u l a r i t y = [ Σ m = 1 n ′ P G m ] 2 n ′ Σ m = 1 n ′ P G m 2 = 1
⇔ [ Σ m = 1 n ′ P G m ] 2 = n ′ Σ m = 1 n ′ P G m 2
⇔ ( P G 1 + L + P G n ′ ) 2 = n ′ ( P G 1 2 + L + P G n ′ 2 )
Q ( P G 1 + L + P G n ′ ) 2 = P G 1 2 + L + P G n ′ 2 + 2 P G 1 P G 2 + L + 2 P G n ′ - 1 P G n ′
( P G 1 + L + P G n ′ ) 2 = n ′ ( P G 1 2 + L + P G n ′ 2 )
⇔ P G 1 2 + L + P G n ′ 2 + 2 P G 1 P G 2 + L + 2 P G n ′ - 1 P G n ′ = n ′ ( P G 1 2 + L + P G n ′ 2 )
⇔ 2 P G 1 P G 2 + L + 2 P G n ′ - 1 P G n ′ = ( n ′ - 1 ) ( P G 1 2 + L + P G n ′ 2 )
⇔ ( P G 1 - P G 2 ) 2 + ( P G 1 - P G 3 ) 2 + L + ( P G n ′ - P G n ′ - 1 ) 2 = 0
∴ can obtain P G 1 = P G 2 = L = P G n ′
The visible fairness optimization when all kinds of service blocking rate is identical.
Known by above-mentioned elaboration, for improving combination grain business fairness, all kinds of service blocking rate gap need be reduced as far as possible.But the RSA process need of elastic optical network meets frequency spectrum adjacency and successional constraint.Meanwhile, business produces a large amount of frequency spectrum fragments in the process constantly set up and remove, and is degrading the blocking rate of business.For this reason, introducing two-dimensional rectangle Packing model, business duration solve the problem successively.Finally, the difference of blocking rate between different grain size business is reduced by combination grain business fairness model.
Two, two-dimensional rectangle Packing model
Elastic optical network intermediate frequency spectrum unit is FS, for meeting the transmission requirement of combination grain business, and the double constraints of demand fulfillment spectral continuity and adjacency in RSA process.For this reason, Internet resources can be considered as i.e. " resource pool " by unified to route resource and frequency spectrum resource, according to adjacency and successional constraints, put into virtual " resource pool " and come integratedly to complete Route Selection and spectrum allocation may.For this reason, two-dimensional rectangle Packing model is introduced.
Two-dimensional rectangle Packing refers to, at any given large rectangle frame of length and width and any given respectively little rectangular block of limited length and width, require these little rectangular blocks to put into large rectangle frame as much as possible, make the area utilization of frame reach maximum.Similar, large rectangle frame can be considered as " resource pool " that one holds Internet resources, as shown in Fig. 1 (a).This resource pool is made up of, as shown in Fig. 1 (b) different paths rectangle frame.Wherein, transverse axis is link all in network, and the longitudinal axis is frequency spectrum.But the feature of conventional two-dimensional rectangle Packing model is that little rectangular block directly inserts large rectangle frame; Transverse axis, the longitudinal axis of little rectangular block all have arbitrariness.And two-dimensional rectangle Packing model of the invention process, business is at business rectangle frame, instead of transmission in large rectangle frame; Just because of this reason, transverse axis, namely transmission path is fixing.Only have the longitudinal axis, namely frequency spectrum is flexibility and changeability.
After business determines path, the respective links in large rectangle frame is taken out composition business rectangle frame, as shown in Fig. 1 (b).Visible, what only need ensure business transverse axis can meet frequency spectrum adjacency restriction requirement continuously.Meanwhile, on the longitudinal axis can ensure continuously spectral continuity constraint be met.So, just the problem meeting two constraintss in RSA process is converted into solution two-dimensional rectangle Packing.The basic principle of two-dimensional rectangle Packing be " preemption angle, after account for limit, finally accounting for " center ", determining thus to distribute from the lower left corner of each business rectangle frame as far as possible when distributing frequency spectrum.
Resource Allocation in Networks problem is converted into and solves two-dimensional rectangle Packing, frequency spectrum adjacency and successional constraint can be ensured.Can be ensured by two-dimensional rectangle Packing basic principle " compactedness " of distributing.But it is already present for being different from the medium and small rectangle of conventional two-dimensional rectangle Packing, business to be allocated dynamically arrives, and may arrive at any time and leave.For this reason, when carrying out Route Selection and spectrum allocation may, the business duration must be taken into account, to realize dynamic little block filling large rectangle frame.
