CN116545853A - Integrated network multi-objective optimized resource management method based on quantum particle swarm - Google Patents
Integrated network multi-objective optimized resource management method based on quantum particle swarm Download PDFInfo
- Publication number
- CN116545853A CN116545853A CN202310806716.XA CN202310806716A CN116545853A CN 116545853 A CN116545853 A CN 116545853A CN 202310806716 A CN202310806716 A CN 202310806716A CN 116545853 A CN116545853 A CN 116545853A
- Authority
- CN
- China
- Prior art keywords
- integrated network
- data stream
- data
- network node
- particle swarm
- 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
Links
- 239000002245 particle Substances 0.000 title claims abstract description 48
- 238000007726 management method Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000033001 locomotion Effects 0.000 claims abstract description 9
- 238000004891 communication Methods 0.000 claims abstract description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 36
- 238000013468 resource allocation Methods 0.000 claims description 18
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000006399 behavior Effects 0.000 claims description 3
- 238000013439 planning Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 2
- 239000002096 quantum dot Substances 0.000 claims 1
- 230000006870 function Effects 0.000 abstract description 11
- 238000005457 optimization Methods 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 description 6
- 230000001934 delay Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000010152 pollination Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 108090000623 proteins and genes Proteins 0.000 description 3
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 206010003694 Atrophy Diseases 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 241001134453 Lista Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000037444 atrophy Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010154 cross-pollination Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000005610 quantum mechanics Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an integrated network multi-objective optimized resource management method based on a quantum particle swarm, which comprises the following steps: by analyzing data information of deterministic service and combining with motion trail of integrated network node, a dynamic integrated network node communication system model based on multiple services and multiple demands is established; and defining packet loss rate, time delay and service acceptance rate as objective functions, establishing a problem constraint model, and solving a network slice optimization scheme under different service requirements by improving a quantum particle swarm method. The invention can effectively improve the service acceptance rate and reduce the service packet loss rate on the basis of meeting the QoS requirement of deterministic service at best.
Description
Technical Field
The invention relates to the field of integrated network node network resource management, in particular to an integrated network multi-target optimized resource management method based on a quantum particle swarm.
Background
With the development of an integrated information network, deterministic service is rapidly increased, and the QoS requirement of the service is more and more difficult to be ensured due to the limited resources of the integrated network nodes. In performing resource allocation, various resources, such as bandwidth resources, computing resources, storage resources, and the like, need to be considered. And differentiated services have different demands for each resource. Therefore, different resource requirements of different services are considered to ensure the rationality and effectiveness of resource allocation.
In order to meet the requirements for deterministic traffic while performing resource allocation, expert scholars have proposed the concept of network slicing. Logically, each Network slice is an independent end-to-end Network, and is composed of a set of Network Functions (NFs) and corresponding resources, and is optimized for specific traffic scenarios, and provides end-to-end custom services as required. Because the artificial intelligence method is adopted to construct the network slice, the solving time is too long, and the time delay requirement of deterministic service in the actual world integration network is difficult to meet. Compared with an artificial intelligence method, the heuristic algorithm can obtain a better network slicing scheme in a limited time.
Disclosure of Invention
The invention aims to provide an integrated network multi-objective optimized resource management method based on a quantum particle swarm, which is used for meeting the time delay and bandwidth requirements of deterministic service, improving the acceptance rate of the service and reducing the network packet loss rate.
The technical scheme for realizing the purpose of the invention is as follows: in a first aspect, the present invention provides a method for managing multi-objective optimized resources of an integrated network based on a quantum particle swarm, including:
establishing an integrated network node communication system module for simulating the real integrated network node movement track by analyzing the data information of deterministic service and combining with the integrated network node movement track, and obtaining a route planning strategy in each time period through the track;
the method comprises the steps of establishing a packet loss rate, a time delay and a service acceptance rate construction module, wherein the construction module is used for constructing the relations between the three resources including bandwidth, storage and calculation, the packet loss rate and the service acceptance rate, and establishing a problem constraint model;
and solving resource allocation strategies under different routing strategies through a Levy flight improved quantum particle swarm algorithm.
In a second aspect, the invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of the first aspect when the program is executed.
In a third aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
(1) Because the quantum particle swarm algorithm is adopted, the network slice configuration can be dynamically adjusted according to the change of service requirements and the change of the motion topology of the integrated network node, and the optimal network slice scheme is found, so that the service acceptance rate can be improved;
(2) The updating step length of the particles is controlled by adopting the Levy flight random walking strategy, and compared with the probability density function control step length, the particle optimizing strategy is more met, so that the convergence speed of an algorithm can be improved.
Drawings
Fig. 1 is a schematic diagram of a quantum particle swarm solution process.
Fig. 2 is a graph of a multi-algorithm convergence effect versus.
Fig. 3 is a graph of algorithmic average packet loss versus time.
Fig. 4 is a graph of algorithmic average traffic acceptance versus.
Fig. 5 is a graph of packet loss ratio versus amount of data.
Fig. 6 is a graph of traffic acceptance rate versus amount of data.
Detailed Description
An integrated network multi-target optimized resource management method based on quantum particle swarm comprises a multi-service multi-demand integrated network node communication system module, a packet loss rate, time delay and service acceptance rate construction module and a quantum particle swarm solving module.
The integrated network node communication system module is used for simulating a real integrated network node movement track, further obtaining a route planning strategy in each time period through the track, and further providing path constraint for resource allocation; the packet loss rate, time delay and service acceptance rate construction module is used for constructing the relation between 3 resources and the packet loss rate and service acceptance rate, and 2 indexes can objectively evaluate the quality of the resource allocation scheme; the quantum particle swarm solving module solves the resource allocation scheme through a quantum particle swarm algorithm, defines the problem constraint model by taking the packet loss rate and the service acceptance rate as objective functions, and finally solves the resource allocation scheme to ensure the rationality of resource allocation, reduce the packet loss rate and improve the service acceptance rate.
Further, the running period of the integrated network node network is T, and the T is divided into time intervalsAssuming an initial time of operation of the network of integrated network nodes asAt time intervals, the integrated network nodeWhich can be regarded as a static network topology in which data flows exist during a time intervalAlternate path, willNodes traversed by the alternate paths are denoted as a setWhereinIs thatNumber of nodes traversed by the alternate path.Each node among the nodes isThe number of times the alternative path passes is recorded as a set。
Further, it is assumed that there is a batch of data streams to be processed in the controller, which is usedThe representation, wherein,the number of unknown data streams to be processed currently. Will be the firstThe bar data stream is represented as. Wherein the method comprises the steps ofThe categories representing the current data stream are classified into 3 categories: delay sensitive, bandwidth sensitive and normal data streams,indicating in which group the data stream is transmitted,indicating the amount of data to be transmitted,indicating the minimum delay that the data stream can tolerate,indicating the maximum delay that the data stream can tolerate,representing a data stream mayThe minimum bandwidth to be accepted is set,indicating the maximum bandwidth that the data stream can accept.
Further, let the time required for the transmission of the data stream beThe data stream is on a time sliceAnd when no transmission is completed, adding the residual data stream to the next time slice for processing. The path scheme of the batch of data flows at the low-rail integrated network node is expressed as a set. The nodes traversed by each data stream at the corresponding path scheme are represented asWhereinIn order to determine the number of nodes to be traversed,represent the firstAnd the number of the integrated network node. The packet loss rate, time delay and service acceptance rate building module respectively represents the calculation, bandwidth and storage resource availability on the path as,,To represent. Establishing time delay through 3 kinds of resourcesModel, packet loss rateModel, business acceptance rateAnd (5) a model.
Furthermore, the quantum particle swarm solving module can take the topology structure and the data volume of the integrated network node as input in the t-th iteration, and adjust the network slicing scheme of 3 resources according to 3 step search strategies based on Levy fly behavior, so that the service acceptance rate is improved and the service packet loss rate is reduced on the basis of meeting the QoS requirement of deterministic service in a best effort manner.
The present invention will be described in detail with reference to the accompanying drawings.
Due to the explosive growth of deterministic traffic, the integrated network nodes can allocate a diversity of resources. How to consider constructing a multi-resource network slicing scheme on the premise of various requirements of deterministic traffic becomes a working focus of resource allocation. The method comprises the steps of firstly establishing an integrated network node motion snapshot model and a diversified demand model of deterministic service, and then modeling and optimizing a multi-resource network slice.
A. Integrated network node movement snapshot model and deterministic service model
The iridium-like star base is adopted as an integrated network node network model, and the operation period of the integrated network node network is assumed to beWill beDivided into time intervals ofTime slices of integrated network node network operation initial timeTime intervalCan be regarded as a stationary integrated network node network topology, in which the data flows are present during a time intervalAlternate path, willNodes traversed by the alternate paths are denoted as a setWhereinIs thatNumber of nodes traversed by the alternate path.Each node among the nodes isThe number of times the alternative path passes is recorded as a set。
Assuming that there is a batch of data streams to be processed in the controller, it is usedThe representation, wherein,the number of unknown data streams to be processed currently. Will be the firstThe bar data stream is represented as. Wherein the method comprises the steps ofThe categories representing the current data stream are classified into 3 categories: 1. a time delay sensitive Type (TS) 2, a bandwidth sensitive type (BS) 3, and a common data stream (CD), wherein the time delay sensitive type data stream has definite minimum time delay requirement, and the requirement on the bandwidth is not high because the data volume is smaller; the bandwidth sensitive data stream has higher bandwidth requirement due to the larger data volume per se and the used bandwidth is more; the common data flow has no higher time delay and bandwidth requirement, and the QoS requirement of the data flow is satisfied as much as possible.Indicating in which group the data stream is transmitted,indicating the amount of data to be transmitted,indicating the minimum delay that the data stream can tolerate,indicating the maximum delay that the data stream can tolerate,indicating the minimum bandwidth that the data stream can accept,indicating the maximum bandwidth that the data stream can accept.
B. Time delay modeling
Assuming that the time required for this data stream transmission isThe data stream is on a time sliceAnd when no transmission is completed, adding the residual data stream to the next time slice for processing. The path scheme of the batch of data flows at the low-rail integrated network node is expressed as a set. The nodes traversed by each data stream at the corresponding path scheme are represented asWhereinIn order to determine the number of nodes to be traversed,represent the firstAnd the number of the integrated network node. The computation, bandwidth, and storage resource availability on the path are respectively expressed as,,Wherein the firstThe candidate paths are marked as,The computing resources allocated for a data stream are represented asBandwidth resources are expressed asThe storage resources are represented as. Data flowOn the pathResource allocation on is as. After the resources are allocated, the time delay can be calculatedThree main delay components in the transmission process are mainly considered: and (5) transmitting delay, propagation delay and calculating delay. Calculating time delayThe formula is as follows:
wherein,,indicating that the ith data stream is passingThe sum of the calculated delays accumulated by the individual nodes,indicating that the ith data stream is passingThe sum of the propagation delays accumulated for the links,indicating that the ith data stream has passedThe sum of the transmission delays accumulated by the individual nodes,in order to calculate the time delay,for the data size of the ith data stream,is the firstThe stripe data stream is at the firstThe first candidate pathThe size of the computing resources allocated by the individual nodes;in order to transmit the time delay,in the ith data streamThe first candidate pathThe size of bandwidth resources allocated by the individual nodes;in order for the propagation delay to be sufficient,is the distance between two adjacent nodes on a path,is the speed of light.
C. Packet loss rate and traffic acceptance rate modeling
The resource allocation is followed by considering the packet loss, and in a time interval, waiting for the incoming packet size to be larger than the allocated storage resource size.
First we calculate the firstThe stripe data stream is at the firstThe first candidate pathUnder each node, the size of a data packet waiting to be input in the current time sliceAnd then, according to the difference value between the size of the data packet waiting to be input and the allocated storage resource, the packet loss size is obtained. Wherein,,represent the firstThe stripe data stream is at the firstThe size of the data packet pre-lost by the individual nodes,represent the firstThe stripe data stream is at the firstThe size of the number of data packets actually lost under the candidate path,represent the firstThe amount of data to be transmitted by the stripe data stream,representation ofAverage packet loss rate size for a stripe data stream.
Consider the traffic acceptance rate as the firstThe closer the actual allocated bandwidth resources of the stripe data stream are to the bandwidth requirements of the data stream, and the closer the actual delay is to the delay requirements of the data stream, the higher the acceptance rate of the service.
The actually allocated bandwidth resource is considered as the bandwidth size of the node with the smallest allocated bandwidth resource among all nodes on a path, and is usedTab listAs shown in the drawing,。indicating the minimum delay that the data stream can tolerate,indicating the maximum delay that the data stream can tolerate,indicating the minimum bandwidth that the data stream can accept,indicating the maximum bandwidth that the data stream can accept.Representation ofThe average of the difference between the bandwidth resources actually allocated by the data stream and the bandwidth requirements of the data stream,representing the average of the difference between the actual delays of the s data streams and the delay requirement of the data streams.
In the above-mentioned method, the step of,representing the variance of the difference between the bandwidth resources actually allocated by the s data streams and the bandwidth requirements of the data streams.Representing the actual delay of s data streams and the delay requirement difference of the data streamsVariance of values.
Service acceptance rateRepresented as a normalized representation of the variance of the actual allocated bandwidth resources and delays in the s data streams and the difference in the bandwidth requirements and delay requirements of the data streams.,As the weight value of the weight,. Finally, the resource allocation problem is considered as a multi-constraint multi-dimensional resource optimization problem.
The objective function is:
the constraint conditions are as follows:
wherein,,is the firstThe stripe data stream is at the firstThe first candidate pathThe size of the computing resources allocated by the individual nodes,is the firstThe stripe data stream is at the firstThe first candidate pathThe individual nodeThe size of bandwidth resources allocated on the links between the individual nodes,is the firstThe stripe data stream is at the firstThe first candidate pathThe size of the storage resources allocated by the individual nodes,representing an integrated network node at a firstThe total computation resources available to the individual nodes,representing an integrated network node at a firstThe total memory resources available to the individual nodes,representing an integrated network node at a firstThe individual nodeTotal bandwidth resources available on links between individual nodes; the constraint condition ensures that the allocated resources do not exceed the total amount of resources which can be allocated currently and practically;
is indicated at +.>The third part of the path scheme>The size of the computing resource actually allocated by the individual node, < >>Is indicated at +.>The third part of the path scheme>The size of the storage resources actually allocated by the individual nodes; the constraint condition ensures that the computing and storage resources allocated to each path scheme passing through the node on one node are equal to the total computing resources and storage resources of the node;
is shown in the firstPath scheme pair numberThe individual nodeThe size of the bandwidth resources actually allocated on the links between the individual nodes; the constraint condition ensures that the bandwidth resources allocated between nodes for each path scheme passing through the node are equal to the total bandwidth resources of the link.
D. Solution of quantum particle swarm algorithm
The standard PSO algorithm has high convergence speed, but for an objective function with high dimension, the solving result is very easy to fall into local optimum. The global searching capability of the QPSO algorithm is far better than that of the general PSO algorithm. In the QPSO algorithm, since the particles refer to the quantum uncertainty principle in quantum mechanics and have quantum behaviors, the precise values of the position and the velocity of the particles cannot be determined at the same time. The updating of particles in the QPSO algorithm is to obtain new individuals through observation, firstly observe by utilizing the Monte Carlo idea to obtain a plurality of individual positions, then select the optimum individual, and evaluate the rest individuals in turn to finally obtain next generation individuals, wherein the QPSO solving process is shown in figure 1. The QPSO evolution equation is as follows:
wherein N is the population number, t is the iteration number;for the optimal average position of the population,at objective function for the ith particleIs provided with a history of optimal positions in the database,is an updated formula for a local step size,in objective function for all particles in the populationIs provided with a plurality of the optimal positions,is the firstThe current position of the individual particles is determined,is the next time position of the ith particle;,andall satisfy the normal distribution of (0, 1), namely.
The improved quantum particle swarm method based on flower pollination provided by the invention is as followsThe improvement from random numbers meeting (0, 1) normal distribution to a partial step length solving formula of cross pollination in a pollination algorithm is as follows:
wherein,,representing a standard gamma function, whereinThe step length solved by the formula is applied to an improved quantum particle swarm algorithm, so that the convergence speed and the convergence precision can be effectively improved.
In the QPSO algorithm, the historical optimal position of particles is continuously improved, and the historical optimal position is closely related to the update equation of the local attractors.The 'excellent gene' is obtained, so that the 'excellent' is changed, but the 'attribute' of the particles is destroyed to a certain extent, so that the 'atrophy' is caused to the diversity of the whole population, and the particles of the population are premature 'mature'. Therefore, a multi-strategy local step update mechanism is proposed. Particle-settingThe optimal particle position in the neighborhood of (2) is noted asThe invention proposes 3 search strategy update equations for adjusting local step sizes as follows:
the above is the original local step update equation in the QPSO algorithm, representing the absorption of the "excellent genes" from the optimal position of the population;
the above-described substitution of the global optimum for the neighborhood optimum represents the absorption of the "excellent genes" from the neighborhood optimum, wherein,。
the multi-strategy local attractor updating equation provided by the invention enriches the direction of particle evolution and can effectively inhibit particle premature.
Multi-strategy local step length updating algorithm
Input of population number N, globally optimal particlesIndividual history optimal particles,Particle neighborhood optimization particle,Weighting of
And (3) outputting: local step size of particles
The process comprises the following steps:
for do
=rand
=rand
if
else if
else
end if
end for
examples
The invention uses the digital experimental result to verify the effectiveness of the proposed algorithm, firstly considers the integrated network node topology structure under each time slice, and calculates the selectable path under each topology structure. Based on the selectable paths and the remaining resources of each integrated network node, network slicing schemes under deterministic traffic diversity requirements are calculated.
According to the invention, an iridium-like star seat simulation is built based on MATLAB, the constellation is provided with 6 track planes, each track is provided with 11 integrated network nodes, 66 integrated network nodes are uniformly distributed, the track height is 780km, and the track inclination angle is 86.4 degrees. The algorithm iterates 100 times, the final iteration result is used as the current network slicing scheme, and the packet loss rate of the network system is calculatedService acceptance rateThe system simulation parameters are shown in table 1.
Table 1 simulation parameters
In the parameter setting of the data flow, in 20 data packets average per minute, the time delay sensitive service accounts for 10 percent, the bandwidth sensitive service accounts for 30 percent, the common service accounts for 60 percent, the time delay requirement of the time delay sensitive service is set to be 100,150 Ms, the bandwidth requirement is set to be 5,10 Mbps, and the data volume is set to be 10,100 Mb. The delay requirement of the bandwidth sensitive service is set to [300,550] Ms, the bandwidth requirement is set to [25,70] Mbps, and the data amount is set to [500,2000] Mb. The delay requirement of the common service is set to [150,500] Ms, the bandwidth requirement is set to [10,25] Mbps, and the data volume is set to [100,500] Mb. The experiment was conducted with a solution of the resource allocation scheme for 10 different time slices, each time slice was set to 5 minutes, and the solution results were averaged.
As shown in fig. 2, PSO is a standard particle swarm algorithm, NDPSO is a nonlinear dynamic particle swarm algorithm, IQPSO is a modified quantum particle swarm algorithm, and FIQPSO is a flower pollination-based quantum particle swarm algorithm of the present invention. And solving the network slicing scheme of 10 different time slices, averaging the solving results, taking the performance of the PSO algorithm as a reference, wherein the abscissa represents the iteration times, and the ordinate represents the objective function value calculated by each iteration. As can be seen from the graph, the NDPSO algorithm has relatively slow solving time in the initial stage of simulation, but has certain capability of jumping out of a local optimal solution along with the progress of the simulation, so that the NDPSO algorithm is superior to classical PSO in the final calculation result. The IQPSO algorithm and the FIQPSO algorithm have better convergence speed in the initial simulation stage, and the FIQPSO algorithm provided herein can further improve the convergence speed and finally obtain better objective function values, namely a better resource allocation scheme, than other methods.
As shown in fig. 3, after the resource allocation scheme in the time slice is obtained, the size of the packet loss rate is calculated according to the time delay, the bandwidth requirement and the allocated storage resource condition of different data packets in the time slice. From the results, the FIQPSO algorithm through the Levy flight control step length can reduce the packet loss rate by 4.7% compared with the IQPSO algorithm.
As shown in fig. 4, after the resource allocation scheme in the time slice is obtained, the variance of the bandwidth demand difference between the minimum bandwidth resource allocated on each path in the time slice and the data flow is calculated, and finally normalization is performed to represent the service acceptance rate. From the results, the FIQPSO algorithm through the Levy flight control step can improve the service acceptance rate by 5.1% compared with the IQPSO algorithm.
As shown in fig. 5, the packet loss ratio comparison of different time slices and different packet sending sizes is shown, and the network slice optimization algorithm provided by the invention has good effect within 8000MB of data size. However, when 10000MB of data is transmitted, the data size approaches the bearing capacity of the node due to limited resources of the integrated network node, so that the optimization effect is not obvious when the data size is too large.
As shown in fig. 6, compared with service receiving rates of different time slices and different packet sending sizes, the service receiving rate is severely reduced when the data volume is too large in a similar situation as the packet loss rate in fig. 5, and the improved quantum particle swarm algorithm provided by the invention has good effect within the bearing capacity of the node.
While the invention has been described with reference to the 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 spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. The integrated network multi-target optimized resource management method based on the quantum particle swarm is characterized by comprising the following steps of:
establishing an integrated network node communication system module for simulating the real integrated network node movement track by analyzing the data information of deterministic service and combining with the integrated network node movement track, and obtaining a route planning strategy in each time period through the track;
the method comprises the steps of establishing a packet loss rate, a time delay and a service acceptance rate construction module, wherein the construction module is used for constructing the relations between the three resources including bandwidth, storage and calculation, the packet loss rate and the service acceptance rate, and establishing a problem constraint model;
and solving resource allocation strategies under different routing strategies through a Levy flight improved quantum particle swarm algorithm.
2. The method for multi-objective optimized resource management of an integrated network based on quantum-dot groups according to claim 1, wherein the operation period of the integrated network node network is T, and T is divided into time intervals of time intervalsAssuming an initial time of operation of the integrated network node network of +.>The integrated network node is +.>The network topology is regarded as a static network topology, in which data flows are present in a time interval>Alternate pathWill->The nodes through which the alternative paths pass are denoted as set +.>Wherein->Is->The number of nodes traversed by the selectable path; />In the individual nodes, each node is +.>The number of times the alternative path is passed is denoted as set +.>。
3. The method for multi-objective optimized resource management of an integrated network based on quantum-particle swarm according to claim 2, wherein it is assumed that there is a batch of data streams to be processed in the controller, and it is usedRepresentation, wherein->The number of the unknown data streams to be processed currently is the number; will be->The bar data stream is denoted->The method comprises the steps of carrying out a first treatment on the surface of the Wherein->The categories representing the current data stream are classified into 3 categories: delay-sensitive, bandwidth-sensitive, normal data stream,/->Indicating in which group the data stream is transmitted, < ->Representing the amount of data to be transmitted,/-, for example>Representing the minimum delay that the data stream can tolerate, < +.>Represents the maximum delay that the data stream can tolerate, < +.>Representing the minimum bandwidth acceptable for the data stream, < >>Indicating the maximum bandwidth acceptable to the data stream.
4. The method for multi-objective optimized resource management of a quantum-based integrated network according to claim 1, wherein the time required for the transmission of the data stream is set asThe data stream is in time slice +.>When the transmission is not completed, the residual data stream is added to the next time slice for processing; the path scheme of the batch of data flows at the low-rail integrated network node is expressed as a set +.>The method comprises the steps of carrying out a first treatment on the surface of the The node traversed by each data stream in the corresponding path scheme is denoted +.>Wherein->For the number of nodes passed->Indicate->Numbering of individual integrated network nodes; the calculation, bandwidth and storage resource availability on the path are respectively expressed as +.>,/>,The method comprises the steps of carrying out a first treatment on the surface of the And establishing a time delay model, a packet loss rate model and a service acceptance rate model through 3 resources.
5. The integrated network multi-objective optimized resource management method based on quantum particle swarm according to claim 1, wherein the quantum particle swarm solving module is capable of adjusting network slicing schemes of 3 resources according to 3 step search strategies based on Levy fly behavior by taking an integrated network node topology and data amount as inputs in the t-th iteration.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-5 when the program is executed.
7. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of the method according to any of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310806716.XA CN116545853B (en) | 2023-07-04 | 2023-07-04 | Integrated network multi-objective optimized resource management method based on quantum particle swarm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310806716.XA CN116545853B (en) | 2023-07-04 | 2023-07-04 | Integrated network multi-objective optimized resource management method based on quantum particle swarm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116545853A true CN116545853A (en) | 2023-08-04 |
CN116545853B CN116545853B (en) | 2023-10-13 |
Family
ID=87449133
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310806716.XA Active CN116545853B (en) | 2023-07-04 | 2023-07-04 | Integrated network multi-objective optimized resource management method based on quantum particle swarm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116545853B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118433206A (en) * | 2024-07-04 | 2024-08-02 | 山东木丁信息技术有限公司 | Data efficient storage method, medium and device based on big data analysis |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699446A (en) * | 2013-12-31 | 2014-04-02 | 南京信息工程大学 | Quantum-behaved particle swarm optimization (QPSO) algorithm based multi-objective dynamic workflow scheduling method |
CN104536828A (en) * | 2014-12-26 | 2015-04-22 | 湖南强智科技发展有限公司 | Cloud computing task scheduling method and system based on quantum-behaved particle swarm algorithm |
CN105960024A (en) * | 2016-06-08 | 2016-09-21 | 北京邮电大学 | User discovery and resource allocation method based on social perception in D2D communication |
CN108809860A (en) * | 2018-05-03 | 2018-11-13 | 北京航空航天大学 | Network node resource distribution method based on service reliability |
CN108880663A (en) * | 2018-07-20 | 2018-11-23 | 大连大学 | Incorporate network resource allocation method based on improved adaptive GA-IAGA |
CN111314889A (en) * | 2020-02-26 | 2020-06-19 | 华南理工大学 | Task unloading and resource allocation method based on mobile edge calculation in Internet of vehicles |
US20210293881A1 (en) * | 2017-11-09 | 2021-09-23 | Hefei University Of Technology | Vector-valued regularized kernel function approximation based fault diagnosis method for analog circuit |
CN113572686A (en) * | 2021-07-19 | 2021-10-29 | 大连大学 | Heaven and earth integrated self-adaptive dynamic QoS routing method based on SDN |
CN114466407A (en) * | 2022-01-04 | 2022-05-10 | 中国船舶重工集团公司第七一六研究所 | Network slice arranging algorithm based on particle swarm heredity |
CN115633033A (en) * | 2022-10-08 | 2023-01-20 | 南京邮电大学 | Cooperative energy-saving calculation migration method integrating radio frequency energy acquisition |
-
2023
- 2023-07-04 CN CN202310806716.XA patent/CN116545853B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699446A (en) * | 2013-12-31 | 2014-04-02 | 南京信息工程大学 | Quantum-behaved particle swarm optimization (QPSO) algorithm based multi-objective dynamic workflow scheduling method |
CN104536828A (en) * | 2014-12-26 | 2015-04-22 | 湖南强智科技发展有限公司 | Cloud computing task scheduling method and system based on quantum-behaved particle swarm algorithm |
CN105960024A (en) * | 2016-06-08 | 2016-09-21 | 北京邮电大学 | User discovery and resource allocation method based on social perception in D2D communication |
US20210293881A1 (en) * | 2017-11-09 | 2021-09-23 | Hefei University Of Technology | Vector-valued regularized kernel function approximation based fault diagnosis method for analog circuit |
CN108809860A (en) * | 2018-05-03 | 2018-11-13 | 北京航空航天大学 | Network node resource distribution method based on service reliability |
CN108880663A (en) * | 2018-07-20 | 2018-11-23 | 大连大学 | Incorporate network resource allocation method based on improved adaptive GA-IAGA |
CN111314889A (en) * | 2020-02-26 | 2020-06-19 | 华南理工大学 | Task unloading and resource allocation method based on mobile edge calculation in Internet of vehicles |
CN113572686A (en) * | 2021-07-19 | 2021-10-29 | 大连大学 | Heaven and earth integrated self-adaptive dynamic QoS routing method based on SDN |
CN114466407A (en) * | 2022-01-04 | 2022-05-10 | 中国船舶重工集团公司第七一六研究所 | Network slice arranging algorithm based on particle swarm heredity |
CN115633033A (en) * | 2022-10-08 | 2023-01-20 | 南京邮电大学 | Cooperative energy-saving calculation migration method integrating radio frequency energy acquisition |
Non-Patent Citations (2)
Title |
---|
MEITONG CHEN: "Micro grid scheduling optimization based on quantum particle swarm optimization(QPSO)algorithm", 2018 CHINESE CONTROL AND DECISION CONFERENCE * |
赵莉;: "基于改进量子粒子群算法的云计算资源调度", 南京理工大学学报, vol. 40, no. 02, pages 223 - 228 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118433206A (en) * | 2024-07-04 | 2024-08-02 | 山东木丁信息技术有限公司 | Data efficient storage method, medium and device based on big data analysis |
Also Published As
Publication number | Publication date |
---|---|
CN116545853B (en) | 2023-10-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110717300B (en) | Edge calculation task allocation method for real-time online monitoring service of power internet of things | |
CN114143264B (en) | Flow scheduling method based on reinforcement learning under SRv network | |
CN116545853B (en) | Integrated network multi-objective optimized resource management method based on quantum particle swarm | |
CN111984419B (en) | Complex task computing migration method for edge environment reliability constraint | |
CN113225370B (en) | Block chain multi-objective optimization method based on Internet of things | |
CN107948083A (en) | A kind of SDN data centers jamming control method based on enhancing study | |
Raeis et al. | A deep reinforcement learning approach for fair traffic signal control | |
CN114710410B (en) | Network data acquisition efficiency optimization method and system based on deep reinforcement learning | |
CN116541106A (en) | Computing task unloading method, computing device and storage medium | |
CN115314399A (en) | Data center flow scheduling method based on inverse reinforcement learning | |
CN110049018B (en) | SPMA protocol parameter optimization method, system and medium based on reinforcement learning | |
Abdel-Kader | An improved discrete PSO with GA operators for efficient QoS-multicast routing | |
CN113672372B (en) | Multi-edge collaborative load balancing task scheduling method based on reinforcement learning | |
Dai et al. | Iqor-lse: An intelligent qos on-demand routing algorithm with link state estimation | |
Ikeda et al. | Performance evaluation of an intelligent CAC and routing framework for multimedia applications in broadband networks | |
CN113094857A (en) | Energy-saving controller layout method for software defined vehicle network | |
Sun et al. | Fuzzy controller based QoS routing algorithm with a multiclass scheme for MANET | |
CN107171957A (en) | A kind of adaptive DTN routing algorithms based under the conditions of resource-constrained | |
Nyirenda et al. | Multi-objective particle swarm optimization for fuzzy logic based active queue management | |
Jin et al. | Joint qos control and bitrate selection for video streaming based on multi-agent reinforcement learning | |
CN111555978B (en) | SDN routing arrangement method with energy saving and service quality guarantee functions in smart grid | |
Yang et al. | Beyond the Edge: An Advanced Exploration of Reinforcement Learning for Mobile Edge Computing, its Applications, and Future Research Trajectories | |
Sun et al. | MAMRL: Exploiting Multi-agent Meta Reinforcement Learning in WAN Traffic Engineering | |
Pandey et al. | An Exploratory Study of Optimization Techniques for Congestion Control in Wireless Sensor Networks. | |
Zhang et al. | Dynamic Routing of Multiple QoS-Required Flows in Cloud-Edge Autonomous Multi-Domain Data Center Networks. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |