Disclosure of Invention
In view of this, an object of the present invention is to provide a spectrum slicable resource allocation method based on crosstalk sensing in a multi-core elastic optical network, which reduces the bandwidth blocking rate of a service while reducing the influence of crosstalk and fragmentation on resource allocation.
In order to achieve the purpose, the invention provides the following technical scheme:
a spectrum switchable piece resource allocation method based on crosstalk perception in a multi-core elastic optical network comprises the following steps:
s1: inputting a multi-core elastic optical network topological structure G (N, E, F, C), a node set N and a link set E, wherein each link comprises a frequency gap set F and a fiber core set C; according to the average value of the node degrees in the network topological structure, marking as D, configuring the frequency spectrum slicers with the number of D for each node, wherein the configured connection mode is an on-demand node structure; inputting a service request, and grouping services according to the required bandwidth value of the services in the network;
s2: according to the vertex coloring principle in the graph theory, dividing nonadjacent fiber cores in an input multi-core fiber set C into a group, dividing the grouped fiber cores into fiber core groups, selecting high priority, medium priority and low priority, and carrying out spectrum partitioning on spectrum resources selected by the fiber core groups of each grade;
the calculation formula of the number of frequency slots contained in the frequency spectrum partition i of each fiber core group is as follows:
in the formula (1), m and n are the number of kinds of i-th group service and the total number of kinds of various services, R
k And R
ij The number of FS required by the kth service and the ith group service of the jth type respectively, m is less than or equal to n, P
k And P
ij The ratio of the kth service and the jth group service to the total service is respectively, the fraction in the calculation formula is used for calculating the ratio of the interval service request to all service requests, S is the number of frequency slots contained in a fiber core, and T is the number of frequency slots contained in a fiber core
ime Which represents a positive integer value of the number,
representing taking a maximum integer for x;
s3: selecting K candidate routes in the multi-core elastic optical network by using a shortest path method according to a source node and a destination node of a service request, and storing the K candidate routes into a set P, wherein the value of K is a maximum integer value of the average degree of nodes of a network topology;
s4: determining a modulation format adopted by the service according to the length of the candidate path, calculating the number of frequency slots required by the service according to the modulation format, sequentially finding out idle continuous available frequency spectrum blocks meeting the requirement of the service frequency slots in the candidate path set P, if idle continuous frequency spectrum resources meeting the number of frequency spectrums required by the service exist in the candidate path set P, turning to the step S5, and if not, turning to the step S6;
the calculation formula of the frequency slot number required by the service is as follows:
in equation (2), b represents the bandwidth requirement of the service request, b fs For bandwidth per frequency slot, in elastic optical networks, b fs Modulation efficiency of modulation format selected for traffic, =12.5ghz, m, F G To protect the number of frequency slots;
s5: executing a crosstalk sensing spectrum resource allocation strategy based on the candidate resource windows, and selecting a spectrum window with minimum inter-core crosstalk for the service;
s6: according to spectrum slice module resources configured in the elastic optical network of the on-demand node architecture, performing spectrum slicing on a transit node of a candidate path from a source node to a destination node of a service, and executing a distribution cost perception resource distribution strategy based on the spectrum slice.
Further, S5 is specifically implemented by the following steps:
s501: according to the vertex coloring principle of graph theory, non-adjacent fiber cores are divided into a group, the fiber cores are divided into high, medium and low priority levels, and the service selects the fiber cores to be distributed from the high, medium and low priority levels in sequence;
s502: sorting the idle continuous available frequency spectrum blocks of the candidate paths in the candidate path set P according to the fiber core grouping priority, and storing the sorted continuous available frequency spectrum blocks in the candidate path set P in the candidate frequency spectrum window set
The preparation method comprises the following steps of (1) performing;
s503: compute collections in turn
If a candidate spectral window with the intercore crosstalk value smaller than the intercore crosstalk threshold value is found, the spectral window is allocated to a service, and the fiber core and path information of the spectral window are recorded; otherwise, the spectrum resource allocation of the service fails, the service is blocked, and the algorithm is finished;
the calculation formula of the crosstalk value between cores of each candidate spectral window is as follows:
in the formula (3), the compound represented by the formula (I),
is a core c
i And its adjacent core c
i ' Cross talk between wherein
L is the transmission length of the optical fiber, h represents the incremental crosstalk per unit transmission length,
wherein, alpha, r, beta, omega
th Respectively representing the coupling coefficient, the bending radius, the propagation constant and the distance between cores of the optical fiber;
the reference value of the crosstalk threshold value among cores under different modulation formats is as follows:
further, S6 is specifically implemented by:
s601: sequentially counting the idle available spectrum blocks of each candidate path in the set P, and sequentially arranging the available spectrum blocks in the set { b } from large to small according to the number of frequency slots contained in the idle spectrum blocks 1 ,b 2 ,...,b j In (1) };
s602: checking the switching node whether there is a spectrum slicer left according to the frequency slot number required by the service, if so, according to the termination condition of the spectrum slicer, selecting the available spectrum block set { b } 1 ,b 2 ,...,b j Finding out the spectrum block combination meeting the number of frequency slots needed by the service, and storing the spectrum block combination into a spectrum block set psi, psi = { (psi) 1 (b 1 ,b 2 ,...),ψ 2 (b 1 ,b 3 ,...),...,ψ r (b 2 ,b 3 ,...,b j ),...};
Wherein the spectral slice termination condition is as follows:
in equation (4), T represents the number of sub-slice portions of the traffic r sliced into T on the candidate path, corresponding to the spectral block combination ψ
r Including the number of free-spectrum blocks,
|b
t | represents a spectral block combination ψ
r The frequency slot number contained in the t-th idle spectrum block and the frequency slice termination condition formula show that: when the sum of the total frequency slot number required by the sliced service and the total number of the guard frequency slots between the sliced sub-slices is less than or equal to the total idle frequency slot number on the path, the slicing is stopped;
s603: according to a path distribution cost calculation formula, after service request slices are calculated, each sub-slice selects each spectral window combination psi from candidate spectral window combinations psi of paths r The total assigned cost value of;
wherein the sub-slice is in the candidate path P j Spectral window combination psi r The calculation formula of the total distribution cost is as follows:
in the formula (5), the compound represented by the formula (I),
indicating that a sub-slice is allocated on path P
j Over the link (a, b) of (c) is combined with the spectral window psi
r The t-th block b
t The cost value of (2) is calculated by the following method:
in the formula (6), S
a,b (i) Indicating the status of whether the ith frequency slot of the link (a, b) is occupied or not, when S
a,b (i) =0, meaning that the i-th frequency slot is occupied, then the frequency slot cannot be allocated to the current sub-slice request, S
a,b (i) =1 denotes that the ith frequency slot is free;
the calculation method is divided into 3 cases: requesting allocation of the t-th frequency slot block b in the current subslice
t When the initial index value of the spectrum window is 1, the cost value is determined by the occupation condition of the end frequency gap of the spectrum block; when the sub-slice requests allocation in the t-th frequency slot block b
t When the frequency spectrum window ending frequency slot is positioned at the rightmost side of the link frequency slot, the cost value is determined by the occupation condition of the adjacent frequency slot at the right side; requesting allocation of the block b at the t-th frequency slot in the current subslice
t When the index values of the start frequency slot and the end frequency slot of the frequency spectrum window are positioned in the middle of the link frequency slot, the distribution cost is determined by the occupation conditions of the adjacent frequency slots on the left side and the right side of the frequency spectrum block to be occupied by the sub-slice;
s604: according to the total assigned cost value of the service slices, combining the frequency spectrum windows in an ascending order;
s605: sequentially calculating the crosstalk value between cores after the combination of the frequency spectrum windows is allocated to the sub-slices, if the crosstalk value between the cores is smaller than a crosstalk threshold value, allocating the combination of the frequency spectrum windows to each sub-slice service, and ending the algorithm; and if the crosstalk values among the cores of all the sequenced spectral window combinations do not meet the requirement of the crosstalk threshold, the resource allocation of the sub-slices fails, the service is blocked, and the algorithm is ended.
The invention has the beneficial effects that: the invention provides a frequency spectrum switchable chip resource allocation method based on crosstalk sensing in a multi-core elastic optical network, which considers the matching problem of node port data and link chip resources, the crosstalk influence among cores and the path allocation cost when allocating frequency spectrum resources for services. Firstly, providing an on-demand node structure for configuring a spectrum slicer based on MCF-EONs, improving the flexibility of node exchange and reducing the transmission cost of a system; a fiber core selection stage, wherein a fiber core selection strategy based on fiber core grouping and frequency spectrum partition is provided; the spectrum resource allocation stage is divided into two parts, when available resources are sufficient, a crosstalk sensing resource allocation algorithm based on candidate resource windows is provided, and a resource window which is influenced the least by crosstalk in a selected path is selected for a service; when the resource allocation fails, an allocation cost perception resource allocation algorithm based on the frequency spectrum slice is provided, so that the generation of frequency spectrum fragments is reduced, and the frequency spectrum utilization rate is improved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The invention provides a frequency spectrum switchable resource allocation method based on crosstalk sensing in a multi-core elastic optical network, and aims to better solve the problem of crosstalk between cores and the problem of frequency spectrum fragments of services in the multi-core elastic optical network. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and embodiments may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some components in the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The embodiment is as follows:
fig. 1 is a schematic diagram of an on-demand node architecture based on spectrum slicing. Fig. 1 (a) shows three switching patterns of a node on Demand Architecture (AoD). The first mode is fiber switching, as shown by the blue dashed line in fig. 1 (a), where a multicore fiber enters from an input port 1 and all traffic in the core is connected to an output port 1. In this case, traffic on the entire core is switched, so it is not necessary to use a spatial Multiplexer/Demultiplexer (MUX/(De) MUX, multiplexer/Demultiplexer) and a Spectrum Selective Switch (SSS); the second mode is core switching, as shown by the green dashed line in fig. 1 (a), in the case of non-fiber switching, if all the services on a certain core in the fiber need to be switched to another core, MUX/(De) MUX needs to be used, and when the services in one fiber need to be transferred to a different core, or needs to receive the services transmitted by one core in the fiber, SSS needs to be used; the third mode is frequency slot switching, as shown by the red dotted line in fig. 1 (a), if the output ports to which the traffic transmitted on the core is destined are different, the traffic must be separated by the SSS and then forwarded to the respective output ports (output ports) of the traffic. In addition, when the idle spectrum resource in the link does not meet the bandwidth required by the service, the service is blocked. In order to avoid this, a spectrum slicing (Slicer) machine is required to relax the spectrum continuity constraint on the resource blocks on the fiber cores in resource allocation, so as to implement the combined use of spectrum fragmentation resources. Fig. 1 (b) shows an on-demand node switching fabric configured with a spectrum slicer, in which various optical devices are flexibly cross-connected according to service requirements of arriving nodes on a programmable optical AoD backplane of the on-demand node, and the node structure can save network cost and improve utilization rate of the optical devices compared with a conventional static node structure based on hard connection to reduce device consumption.
A core selection diagram based on core grouping and spectral zoning for a 7-core multicore fiber as shown in fig. 2. Adopting 7-core multi-core fiber, dividing non-adjacent fiber cores into a group according to the vertex coloring principle in the graph theory, and defining the priority of fiber core selection, wherein the grouping result is shown as the attached figure 2 (a), and is respectively high priority: a fiber core group (1, 3, 5); medium priority: a set of fibre cores (2, 4, 6); low priority: a core (7); carrying out frequency spectrum partitioning on each fiber core resource in the same fiber core group, namely, the fiber core 1 only transmits services with the same common divisor S1 in the required frequency spectrum number; the fiber cores with medium priority are divided into public fiber cores with specific frequency spectrum partitions, the frequency spectrum partition result of the fiber cores with medium priority is shown in figure 2 (b), the frequency spectrum resources in the same fiber core are divided into three areas (S1, S2 and S3), and in the same frequency spectrum area, the frequency spectrum allocation criterion is jointly used according to the adoption of a first matching strategy and a last matching strategy.
An exemplary graph of spectral window-to-core crosstalk as shown in figure 3. FIG. 3 shows a hexagonal dense structure of a porous left-hand arrangement model of a 7-core multicore fiber in an optical fiber; each square represents each frequency slot, with light blue squares representing occupied frequency slots and white squares representing idle frequency slots. Assuming that the
core 6 is selected by the service request, the required frequency slot number is 5FSs, and the candidate spectral window is
The actual inter-core crosstalk value of the traffic
Is composed of
Comparing the calculated crosstalk value between cores with a crosstalk threshold value under the current modulation format, and if the calculated crosstalk value is smaller than the crosstalk threshold value, successfully transmitting the service; otherwise, the next candidate path is searched, if the K candidate paths do not satisfy the crosstalk threshold condition, the service resource allocation fails, and the service is blocked.
Wherein, the crosstalk threshold under different modulation formats is:
fig. 4 is a diagram illustrating an example of path allocation cost. Each square represents each frequency slot, with the gray squares representing occupied frequency slots and the white squares representing idle frequency slots. Assuming that the required frequency slots are 4FSs, selecting candidate paths containing service di of 3 links to be distributed on the paths A-B-C-D, observing idle spectral windows on the paths A-B-C-D to obtain a candidate spectral window combination psi of the service 1 :(SW 2-3 ,SW 5-6 ),ψ 2 :(SW 2-3 ,SW 9-10 ),ψ 3 :(SW 5-6 ,SW 9-10 ) }; according to path allocationA total cost calculation formula for calculating the distribution cost values of the three candidate spectral window combinations respectively, wherein psi 1 :(SW 2-3 ,SW 5-6 ) If the allocation cost is the minimum, the corresponding spectrum-time resource window combination is allocated to the sub-slice service after the spectrum slice.
Wherein, the distribution cost calculation result of the three candidate spectrum window combinations is:
a general flow of spectrum switchable resource allocation based on crosstalk sensing in the multi-core elastic optical network according to the present invention will be described below with reference to fig. 5, where the general flow may be divided into the following steps:
s1: the method comprises the steps of inputting a network topology structure G (N, E, F and C), a node set N and a link set E, wherein each link comprises a frequency slot set F and a fiber core set C. According to the average value of the node degrees in the network topological structure, marked as D, configuring frequency spectrum slicers with the number of D for each on-demand node architecture, inputting a service request, and grouping services according to the required bandwidth value of the services in the network;
s2: according to a vertex coloring principle in a graph theory, dividing nonadjacent fiber cores in an input multi-core fiber set C into a group, dividing the grouped fiber cores into fiber core groups, selecting high priority, medium priority and low priority, and performing spectrum partitioning on spectrum resources selected by the fiber core groups of each grade;
the calculation formula of the number of frequency slots contained in the frequency spectrum partition i of each fiber core group is as follows:
in the above formula, m and n are the number of kinds of i-th group service and the total number of kinds of various services, R
k And R
ij The number of FS required by the kth service and the jth group service of the jth type respectively, m is less than or equal to n, P
k And P
ij Respectively, the kth service andthe j-th group service accounts for the total service, the fraction in the calculation formula is used for calculating the ratio of the interval service request to all service requests, S is the number of frequency slots contained in a fiber core, T is the number of frequency slots contained in a fiber core
ime Which represents a positive integer value of the number,
representing taking a maximum integer for x;
s3: selecting K candidate routes in a multi-core elastic optical network by using a shortest path method according to a service request, storing the K candidate routes into a set P, wherein the value of K is a maximum integer value of the node average degree of the network topology upwards, determining a service modulation format according to the candidate path length, and calculating the frequency slot number required by the service according to the service request bandwidth and the modulation format;
the calculation formula of the frequency slot number required by the service is as follows:
in the above equation, b represents the bandwidth requirement of the service request, b fs For bandwidth per frequency slot, in elastic optical networks, b fs Modulation efficiency of modulation format selected for traffic, =12.5GHz, m G To protect the number of frequency slots;
s4: judging whether a frequency spectrum block meeting the frequency slot number required by service transmission exists in the set P, if so, turning to S10, otherwise, turning to S5;
s5: calling a resource allocation algorithm based on allocation cost perception of the frequency spectrum slices, using a frequency spectrum slicer module of an on-demand node structure to perform frequency spectrum slicing on the service from a source node to a destination node, and turning to S6;
s6: observation candidate path P i The spectrum resource occupation of P i The middle idle frequency spectrum blocks are arranged from large to small according to the size of the contained frequency slots and are stored in a set { b } 1 ,b 2 ,...,b j Fourthly, turning to S7;
s7: checking the switching node for a remaining spectrum slicer, if any, based onSpectral slicer termination condition from set of available spectral blocks b 1 ,b 2 ,...,b j Finding out the spectrum block combination meeting the number of frequency slots needed by the service, and storing the spectrum block combination into a spectrum block set psi, psi = { (psi) 1 (b 1 ,b 2 ,...),ψ 2 (b 1 ,b 3 ,...),...,ψ r (b 2 ,b 3 ,...,b j ) A check, go to S8;
the calculation formula of the termination condition of the frequency spectrum slice is as follows:
in the above equation, T represents the number of sub-slice portions on the candidate path representing the traffic r sliced into T, corresponding to the spectral block combination ψ
r Including the number of free-spectrum blocks,
|b
t | represents a spectral block combination ψ
r The frequency slot number contained in the t-th idle spectrum block and the frequency slice termination condition formula show that: when the sum of the total frequency slot number required by the sliced service and the total number of the guard frequency slots among the sliced sub-slices is less than or equal to the total idle frequency slot number on the path, the slicing is stopped;
s8: according to a path distribution cost calculation formula, after service request slices are calculated, each sub-slice selects each spectral window combination psi from candidate spectral window combinations psi of paths r The total assigned cost value of (a); turning to S9;
wherein the subslice are at the candidate path P j Spectral window combination psi r The total distribution cost is calculated by the following formula:
in the above-mentioned formula, the compound of formula,
indicating that a sub-slice is allocated on path P
j On the link (a, b) of
r Middle t frequency slot block b
t The cost value of (2) is calculated by the following method:
in the above formula, S
a,b (i) Indicating the status of whether the ith frequency slot of the link (a, b) is occupied or not, when S
a,b (i) =0, meaning that the i-th frequency slot is occupied, then the frequency slot cannot be allocated to the current sub-slice request, S
a,b (i) =1 denotes that the ith frequency slot is idle;
the calculation method is divided into 3 cases: requesting allocation of the block b at the t-th frequency slot in the current subslice
t When the initial index value of the spectrum window is 1, the cost value is determined by the occupation condition of the end frequency slot of the spectrum block; when the sub-slice requests allocation in the t-th frequency slot block b
t When the frequency spectrum window ending frequency slot is positioned at the rightmost side of the link frequency slot, the cost value is determined by the occupation condition of the adjacent frequency slot at the right side; requesting allocation of the t-th frequency slot block b in the current subslice
t When the index values of the start frequency slot and the end frequency slot of the frequency spectrum window are positioned in the middle of the link frequency slot, the distribution cost is determined by the occupation conditions of the adjacent frequency slots on the left side and the right side of the frequency spectrum block to be occupied by the sub-slices;
s9: according to the total assigned cost value of the service slices, the combination of the frequency spectrum windows is arranged in an ascending order; then, sequentially calculating the crosstalk value between cores after the combination of the frequency spectrum windows is allocated to the sub-slices, if the crosstalk value between cores is smaller than a crosstalk threshold value, allocating the combination of the frequency spectrum windows to each sub-slice service, and ending the algorithm; and if the crosstalk values among the cores of all the sequenced spectral window combinations do not meet the requirement of the crosstalk threshold, the resource allocation of the sub-slices fails, the service request is blocked, and the algorithm is ended.
Wherein, the calculation formula of the crosstalk between the cores is as follows:
s10: selecting fiber cores and frequency spectrum blocks for the service from high priority according to the priority of the selection of the fiber core groups, and calculating a candidate path P according to a calculation formula of the crosstalk value between the cores
i The bearer service is
Interchip crosstalk value in candidate spectral window
Turning to S11;
s11: comparing different candidate spectral windows
Selection for service request
The spectral window of the minimum value, judging
Whether the value meets the crosstalk threshold requirement under the current modulation format or not is judged, if yes, the S13 is carried out, and if not, the S12 is carried out;
s12: checking the spectrum resource state of the remaining candidate paths in the set P, and turning to S10: if k > | P |, where | P | represents the number of candidate paths, the request is blocked;
s13: and successfully distributing the resources requested by the service, outputting the path, the fiber core and the spectrum block selected by the service, and updating the number of the available spectrum slicers of the nodes according to the requirement and the spectrum resource occupation state on each fiber core.
Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, while the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.