Three, the perception of business duration
Two-dimensional rectangle Packing model can map frequency spectrum double constraints preferably.But the path optimizing how determining dynamic service need be further considered.For this reason, the concept of business duration is introduced.The business duration refers to the duration that certain business exists in a network.And path-calculating element (PathComputationElement, PCE) technology and general multiple protocols mark exchange agreement (GeneralizedMulti-ProtocolLabelSwitching, GMPLS) fast development, network online service level is significantly promoted, after business reaches, effectively can perceive the duration of this business.Visible, more effectively dynamic service is held for making " resource pool ", business time departure is calculated by business time of advent, business duration, and be the immediate traffic assignments adjacent spectra of time departure, thus make the large rectangle frame of playing the part of " resource pool " role insert more little rectangular block (i.e. business) as far as possible, thus reach the object reducing blocking rate, reduce frequency spectrum fragment.
Can calculate business time departure by the business duration is: t leaving=t arrival+ t holding, wherein t leaving, t arrivaland t holdingbe respectively the time departure of business, reach time and duration.The distribution service that multiple time departure is identical may being there is in network, for determining path optimizing, proposing path optimizing back-and-forth method (OptimalPathSelectedPolicy, OPSP).
Step 1: find available frequency gap near the immediate distribution service of time departure, if existed, is then defined as path optimizing.
Step 2: if the path that step 1 obtains is not unique, then after selecting business to leave away, gap is maximum frequently continuously in release.
Step 3: if path optimizing is still not unique, selects high the carrying out of frequency spectrum continuity degree to distribute.Frequency spectrum continuity degree is defined as: Maxmize (2*x+y), t>=t leave.Wherein, x, y are followed successively by business and leave the continuation degree produced at transverse axis and the longitudinal axis, as shown in Figure 2.In Fig. 2 (a), in dotted line, business is left rear Ru shown in 2 (b), and now, the continuation degree of generation is 3*x+4*2*y.
The business duration is according to now network condition selection alternative path.Produce maximum available frequency gap determination path optimizing according to the business departure time simultaneously.Just because of the selection of this maximum available frequency gap, reach the object reducing frequency spectrum fragment, improve resource utilization.But the problem of the fairness of combination grain is still unresolved.
Four, combination grain fairness model
Spectrum allocation may is instantly avoided to affect subsequent allocations by the introducing of business duration.But compared to WDM optical-fiber network, the fine granularity of elastic optical network and frequency spectrum fragment cause network fairness to decline, thus, combination grain business fairness problem is urgently to be resolved hurrily.For this reason, combination grain fairness model is proposed to ensure the fairness of combination grain business in elastic optical network.
Suppose in network, to there is the different business of n' class granularity, i.e. G={G 1, G 2lG n'.G n'represent n-th ' business that class granularity is identical.As represent that every class business packet contains the identical business of j' size. represent jth in m class business ' individual business.From all kinds of business, now take out a jth business composition transmission block B j, as B={B 1, B 2lB jwith shown in.B={B 1, B 2lB jrepresent in network and comprise j transmission block. represent that each transmission block comprises n' business, B jrepresent a jth transmission block, it comprises the business being designated as j under in n' class business.When there is not the business of a certain label in certain class business, use represent.
After business reaches, complete the division of transmission block according to above process.If process one by one business, due to the difference of business granularity, cause the difference of blocking rate between different grain size business.Now by different grain size business composition transmission block, when certain (individual) in block a little business cannot be transmitted, block all business in this transmission block.So, avoid the difference of coarsegrain business and small grain size service blocking rate, ensure that the fairness of combination grain business.
In sum, combination grain fairness can reduce the difference of blocking rate between combination grain, ensures the fairness of combination grain business in EONs.
Five, route and frequency spectrum distributing method
The basic ideas of RSA problem are: first according to combination grain fairness model, different grain size business is packaged as transmission block.Then K bar alternative path is obtained by KSP method, again by the perception of business duration, to be optimized path according to OPSP, if now path optimizing is not unique, select the position in the lower left corner according to two-dimensional rectangle Packing model basic principle as far as possible, carry out compact layout.Lower mask body sets forth the performing step of the method.
Step 1: business is packaged into transmission block.
Step 2: to each transmission block, determines k bar optional path set by k shortest-path method (k-shortest-path, KSP).For business each in block, according to OPSP step one selecting paths, if this path is unique, then determine that it is path optimizing.If there are many such paths, then jump to step 3.If there is a business path optimizing cannot be found, then jump to step 5.
Step 3: select path optimizing according to OPSP step 2.If there are many path optimizings, then jump to step 4.If such path cannot be found, then jump to step 5.
Step 4: for determining path optimizing, is screened by OPSP step 3.If the selection result is unique, then determine that it is path optimizing.If still cannot determine, then according to two-dimensional rectangle Packing model basic principle, select lower left corner frequency spectrum as far as possible.Otherwise, jump to step 5.
Step 5: block all business in this transmission block.
Step 6: method terminates.
Be described in detail for Fig. 3 below.Suppose to there are 4 links in network, every bar link comprises 8 gaps frequently.There are 8 business: N1-N8.Pending business is N9, and size is 1FS, and alternative path is C-F-G, C-B-G.Each business time departure is as shown in table 1, table 2.
Table 1 business N1-N4 time departure
Business time departure N1 N2 N3 N4
Fig. 3 (b) t 1 t 1 t 2 t 3
Fig. 3 (d) t 1 t 2 t 1 t 3
Fig. 3 (f) t 1 t 2 t 3 t 1
Fig. 3 (h) t 1 t 1 t 3 t 2
Table 2 business N5-N9 time departure
Fig. 3 (a) is " resource pool ", convenient service condition of observing each link in whole network.Fig. 3 (b)-3 (g) is corresponding in turn to the process selecting path optimizing according to formula OPSP, and Fig. 3 (i) is the process selecting path optimizing according to two-dimensional rectangle Packing basic principle.In Fig. 3 (b), first determine that 2,4,6 for available frequency gap, and business not identical with N9 time departure in Fig. 3 (c), thus determine that path optimizing is C-F-G.Again in conjunction with time departure, in Fig. 3 (b), N1 and N2 and N9 leaves simultaneously, selectes the frequency gap 2 between N1, N2 according to OPSP step one.In Fig. 3 (d), available frequency gap is 2,6,8.Left by the known business N1 of time departure, N3, N9 simultaneously.In Fig. 3 (e), available frequency gap is 2,4,6.Left by the known business N5 of time departure, N9 simultaneously.Now, select according to OPSP step 2.Be placed on 2, No. 6 positions in Fig. 3 (d) and 5,7 available frequency gaps will be produced successively, and in Fig. 3 (e), No. 2 positions by generation 4 available frequency gaps.Thus determine that C-F-G is path optimizing, and selected No. 6 spectrum positions.In Fig. 3 (f), available position is 2,6, and N1, N4, N9 together leave just simultaneously, and the frequency gap produced is 6FSs.And in Fig. 3 (g), available position only has No. 2 positions, N5, N9 together leave simultaneously, and the frequency gap of generation is 4FSs.Now, can determine that path optimizing is C-F-G according to OPSP step 3, the frequency gap block produced due to position 6 is more complete, selected No. 6 frequency spectrums.In Fig. 3 (h), available position is 2,6.No. 2 position the bests are determined according to formula (6)-(8).In Fig. 3 (i), available position is 1,4, and determines No. 1 position the best.Because No. 1 position more to meet the principle of " lower-left " position compared to No. 2 positions, thus determine that path optimizing is C-B-G, optimizing position is 1.
What finally illustrate is, above preferred embodiment is only in order to illustrate technical scheme of the present invention and unrestricted, although by above preferred embodiment to invention has been detailed description, but those skilled in the art are to be understood that, various change can be made to it in the form and details, and not depart from claims of the present invention limited range.

Claims (3)

1. the fair route based on fragment process and frequency spectrum distributing method, it is characterized in that: in the method, by virtual for the Internet resources such as route and frequency spectrum be " resource pool ", by " the little rectangular block " that utilize two-dimensional rectangle Packing model solution to meet frequency spectrum double constraints condition; By the perception service duration, the time departure of computing service, is the frequency spectrum resource that the close traffic assignments of time departure is contiguous, reduces the generation of frequency spectrum fragment; Build the fairness model of combination grain business, same particle sizes business is packaged into " little rectangular block " to carry out route and spectrum allocation may.
2. a kind of fair route based on fragment process according to claim 1 and frequency spectrum distributing method, is characterized in that: in the method, describedly calculates business time departure by the business duration and be: t leaving=t arrival+ t holding, wherein t leaving, t arrivaland t holdingbe respectively the time departure of business, reach time and duration;
The distribution service that multiple time departure is identical may be there is in network, for determining path optimizing, being optimized routing method (OptimalPathSelectedPolicy, OPSP), specifically comprising the following steps:
Step 1: find available frequency gap near the immediate distribution service of time departure, if existed, is then defined as path optimizing;
Step 2: if the path that step 1 obtains is not unique, then after selecting business to leave away, gap is maximum frequently continuously in release;
Step 3: if path optimizing is still not unique, selects high the carrying out of frequency spectrum continuity degree to distribute; Frequency spectrum continuity degree is defined as: Maxmize (2*x+y), t>=t leave, wherein, x, y are followed successively by business and leave the continuation degree produced at transverse axis and the longitudinal axis.
3. a kind of fair route based on fragment process according to claim 1 and frequency spectrum distributing method, it is characterized in that: the method for described route and spectrum allocation may is: first according to combination grain fairness model, different grain size business is packaged as transmission block, then K bar alternative path is obtained by KSP method, again by the perception of business duration, to be optimized path according to OPSP, if now path optimizing is not unique, the position in the lower left corner is selected as far as possible according to two-dimensional rectangle Packing model basic principle, carry out compact layout, specifically comprise the following steps:
Step 1: business is packaged into transmission block;
Step 2: to each transmission block, determines k bar optional path set by k shortest-path method (k-shortest-path, KSP); For business each in block, according to OPSP step one selecting paths, if this path is unique, then determine that it is path optimizing; If there are many such paths, then jump to step 3; If there is a business path optimizing cannot be found, then jump to step 5;
Step 3: select path optimizing according to OPSP step 2; If there are many path optimizings, then jump to step 4; If such path cannot be found, then jump to step 5;
Step 4: for determining path optimizing, is screened by OPSP step 3; If the selection result is unique, then determine that it is path optimizing; If still cannot determine, then according to two-dimensional rectangle Packing model basic principle, select lower left corner frequency spectrum as far as possible; Otherwise, jump to step 5;
Step 5: block all business in this transmission block, method terminates.
CN201610049605.9A 2016-01-25 2016-01-25 A kind of fair routing and frequency spectrum distributing method based on fragment processing Active CN105490934B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610049605.9A CN105490934B (en) 2016-01-25 2016-01-25 A kind of fair routing and frequency spectrum distributing method based on fragment processing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610049605.9A CN105490934B (en) 2016-01-25 2016-01-25 A kind of fair routing and frequency spectrum distributing method based on fragment processing

Publications (2)

Publication Number Publication Date
CN105490934A true CN105490934A (en) 2016-04-13
CN105490934B CN105490934B (en) 2018-08-21

Family

ID=55677664

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610049605.9A Active CN105490934B (en) 2016-01-25 2016-01-25 A kind of fair routing and frequency spectrum distributing method based on fragment processing

Country Status (1)

Country Link
CN (1) CN105490934B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105933174A (en) * 2016-07-12 2016-09-07 重庆邮电大学 Pre-calculating recovery method based on distance self-adaptive route and spectrum distribution
CN106850427A (en) * 2017-01-17 2017-06-13 北京工业大学 The route frequency spectrum distributing method of the elastic optical multicast network that network-oriented coding is enabled
CN107135056A (en) * 2017-04-28 2017-09-05 重庆邮电大学 It is a kind of to reduce frequency spectrum fragment and the Anycast service resource allocation method of time delay
CN107682134A (en) * 2017-09-28 2018-02-09 全球能源互联网研究院 A kind of frequency spectrum distributing method and system
CN109818803A (en) * 2019-02-26 2019-05-28 重庆邮电大学 A kind of frequency spectrum fragment avoiding method based on machine learning
CN110035337A (en) * 2019-04-16 2019-07-19 郑州轻工业学院 The active multithread frequency spectrum distributing method of frequency block is divided in elastic optical network based on index
CN110365589A (en) * 2019-07-30 2019-10-22 国网福建省电力有限公司 A kind of routing of electric power optical transport and frequency spectrum distributing method based on elastic optical network
CN111182379A (en) * 2020-01-03 2020-05-19 中原工学院 Frequency spectrum allocation method for plastic reserved service in elastic optical network
CN112203167A (en) * 2020-09-10 2021-01-08 烽火通信科技股份有限公司 Routing spectrum allocation method, device, equipment and readable storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102447623A (en) * 2011-11-22 2012-05-09 北京邮电大学 Cascade method of optical network frequency spectrum resource fragments and apparatus thereof
CN102548001A (en) * 2012-03-01 2012-07-04 重庆邮电大学 Coexistence method for WIA-PA network and IEEE 802.15.4 network
CN102752757A (en) * 2011-04-19 2012-10-24 深圳清华大学研究院 Method for optimizing frequency spectrum allocation according to minimal waste criterion in frequency spectrum aggregation process
CN103327501A (en) * 2013-05-17 2013-09-25 北京邮电大学 Resource allocation method based on distributed debris concentration ratio
CN103391258A (en) * 2013-07-03 2013-11-13 北京邮电大学 Resource distribution method and device based on distributed fragment concentration ratios
CN103684726A (en) * 2013-09-30 2014-03-26 广西大学 Routing and frequency spectrum allocating method and routing and frequency spectrum allocating equipment
CN103916926A (en) * 2014-04-22 2014-07-09 国家电网公司 Routing and spectrum allocation method based on non-interval frequency spectrum
US20140379927A1 (en) * 2013-06-24 2014-12-25 Nec Laboratories America, Inc. Network Followed by Compute Load Balancing Procedure for Embedding Cloud Services in Software-Defined Flexible-Grid Optical Transport Networks
CN104836736A (en) * 2015-05-26 2015-08-12 重庆邮电大学 Spectrum sensing-based energy efficiency resource allocation and path reconfiguration method in elastic optical network

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102752757A (en) * 2011-04-19 2012-10-24 深圳清华大学研究院 Method for optimizing frequency spectrum allocation according to minimal waste criterion in frequency spectrum aggregation process
CN102447623A (en) * 2011-11-22 2012-05-09 北京邮电大学 Cascade method of optical network frequency spectrum resource fragments and apparatus thereof
CN102548001A (en) * 2012-03-01 2012-07-04 重庆邮电大学 Coexistence method for WIA-PA network and IEEE 802.15.4 network
CN103327501A (en) * 2013-05-17 2013-09-25 北京邮电大学 Resource allocation method based on distributed debris concentration ratio
US20140379927A1 (en) * 2013-06-24 2014-12-25 Nec Laboratories America, Inc. Network Followed by Compute Load Balancing Procedure for Embedding Cloud Services in Software-Defined Flexible-Grid Optical Transport Networks
CN103391258A (en) * 2013-07-03 2013-11-13 北京邮电大学 Resource distribution method and device based on distributed fragment concentration ratios
CN103684726A (en) * 2013-09-30 2014-03-26 广西大学 Routing and frequency spectrum allocating method and routing and frequency spectrum allocating equipment
CN103916926A (en) * 2014-04-22 2014-07-09 国家电网公司 Routing and spectrum allocation method based on non-interval frequency spectrum
CN104836736A (en) * 2015-05-26 2015-08-12 重庆邮电大学 Spectrum sensing-based energy efficiency resource allocation and path reconfiguration method in elastic optical network

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
FARHAD AKHOONDI 等: "Resource Allocation Using Fragmented-Spectrum Synchronous OFDM-CDMA in Cognitive Radio Networks", 《IEEE》 *
YU XIONG 等: "Fast Fault Localization Mechanism Based on Minimum Dominating Set Clustering in WDM Networks", 《OPTICAL SWITCHING AND NETWORKING》 *
鞠卫国 等: "面向频谱融合的路由频谱分配和碎片整理算法", 《光子学报》 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105933174B (en) * 2016-07-12 2018-10-19 重庆邮电大学 A kind of precomputation restoration methods based on apart from adaptive routing and frequency spectrum distribution
CN105933174A (en) * 2016-07-12 2016-09-07 重庆邮电大学 Pre-calculating recovery method based on distance self-adaptive route and spectrum distribution
CN106850427A (en) * 2017-01-17 2017-06-13 北京工业大学 The route frequency spectrum distributing method of the elastic optical multicast network that network-oriented coding is enabled
CN107135056B (en) * 2017-04-28 2019-12-27 重庆邮电大学 Anycast service resource allocation method for reducing frequency spectrum fragments and time delay
CN107135056A (en) * 2017-04-28 2017-09-05 重庆邮电大学 It is a kind of to reduce frequency spectrum fragment and the Anycast service resource allocation method of time delay
CN107682134A (en) * 2017-09-28 2018-02-09 全球能源互联网研究院 A kind of frequency spectrum distributing method and system
CN107682134B (en) * 2017-09-28 2022-07-01 全球能源互联网研究院 Spectrum allocation method and system
CN109818803A (en) * 2019-02-26 2019-05-28 重庆邮电大学 A kind of frequency spectrum fragment avoiding method based on machine learning
CN110035337A (en) * 2019-04-16 2019-07-19 郑州轻工业学院 The active multithread frequency spectrum distributing method of frequency block is divided in elastic optical network based on index
CN110365589A (en) * 2019-07-30 2019-10-22 国网福建省电力有限公司 A kind of routing of electric power optical transport and frequency spectrum distributing method based on elastic optical network
CN111182379A (en) * 2020-01-03 2020-05-19 中原工学院 Frequency spectrum allocation method for plastic reserved service in elastic optical network
CN112203167A (en) * 2020-09-10 2021-01-08 烽火通信科技股份有限公司 Routing spectrum allocation method, device, equipment and readable storage medium
CN112203167B (en) * 2020-09-10 2021-09-07 烽火通信科技股份有限公司 Routing spectrum allocation method, device, equipment and readable storage medium

Also Published As

Publication number Publication date
CN105490934B (en) 2018-08-21

Similar Documents

Publication Publication Date Title
CN105490934A (en) Fair routing and spectrum assignment method based on fragment processing
Abkenar et al. Study and analysis of routing and spectrum allocation (RSA) and routing, modulation and spectrum allocation (RMSA) algorithms in elastic optical networks (EONs)
US7249169B2 (en) System and method for network control and provisioning
US10009287B2 (en) Hierarchical software-defined network traffic engineering controller
US9654248B2 (en) Optical data transmission method and apparatus
US7403988B1 (en) Technique for autonomous network provisioning
CN105827528B (en) A kind of route selection method suitable for the flexible optical-fiber network of frequency spectrum
Fadini et al. A subcarrier-slot partition scheme for wavelength assignment in elastic optical networks
WO2013037570A1 (en) Allocation of spectral capacity in a wavelength-division multiplexing optical network
CN101127689B (en) Method for minimizing priority seizing cost
CN108377220B (en) Transparent virtual optical network collaborative mapping method for sensing node importance
CN109743217B (en) Self-adaptive resource adjusting method based on SVRA algorithm
WO2013000116A1 (en) Method and device for leaky bucket speed-limitation
CN111010627B (en) Route distribution method and device applied to mixed grid optical network
CN115396514A (en) Resource allocation method, device and storage medium
CN106961641A (en) A kind of frequency spectrum distributing method perceived in elastic optical network based on service priority
CN108616782A (en) A kind of frequency spectrum distributing method based on the optimization of non-intersecting connection group in elastic optical network
Sun et al. Hop-by-hop bandwidth allocation and deployment for SFC with end-to-end delay QoS guarantees
CN105634990A (en) Resource reservation method, device and processor based on time spectrum continuity
CN104301255B (en) A kind of method of optical-fiber network multi-user fair bandwidth sharing
Enderle et al. Modeling dynamic traffic demand behavior in telecommunication networks
Maher et al. A hybrid bandwith allocation algorithm for EPON-WiMAX integrated networks based on auction process
Lan et al. A fragmentation-aware load-balanced RMSCA algorithm in space-division multiplexing elastic optical networks
CN104202111B (en) Triggered aggregation optical path reconstruction method based on channel density and minimum clearance
CN113890829B (en) Network virtualization resource allocation method and system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant