CN109362093B - Resource optimization method for maximizing total throughput of network function virtualization - Google Patents

Resource optimization method for maximizing total throughput of network function virtualization Download PDF

Info

Publication number
CN109362093B
CN109362093B CN201811537839.3A CN201811537839A CN109362093B CN 109362093 B CN109362093 B CN 109362093B CN 201811537839 A CN201811537839 A CN 201811537839A CN 109362093 B CN109362093 B CN 109362093B
Authority
CN
China
Prior art keywords
total
frog
controller
throughput
logic
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.)
Active
Application number
CN201811537839.3A
Other languages
Chinese (zh)
Other versions
CN109362093A (en
Inventor
谢宁
张齐齐
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.)
Shenzhen University
Original Assignee
Shenzhen University
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 Shenzhen University filed Critical Shenzhen University
Priority to CN201811537839.3A priority Critical patent/CN109362093B/en
Publication of CN109362093A publication Critical patent/CN109362093A/en
Application granted granted Critical
Publication of CN109362093B publication Critical patent/CN109362093B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • 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
    • 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
    • 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
    • H04L41/0893—Assignment of logical groups to network elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • 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)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The present disclosure describes a resource optimization method for maximizing total throughput of network function virtualization, comprising: the network function virtualization architecture comprises a logic layer, a virtual layer, a physical layer and a management arrangement system, wherein the logic layer comprises logic nodes and a controller, and the physical layer comprises physical nodes; the logic node and the controller transmit information signals to the base station through corresponding physical nodes through orthogonal channels; the base station calculates the effective load bit number, the channel using number, the transmitting power, the receiving signal-to-noise ratio, the throughput and the total transmitting energy of each logic node and each controller based on the information signals; and when the third quantity, the number of payload bits, the received signal-to-noise ratio, the first total transmit energy and the second total transmit energy meet the corresponding requirements and the logical node and the controller meet the ultra-reliability requirements, the management scheduling system adaptively allocates transmit power and the number of used channels based on a hybrid optimization algorithm to maximize the total throughput.

Description

Resource optimization method for maximizing total throughput of network function virtualization
Technical Field
The disclosure relates to the technical field of network virtualization, in particular to a resource optimization method for maximizing total throughput of network function virtualization.
Background
As end users' demand for services with more diverse high data rates continues to increase, Telecommunication Service Providers (TSPs) must correspondingly and continuously purchase, store, and operate new physical equipment. This not only requires the technicians operating and managing the equipment to have rapidly varying skills, but also requires intensive deployment of the network equipment, which results in high operational and Capital expenditures (OPEX). Network Function Virtualization (NFV) has been proposed as a solution to these challenges, providing new methods of designing, deploying and managing Network services using Virtualization technologies, and the NFV concept is expected to serve various emerging services and vertical markets, such as industrial automation, autopilot, robotics, healthcare, virtual and augmented reality.
Resource Allocation (RA) is one of the important factors in Network Function Virtualization (NFV) deployment. Since Physical layer (PHY) resources are limited (e.g., transmit energy and channel usage), RA issues for the PHY layer for NFV deployment have become a rapidly growing issue, especially for Ultra Reliable Low Latency Communications (URLLC). In addition, since the problem of optimizing the maximum total throughput in the Uplink (UL) transmission protocol deployed in NFV is highly non-convex, it is difficult to obtain a global optimal solution, and since an over-constraint condition in the optimization problem may cause a low convergence efficiency phenomenon, the conventional heuristic algorithm cannot directly solve the resource optimization problem of maximizing the total throughput.
Disclosure of Invention
The present disclosure is made to solve the above conventional problems, and an object of the present disclosure is to provide a resource optimization method capable of maximizing total throughput of network function virtualization, which can quickly and efficiently complete a problem of maximizing total throughput optimization in NFV deployment.
Therefore, the present disclosure provides a resource optimization method for maximizing total throughput of network function virtualization, which includes a network function virtualization architecture and a resource optimization method for maximizing total throughput of a base station, and is characterized by including: the network function virtualization architecture comprises a logic layer, a virtual layer, a physical layer and a management and orchestration system, wherein the logic layer comprises a first number of logic nodes and a second number of controllers, the physical layer comprises a third number of physical nodes, the third number is equal to the sum of the first number and the second number, the logic nodes and the controllers form a mapping relation with the physical nodes through the virtual layer, and the logic nodes and the controllers send information signals to the base station through the corresponding physical nodes through orthogonal channels; the base station calculates the number of payload bits, the number of used channels, the transmission power, the received signal-to-noise ratio and the throughput of each logic node and each controller based on the information signal, calculates first total transmission energy and first total throughput of a plurality of the logic nodes, and calculates second total transmission energy and second total throughput of a plurality of the controllers; and when the third number is equal to a preset service number, the number of payload bits is equal to a preset number of bits, and the received signal-to-noise ratio is equal to a preset received signal-to-noise ratio, the logical nodes and the controller meet a super-reliability requirement, the first total transmit energy is not greater than a first energy threshold and the second total transmit energy is not greater than a second energy threshold, and the number of channel usages is not greater than a total number of channel usages, the management orchestration system adaptively allocates the transmit power and the number of channel usages based on a hybrid optimization algorithm to maximize a total throughput, the total throughput being a sum of the first total throughput and the second total throughput, wherein the hybrid optimization algorithm is a hybrid frog-extreme value optimization algorithm that is multidimensional and takes the total throughput as an adaptive value, the multidimensional including the transmit power of each logical node and the transmit power of each controller, and a number of channel usages of each of the logical nodes and each of the controllers.
In the disclosure, based on a network function virtualization architecture, a logic node and the controller send information signals to a base station, the base station receives the information signals and calculates related parameters, and when a third quantity, a payload bit number, a received signal-to-noise ratio, a first total transmission energy and a second total transmission energy satisfy respective requirements and the logic node and the controller satisfy a super-reliability requirement, a management and deployment system adaptively allocates transmission power and a channel usage quantity based on a hybrid optimization algorithm to maximize a total throughput. Therefore, the problem of optimizing the maximum total throughput in NFV deployment can be quickly and effectively solved, and the method has better global search capability.
In an optimization method to which the present disclosure relates, optionally, the total throughput is satisfied
Figure BDA0001907343690000031
The throughput of the ith logical node and the jth controller are respectively satisfied
Figure BDA0001907343690000032
And
Figure BDA0001907343690000033
wherein k isLiIs the number of payload bits, k, of the ith logical nodeCjFor the j controller to be effectiveNumber of bits of payload, Pt,LiIs the transmission power, P, of the ith logical nodet,CjIs the transmission power of the jth controller, nLiNumber of channel usages, n, for ith logical nodeCjThe number of channel usages for the jth controller. Thereby, the total throughput can be obtained.
In the optimization method to which the present disclosure relates, optionally, the super-reliability requirement is satisfied
Figure BDA0001907343690000034
And
Figure BDA0001907343690000035
and is
Figure BDA0001907343690000036
Wherein p ise,LiPacket error probability, p, for the ith logical nodee,CjThe packet error probability for the jth controller,
Figure BDA0001907343690000037
is the upper limit of the packet error probability of any one of the logical nodes,
Figure BDA0001907343690000038
is the upper limit of the packet error probability of any one of said controllers. Thus, the connection between any one controller and the base station has higher reliability than the connection between any one logical node and the base station.
In the optimization method related to the present disclosure, optionally, the transmission energies of the ith logical node and the jth controller are respectively satisfied with ELi＝Pt,LinLiAnd ECj＝Pt,CjnCjWherein E isLiAnd ECjAll units of (a) are W.Hz.s, Pt,LiIs the transmission power, P, of the ith logical nodet,CjIs the transmission power of the jth controller, nLiNumber of channel usages, n, for ith logical nodeCjThe total number of channel usage for the jth controller satisfies
Figure BDA0001907343690000039
M is the first number and N is the second number. Thereby, the transmission energy of each logical node and each controller and the total number of channel usages can be obtained.
In the optimization method according to the present disclosure, optionally, the input parameters of the mixed frog-extremum optimization algorithm include the first number M, the second number N, and the number k of payload bits of each of the logic nodesLiNumber of payload bits k for each of the controllersCjReceiving signal-to-noise ratio gamma of each logic nodeLiReceiving signal-to-noise ratio gamma of each controllerCjUpper limit of packet error probability of the logical node
Figure BDA00019073436900000310
Upper limit of packet error probability of the controller
Figure BDA00019073436900000311
The first energy threshold
Figure BDA00019073436900000312
The second energy threshold
Figure BDA00019073436900000313
And said total number of channel uses n∑The output parameter includes the total throughput RΣTransmitting power P of each logic nodet,LiThe transmission power P of the controllert,CjNumber of channel usages n of each logical nodeLiAnd the number of channel usages n of each of the controllersCj. Thus, the mixed frog-leap-extremum optimization algorithm can be optimized based on the input parameters and obtain optimized output parameters.
In the optimization method according to the present disclosure, optionally, the mixed frog-leap-extremum optimization algorithm includes: setting initialization parameters; randomly generating a population p including F frogs; x with dimension t ═ 2(M + N) for each frog positioniRepresents; tong (Chinese character of 'tong')The over-evaluation algorithm calculates the adaptive value f (X) of each frogi) (ii) a Judging whether a convergence criterion is met; when the convergence criterion is met, obtaining an optimal output parameter and ending the process; when the convergence criterion is not met, sorting the corresponding adaptive values of the F frogs according to a descending order; constructing p groups of frog and sub-factor complexes; performing for-circulation on each group of frogs, locally searching in the sub-factor complex in each circulation, calculating the fitness of each frog by using the evaluation algorithm, and performing extreme value optimization on each frog, wherein the fitness of each frog is obtained by the evaluation algorithm; all frogs are shuffled. Thereby, rapid and stable convergence can be ensured.
In the optimization method according to the present disclosure, optionally, the input parameter of the evaluation algorithm is the location X of the ith frogiThe output parameter is the adaptive value f (X) of the position of the ith frogi) Calculating the fitness f (X) of the position of each frogi) (ii) a Wherein
Figure BDA0001907343690000041
T represents a penalty coefficient, and T is 2 × 104(ii) a When in use
Figure BDA0001907343690000042
Or
Figure BDA0001907343690000043
When f (X)i)＝f(Xi) -T; when in use
Figure BDA0001907343690000044
Or
Figure BDA0001907343690000045
When f (X)i)＝f(Xi) -T; when in use
Figure BDA0001907343690000046
When f (X)i)＝f(Xi) -T; otherwise, f (X) is returnedi). Thus, an adaptation value for each frog is obtained based on the evaluation algorithm.
In the optimization method related in the present disclosure, optionally, the process of extremum optimization includes: randomly generating individuals having a plurality of components; calculating the fitness value for each component of each individual; setting the current individual as an optimal individual; judging whether the set standard is met, and ending the process when the set standard is met; when the set standard is not satisfied, calculating an adaptive value of each component of the current individual; searching for components with adaptation values less than or equal to the adaptation value of the current individual; obtaining a target individual; taking the target individual as a new current individual; and when the adaptive value of the target individual is smaller than that of the optimal individual, the target individual is taken as the optimal individual. This can improve the local search capability.
In the optimization method according to the present disclosure, optionally, the adaptive value λ of each of the componentsijSatisfy the requirement of
Figure BDA0001907343690000051
Wherein, Δ xijIs the offset between the current position and the new position of the ith frog. Thus, the fitness value of each mutated component can be obtained.
In the optimization method according to the present disclosure, optionally, the hybrid frog-extreme optimization algorithm has a running time satisfying complexity of
Figure BDA0001907343690000052
Wherein, the extremum optimizes the process jump condition NEOThe value is set to any value between 1 and 100, NiteIs the number of iterations of the extremum optimization, and NiteIs 2, lmaxRepresents the maximum iteration number of the mixed frog leaping algorithm, and t is a dimension. Thus, runtime complexity can be obtained.
The resource optimization method for maximizing the total throughput based on the mixed frog-extreme optimization algorithm (MSFLA-EO) has excellent stability and global search capability, the resource optimization method is simulated according to various performance parameters, and experimental results prove that the resource optimization method has a remarkable effect on maximizing the total throughput.
Drawings
Fig. 1 is a schematic diagram illustrating a system model of an NFV architecture to which examples of the present disclosure relate.
Fig. 2 is a flow chart illustrating a resource optimization method for overall throughput maximization for network function virtualization in accordance with an example of the present disclosure.
Fig. 3 is a flow chart illustrating a hybrid optimization algorithm of a resource optimization method of network function virtualization according to an example of the present disclosure.
Fig. 4 is a flow chart illustrating an extremum optimization process of a resource optimization method of network function virtualization according to an example of the present disclosure.
Fig. 5 is a waveform diagram illustrating the total throughput at different numbers of payload bits for a resource optimization method of network function virtualization according to an example of the present disclosure.
Fig. 6 is a waveform diagram illustrating the total throughput at different received signal-to-noise ratios for a resource optimization method of network function virtualization according to examples of the present disclosure.
Detailed Description
Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the following description, the same components are denoted by the same reference numerals, and redundant description thereof is omitted. The drawings are schematic and the ratio of the dimensions of the components and the shapes of the components may be different from the actual ones.
It is noted that the terms "comprises," "comprising," and "having," and any variations thereof, in this disclosure, for example, a process, method, system, article, or apparatus that comprises or has a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include or have other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Fig. 1 is a schematic diagram illustrating a system model of an NFV architecture to which examples of the present disclosure relate. The disclosure relates to a resource optimization method for maximizing total throughput of network function virtualization, which comprises a network function virtualization framework and a base station.
In some examples, as shown in fig. 1, a network function virtualization architecture includes a logical layer, a virtual layer, a physical layer, and a management orchestration system. The logical layer may include a first number of logical nodes and a second number of controllers. Multiple controllers coexist to achieve the resilience function of the network. The first number may be M, for example, and the second number may be N, for example.
In some examples, the physical layer may include a third number of physical nodes. The third number is equal to the sum of the first number and the second number. The third number may be, for example, X. Wherein X is M + N. The physical node may be, for example, a hardware device on which the VNF operates. The physical layer may also include physical links. The physical link may be, for example, a wired or wireless actual communication device. In addition, since NFVs may deploy redundancy to recover from software or hardware failures, resources in a physical node may serve multiple logical nodes and controllers. The logical nodes and the controller can form a mapping relation with the physical nodes through the virtual layer.
In some examples, a management orchestration system (NFV MANO) may provide the functionality required to configure VNFs and related operations. The management orchestration system may also be responsible for resource allocation throughout the network. For example, the resources may include transmission power and channel usage times.
Fig. 2 is a flow chart illustrating a resource optimization method for overall throughput maximization for network function virtualization in accordance with an example of the present disclosure. The resource optimization method for maximizing the total throughput of network function virtualization related to the present disclosure may be simply referred to as an optimization method.
In some examples, as shown in fig. 2, based on the NFV architecture described above, the optimization method may include a sending phase (step S100), a receiving calculation phase (step S200), and an optimization phase (step S300).
In step S100, the logical node and the controller may transmit an information signal to the base station through the orthogonal channel through the corresponding physical node. The orthogonal channel may be, for example, an orthogonal frequency band channel or an orthogonal time slot channel. Therefore, the co-channel interference in the information signal transmission process can be ignored.
In some examples, the connection between any one of the controllers and the base station may have a higher reliability than the connection between any one of the logical nodes and the base station. Therefore, the problems of random faults, intentional attacks, software faults and the like caused by connection failure of the controller and the base station can be reduced.
In step S200, the base station may calculate the number of payload bits, the number of channel usages, a reception signal-to-noise ratio, and a throughput of each logical node and each controller based on the information signal. Wherein, the logical nodes may include M. The relevant parameters of the ith logical node may be denoted by the subscript Li. E.g. kLiIs the number of payload bits, n, of the ith logical nodeLiNumber of channel usages, P, for the ith logical nodet,LiIs the transmission power of the ith logical node, gammaLiThe received signal-to-noise ratio of the ith logical node. The controller may include N. The relevant parameters of the jth controller may be represented by the subscript Cj. E.g. kCjNumber of payload bits, n, for jth controllerCjNumber of channel usages, P, for jth controllert,CjIs the transmission power of the jth controller, gammaCjIs the received signal-to-noise ratio of the jth controller. In addition, the total number of channel uses can be expressed as
Figure BDA0001907343690000071
Additionally, in some examples, the first throughput R of the ith logical nodeLiCan satisfy
Figure BDA0001907343690000072
And a second throughput R of the jth controllerCjCan satisfy
Figure BDA0001907343690000073
Thereby, a first throughput sum of each logical node can be obtainedA second throughput for each controller. Thereby, transmission energy, total number of channel use, and throughput of each logical node and each controller can be obtained.
In step S200, the base station may further calculate a first total transmission energy and a first total throughput of the plurality of logical nodes, and calculate a second total transmission energy and a second total throughput of the plurality of controllers.
Specifically, the base station may calculate the transmit energy E of the ith logical nodeLiAnd the emission energy E of the jth controllerCj. Wherein E isLiAnd ECjHas the unit of W.Hz.s, ELi＝Pt,LinLiAnd ECj＝Pt,CjnCj. Based on the transmission energy of each logical node and the transmission energy of each controller, the base station may calculate the total transmission energy of all logical nodes and the total transmission energy of all controllers.
Additionally, in some examples, the total throughput may be a sum of the first total throughput and the second total throughput, and the total throughput satisfies
Figure BDA0001907343690000081
In step S300, the conditions that need to be satisfied by the management orchestration system for adaptive allocation may include that the third number is equal to the preset number of services, the number of payload bits is equal to the preset number of bits, and the received signal-to-noise ratio is equal to the preset received signal-to-noise ratio. That is, the third number X may be a set preset number of services.
In addition, the management and arrangement system needs to meet the requirement of the logic node and the controller on super-reliability for self-adaptive distribution. I.e. the ultra-reliability requirement is satisfied
Figure BDA0001907343690000082
And
Figure BDA0001907343690000083
and is
Figure BDA0001907343690000084
For example
Figure BDA0001907343690000085
And
Figure BDA0001907343690000086
wherein p ise,LiPacket error probability, p, for the ith logical nodee,CjThe packet error probability for the jth controller,
Figure BDA0001907343690000087
is the upper limit of the packet error probability for any logical node,
Figure BDA0001907343690000088
is the upper limit on the probability of any controller packet error. Thus, the connection between any one controller and the base station has higher reliability than the connection between any one logical node and the base station.
In some examples, managing adaptive allocation by the orchestration system further entails satisfying that the first total transmit energy is not greater than the first energy threshold
Figure BDA0001907343690000089
And the second total emission energy is not greater than the second energy threshold
Figure BDA00019073436900000810
The number of the used channels is not more than the total number n of the used channelsΣ. That is to say that the first and second electrodes,
Figure BDA00019073436900000811
Figure BDA00019073436900000812
Figure BDA00019073436900000813
and
Figure BDA00019073436900000814
first energy threshold of each logical node
Figure BDA00019073436900000815
And a second energy threshold of the controller
Figure BDA00019073436900000816
In some examples, the management orchestration system may send adaptive allocation instructions to adaptively allocate transmit power and the number of channel usages. The base station may receive the adaptive allocation command to allocate the transmit power and the number of channel uses.
In some examples, the base station may enable control of the power of each logical node and the power of the controller through automatic power control. For example, a radio frequency signal received by a transceiver station of a base station is sequentially input to a filter and a frequency converter having a filtering function, so as to obtain an intermediate frequency signal, and the intermediate frequency signal is input to an automatic power control module of the base station to control power. The automatic power control module comprises an A/D converter, a DC removal unit, a power estimation unit and a power feedback adjustment unit.
In some examples, the automatic power control process of the automatic power control module includes: the intermediate frequency signal is processed by an A/D converter to obtain a digital signal, the digital signal is processed by a direct current removing unit with variable point number to obtain a digital intermediate frequency signal with zero mean value, the digital intermediate frequency signal is processed by a power estimation unit with variable point number to obtain power estimation of the signal, the power estimation value is processed by a power feedback adjustment unit to obtain a new gain coefficient value, the new gain coefficient is applied to an amplitude limiting adjustment process in the next time period, and finally the output of the digital intermediate frequency signal is maintained near stable power.
In some examples, the base station can stably retransmit the received signal, so that the loss of the communication signal in wireless transmission can be effectively reduced or avoided, and the communication quality of the user can be ensured.
In some examples, the base station may implement allocation of the number of channel usages using frequency division multiplexing. In case the available bandwidth of a physical channel exceeds the bandwidth required by a single information signal (the information signal transmitted by each physical node), the total bandwidth of the physical channel may be divided into several sub-channels of the same bandwidth as the single information signal is transmitted. The information signal sent by a corresponding physical node is transmitted on each sub-channel, so that a plurality of information signals (multipath signals) are transmitted simultaneously in the same channel. Before frequency division multiplexing of multiple signals, the frequency spectrum of each signal needs to be shifted to different segments of the physical channel frequency spectrum by a frequency spectrum shifting technology, so that the bandwidths of the information signals are not overlapped with each other. After the spectrum shifting, each signal needs to be modulated with a different carrier frequency. Each signal is transmitted over a sub-channel of a certain bandwidth centered on its respective carrier frequency. In addition, to prevent mutual interference, anti-interference protection measures are needed to isolate each sub-channel.
In step S300, the management orchestration system may adaptively allocate transmit power and the number of channel uses based on a hybrid optimization algorithm to maximize the total throughput if the above conditions are satisfied. The mixed optimization algorithm is a mixed frog leap-extremum optimization algorithm which is multidimensional and takes the total throughput as an adaptive value. Specifically, the mixed frog-Extreme optimization algorithm replaces the random solution of the mixed frog-Extreme Optimization (EO) with improved Extreme optimization. The method combines extreme value optimization and mixed frog-leaping algorithm, and has strong local searching capability.
In some examples, the hybrid leapfrog-extremum optimization algorithm input parameters may include a first number M, a second number N, a number of payload bits k for each logical nodeLiNumber of payload bits k for each controllerCjReceiving signal-to-noise ratio gamma of each logic nodeLiReceiving signal-to-noise ratio gamma of each controllerCjUpper limit of packet error probability of logical node
Figure BDA0001907343690000101
Upper limit of packet error probability for controller
Figure BDA0001907343690000102
First energy threshold
Figure BDA0001907343690000103
Second energy threshold
Figure BDA0001907343690000104
And total number of channel uses n∑. Thus, the mixed frog-leap-extremum optimization algorithm can be optimized based on the input parameters and obtain optimized output parameters.
Fig. 3 is a flow chart illustrating a hybrid optimization algorithm of a resource optimization method of network function virtualization according to an example of the present disclosure.
In some examples, as shown in fig. 3, the mixed frog-extreme optimization algorithm may include setting initialization parameters (step S310). The initialization parameters may be, for example, input parameters of the above-described leapfrog-extremum optimization algorithm. The initialization parameters may also be an initial population F, a group p, the number of frogs in each group q, and an EO process jump condition NEOAnd the like.
In some examples, as shown in fig. 3, the mixed frog-extreme optimization algorithm may further include randomly generating a population (represented by an F frog) (step S320). That is, the mixed frog-extreme value optimization algorithm can randomly generate a group p comprising F frogs, and the position of each frog is X with the dimension t being 2(M + N)iAnd (4) showing. In step S320, F may be the initial population and q may be the number of frogs per group. The initial population F ═ pq can be generated by random frogs P ═ X1,X2,...XFAnd (4) generating. In some examples, p-20, q-10, and F-200 may be provided. In addition, Xi＝[xi1,xi2,...,xit]Is the position of the ith frog to solve the t-dimension problem. In the present disclosure, the dimension t satisfies t ═ 2(M + N). The multidimensional t may include a transmit power of each logical node, a transmit power of each controller, a number of channel usages of each logical node, and a number of channel usages of each controller.
In some examples, the mixed frog-extreme optimization algorithm as shown in fig. 3 may further include evaluating the fitness of each frog (step S330). That is, mixed frog leaping-extreme value is excellentThe adaptive algorithm can calculate the adaptive value f (X) of each frog by an evaluation algorithmi). Wherein the adaptation value f (X)i) Satisfy the requirement of
Figure BDA0001907343690000105
I.e. total throughput RΣTo an adaptation value. The fitness value is also referred to as fitness.
In some examples, the input parameter to the evaluation algorithm is the location X of the ith frogiThe output parameter is the adaptive value f (X) of the position of the ith frogi). The evaluation algorithm comprises in particular calculating the fitness f (X) of the position of each frogi) (ii) a Wherein
Figure BDA0001907343690000106
When in use
Figure BDA0001907343690000107
Or
Figure BDA0001907343690000111
When f (X)i)＝f(Xi) -T; when in use
Figure BDA0001907343690000112
Or
Figure BDA0001907343690000113
When f (X)i)＝f(Xi) -T; when in use
Figure BDA0001907343690000114
When f (X)i)＝f(Xi) -T; otherwise, f (X) is returnedi). Wherein T represents a penalty coefficient and satisfies T2 × 104. Thus, an adaptation value for each frog is obtained based on the evaluation algorithm.
In some examples, as shown in fig. 3, the mixed frog-extreme optimization algorithm may further include determining whether a convergence criterion is satisfied (step S340) and obtaining an optimal output parameter and ending the process when the convergence criterion is satisfied (step S350). Wherein the output parameter may comprise the total throughput RΣTransmission of each logical nodePower Pt,LiThe transmitting power P of the controllert,CjNumber of channel usages n of each logical nodeLiAnd the number of channel usages n of each controllerCj。
In some examples, as shown in fig. 3, the mixed frog-extreme optimization algorithm may further include sorting the F frogs in descending order when the convergence criterion is not satisfied (step S360). I.e. the corresponding fitness values of the F frogs are sorted in descending order.
In some examples, as shown in fig. 3, the mixed frog-extreme optimization algorithm may further include constructing a swarm and a sub-factor complex (step S370). Thus, the sub-factor complex can prevent the mixed frog-leap algorithm (SFLA) from stopping at the local optimum position.
In some examples, as shown in fig. 3, the shuffled frog-extremum optimization algorithm may further include a for-loop for each group of frogs, each loop searching locally in the sub-factor complex and extremally optimizing each frog (a temporary EO process) (step S380). In each loop of step S380, the fitness of each frog is calculated using an evaluation algorithm.
In some examples, as shown in fig. 3, the mixed frog-extremum optimization algorithm may further include shuffling all frogs (step S390). Thereby, rapid and stable convergence can be ensured.
In step S380, a triangular probability selection rule is generally followed in the sub-factor complex. Updating the moving distance of the worst frog in each group for the first iteration to be Di,w(l)＝r(Xi,b-Xi,w(l) Where r is a random number and r ∈ [0,1 ]]And Xi,bIs the best frog position in the group. In some examples, near the optimal location, there may be other locations that hold more food than the current optimal location, but are outside of the range between the current worst location and the optimal location. To cover this and extend the range of possible searches, D may be usedi,w(l)＝r(Xi,b-Xi,w(l) Modified as Di,w(l)＝wcr(Xi,b-Xi,w(l) ). Wherein c is a jump vision factor and c is more than or equal to 1. In addition, c cannot be extended indefinitely. Thus, c may beC is set to be more than or equal to 1 and less than or equal to 3. For example, c is 1.5. By addition of c to formula Di,w(l)＝wcr(Xi,b-Xi,w(l) Can increase the jump range of the frog at each step, expand the jump horizon of the frog, and enhance the optimization ability of the algorithm. lmaxIndicating the maximum number of iterations allowed in the leapfrog algorithm, e.g./max＝1000。
In some examples, the parameter w has a significant effect on the convergence behavior of the frog-leap blending algorithm, which serves to better control the relationship between the local search and the global search during frog leap. The parameter w represents strong global search capability and weak local search capability when it is large, and represents weak global search capability and strong local search capability when it is small. The parameter w satisfies
Figure BDA0001907343690000121
In some examples, set wmin0.8 and wmax2.5, w is linearly shifted from w as the iteration progressesmaxGradually decreases to wmin。
In some examples, Xi,w(l) Is the frog position with the worst fitness value in the i-th iteration of the i-th group, which is updated to Xi,w(l+1)＝Xi,w(l)+Di,w(l) If the updated adapted value of the new frog position is better than the original adapted value, the new position will replace the old one. Otherwise, Di,w(l)＝wcr(Xi,b-Xi,w(l) In X) ofi,bQuilt XbReplacement wherein XbIs the best frog position in the overall population. If no improvement is observed after the update, then a random solution X will be implementedi,w(l) In that respect This operation is repeated in each group until a specified iteration is reached.
Fig. 4 is a flow chart illustrating an extremum optimization process of a resource optimization method of network function virtualization according to an example of the present disclosure.
In some examples, as shown in fig. 4, the extremum optimization process is performed again for each frog in step S380. Extremum optimization can eliminate the worst component (i.e., constituent part) in the optimal individual.
In some examples, as shown in fig. 4, the extremum optimization process of step S380 may include randomly generating individual X ═ X1,x2,...,xt](step S381). That is, the extremum optimization process may randomly generate individuals having multiple components. In step S381, an individual may be represented as X ═ X1,x2,...,xt]，xiFor each component of the individual X. For example, each component of an individual X may be, for example, the transmit power P of the respective logical nodet,LiTransmitting power P of each controllert,CjNumber of channel usages n of each logical nodeLiAnd the number of channel usages n of each controllerCj。
In some examples, as shown in FIG. 4, the extremum optimization process can include evaluating an adaptive value f (X)i) (step S382). That is, the extremum optimization process may calculate an adaptive value for each component of each individual. The fitness value may also be referred to as a fitness value or fitness. Each component adaptation value satisfies
Figure BDA0001907343690000131
Wherein Δ xijIs the offset between the current and new positions of the ith frog, and this adaptation value can be obtained by mutating the ith component and keeping all other components fixed.
In some examples, as shown in fig. 4, the extremum optimization process may include setting an optimal solution Xb＝Xi(step S383). That is, the extremum optimization process can set the current individual as the optimal individual. The extremum optimizing process may further include determining whether a predetermined criterion is satisfied (step S384). The preset criterion may be simply referred to as a setting criterion. When the set criterion is satisfied, the adaptive value f (X) is outputi) And ends the process.
In some examples, as shown in fig. 4, the extremum optimization process can include evaluating each decision variable x when the set criteria are not satisfiediThe fitness of (1) (step S385). Wherein i is a natural number and a decision variableAlso referred to as components. Thus, the extremum optimization process can calculate the adaptive value for each component of the current individual. The adaptation value of each component satisfies
Figure BDA0001907343690000132
Wherein, Δ xijIs the offset between the current position and the new position of the ith frog. Thus, the fitness value of each mutated component can be obtained.
In some examples, as shown in fig. 4, the extremum optimization process may include finding that λ is satisfiedj≤λiX ofj(step S386). That is, the components having the adaptation values less than or equal to the adaptation value of the current individual are searched for in step S386. In addition, xjMay also be represented by x'i。
In some examples, as shown in fig. 4, the extremum optimization process may include obtaining X' (step S387). Wherein X' represents a target individual. In step S387, the component x of the target individualjCan be through x'i＝xi+ηiChanging its state. Wherein, x'iAnd xiIndicates the i-th component of the individual X before and after mutation.iIndicating the generation of random numbers. For example,iwhich may be a standard cauchy random variable or a standard gaussian random variable.iIt may also be a mixture of gaussian and cauchy operators of mutation. The variable η is an amplification factor and generally decreases linearly with increasing number of mutations. For example, the value of setting η decreases linearly from 1 to 0.1.
In some examples, as shown in fig. 4, the extremum optimization process may include unconditional acceptance X ═ X' (step S388). That is, in step S388, the target individual X' is regarded as the new current individual.
In some examples, as shown in fig. 4, the extremum optimization process may include if f (X) < f (X)b) Then XbX (step S389). That is, in step S389, when the adaptive value f (X) of the target individual is smaller than the adaptive value f (X) of the optimal individualb) And the target individual is taken as the optimal individual. This can improve the local search capability.
In step S390, after performing a deep search in each group, the entire frog population is shuffled and classified. And recording the optimal frog position according to the adaptive value. The population is then repartitioned and a local depth search is performed again.
In some examples, the runtime satisfaction complexity of the mixed frog-extreme optimization algorithm is
Figure BDA0001907343690000141
Wherein, the extremum optimizes the process jump condition NEOThe value may be set to any value between 1 and 100, for example, N may be setEO＝10。NiteIs the number of iterations of the extremum optimization, and NiteMay be set to 2. lmaxRepresenting the maximum number of iterations of the shuffled frog-leaping algorithm. And t is the dimension. Thus, runtime complexity can be obtained.
Fig. 5 is a waveform diagram illustrating the total throughput at different numbers of payload bits for a resource optimization method of network function virtualization according to an example of the present disclosure. Fig. 6 is a waveform diagram illustrating the total throughput at different received signal-to-noise ratios for a resource optimization method of network function virtualization according to examples of the present disclosure.
In some examples, as shown in fig. 5, waveform a is such that the received signal-to-noise ratio satisfies γL1＝γL2＝γC1＝γC2Waveform at 20 (dB). Waveform B is such that the received signal-to-noise ratio satisfies γL110(dB) and γL2＝γC1＝γC2Waveform at 20 (dB). Waveform C is such that the received signal-to-noise ratio satisfies γL1＝γC110(dB) and γL2＝γC2Waveform at 20 (dB).
The conditions satisfied by the waveform diagram shown in FIG. 5 include that the number of payload bits of the logical node and the controller are the same, i.e., kLi＝kCj. In addition, the satisfied condition of the waveform diagram shown in fig. 5 further includes a first energy threshold
Figure BDA0001907343690000142
Satisfy the requirement of
Figure BDA0001907343690000143
Second energy threshold
Figure BDA0001907343690000144
Satisfy the requirement of
Figure BDA0001907343690000145
The total number of channel uses satisfies n∑300. From FIG. 5, the total throughput R∑With kLi＝kCjThe value increases. With kLi＝kCjThe value is further increased and the total throughput R is increased∑Infinitely close to its maximum realizable value.
Waveform C at kLi＝kCjUnder the condition of 24(bytes), the evaluation algorithm has no feasible solution. This is because kLi＝kCjThe larger the value, the more physical layer resources are required. Specifically, at the beginning of FIG. 5, kLi＝kCjThe increase in value is in the total throughput R∑Providing a more dominant contribution. The increase in the number of channel uses then becomes at the total throughput R∑The dominant factor above. If it is not
Figure BDA0001907343690000151
The evaluation algorithm has no feasible solution. Furthermore, as can be seen in fig. 5, the results under waveform a are best because the channel signal-to-noise ratio of the logical node and the controller is best.
In some examples, as shown in fig. 6, waveform D is
Figure BDA0001907343690000152
And n∑300 time waveform diagram. Waveform E is
Figure BDA0001907343690000153
And n∑300 time waveform diagram. Waveform F is
Figure BDA0001907343690000154
And n∑300 time waveform diagram. Waveform G is
Figure BDA0001907343690000155
And n∑300 time waveform diagram. The waveform M is
Figure BDA0001907343690000156
Figure BDA0001907343690000157
And n∑400-hour waveform diagram. Waveform N is
Figure BDA0001907343690000158
And n∑The waveform diagram at 200.
The conditions satisfied by the waveform diagram shown in fig. 6 include that the respective received signal-to-noise ratios are the same. E.g. gammaL1＝γL2. The total throughput R of the waveform shown in FIG. 6∑Value with gammaL1＝γL2The value increases. Of these, the result for waveform E is the best and the result for waveform D is the worst.
In the disclosure, based on a network function virtualization architecture, a logic node and a controller send information signals to a base station, the base station receives the information signals and calculates related parameters, and when a third quantity, a payload bit number, a received signal-to-noise ratio, a first total transmission energy and a second total transmission energy satisfy respective requirements and the logic node and the controller satisfy a super-reliability requirement, a management and arrangement system adaptively allocates transmission power and a channel usage quantity based on a hybrid optimization algorithm to maximize total throughput. Therefore, the problem of optimizing the maximum total throughput in NFV deployment can be quickly and effectively solved, and the method has better global search capability.
The maximum total throughput optimization method based on the mixed frog-extreme optimization algorithm (MSFLA-EO) has excellent stability and global search capability, the resource optimization method is simulated according to various performance parameters, and experimental results prove that the resource optimization method has a remarkable effect on maximum total throughput.
While the present disclosure has been described in detail in connection with the drawings and examples, it should be understood that the above description is not intended to limit the disclosure in any way. Those skilled in the art can make modifications and variations to the present disclosure as needed without departing from the true spirit and scope of the disclosure, which fall within the scope of the disclosure.

Claims (5)

1. A resource optimization method for maximizing the total throughput of network function virtualization is a resource optimization method for maximizing the total throughput of a network function virtualization architecture and a base station,
the method comprises the following steps:
the network function virtualization architecture comprises a logic layer, a virtual layer, a physical layer and a management and orchestration system, wherein the logic layer comprises a first number of logic nodes and a second number of controllers, the physical layer comprises a third number of physical nodes, the third number is equal to the sum of the first number and the second number, the logic nodes and the controllers form a mapping relation with the physical nodes through the virtual layer, and the logic nodes and the controllers send information signals to the base station through the corresponding physical nodes through orthogonal channels;
the base station calculates the effective load bit number, the channel using number, the transmitting power, the receiving signal-to-noise ratio and the throughput of each logic node and each controller based on the information signal, calculates the first total transmitting energy and the first total throughput of all the logic nodes, and calculates the second total transmitting energy and the second total throughput of all the controllers; and is
When the third number is equal to a preset service number, the number of payload bits is equal to a preset number of bits, and the received signal-to-noise ratio is equal to a preset received signal-to-noise ratio, the logical node and the controller meet a super-reliability requirement, the first total transmission energy is not greater than a first energy threshold, the second total transmission energy is not greater than a second energy threshold, and the number of channel usages is not greater than a total number of channel usages, the management orchestration system adaptively allocates the transmission power and the number of channel usages based on a hybrid optimization algorithm to maximize a total throughput, the total throughput being a sum of the first total throughput and the second total throughput,
wherein the hybrid optimization algorithm is a hybrid leapfrog-extremum optimization algorithm that is multidimensional and takes the total throughput as an adaptation value, the multidimensional including a transmit power of each logical node and a transmit power of each controller, and a channel usage number of each logical node and each controller,
wherein the super-reliability requirement is satisfied
Figure FDA0002608790410000021
And
Figure FDA0002608790410000022
and is
Figure FDA0002608790410000023
Wherein p ise,LiPacket error probability, p, for the ith logical nodee,CjThe packet error probability for the jth controller,
Figure FDA0002608790410000024
is the upper limit of the packet error probability of any one of the logical nodes,
Figure FDA0002608790410000025
is the upper limit of the packet error probability of any one of the controllers, and the input parameters of the hybrid leapfrog-extremum optimization algorithm include a first number M, a second number N, and the number k of payload bits of each of the logic nodesLiNumber of payload bits k for each of the controllersCjReceiving signal-to-noise ratio gamma of each logic nodeLiReceiving signal-to-noise ratio gamma of each controllerCjUpper limit of packet error probability for any one of the logical nodes
Figure FDA0002608790410000026
Upper limit of packet error probability for any one of said controllers
Figure FDA0002608790410000027
First energy threshold
Figure FDA0002608790410000028
Second energy threshold
Figure FDA0002608790410000029
And total number of channel uses n∑The output parameter includes the total throughput RΣTransmitting power P of each logic nodet,LiThe transmission power P of each controllert,CjNumber of channel usages n of each logical nodeLiAnd the number of channel usages n of each of the controllersCj，
The mixed frog leap-extremum optimization algorithm comprises the following steps: setting initialization parameters; randomly generating a group p comprising F frogs; x with dimension t ═ 2(M + N) for each frog positioniRepresents; calculating the adaptive value f (X) of each frog by an evaluation algorithmi) (ii) a Judging whether a convergence criterion is met; when the convergence criterion is met, obtaining an optimal output parameter and ending the process; when the convergence criterion is not met, sorting the corresponding adaptive values of the F frogs according to a descending order; constructing p groups of frog and sub-factor complexes; performing for-circulation on each group of frogs, locally searching in the sub-factor complex in each circulation, calculating the adaptive value of each frog by using the evaluation algorithm, and performing extreme value optimization on each frog, wherein the adaptive value of each frog is obtained by the evaluation algorithm; shuffling all frogs, where F ═ pq, q denotes the number of frogs per group,
the input parameter of the evaluation algorithm is the position X of the ith frogiThe output parameter is the adaptive value f (X) of the position of the ith frogi) Calculating an adaptive value f (X) of the position of each frogi) (ii) a Wherein
Figure FDA0002608790410000031
T represents a penalty coefficient, and T is 2 × 104(ii) a When in use
Figure FDA0002608790410000032
Or
Figure FDA0002608790410000033
When f (X)i)＝f(Xi) -T; when in use
Figure FDA0002608790410000034
Or
Figure FDA0002608790410000035
When f (X)i)＝f(Xi) -T; when in use
Figure FDA0002608790410000036
When f (X)i)＝f(Xi) -T; otherwise, f (X) is returnedi)，RLiRepresents the throughput, R, of the ith logical nodeCjRepresents the throughput of the jth controller, ELiRepresenting the transmission energy of the ith logical node, ECjRepresenting the transmitted energy of the jth controller,
the process of extremum optimization comprises: randomly generating individuals having a plurality of components; calculating the fitness value for each component of each individual; setting the current individual as an optimal individual; judging whether the set standard is met, and ending the process when the set standard is met; when the set standard is not satisfied, calculating an adaptive value of each component of the current individual; searching for components with adaptation values less than or equal to the adaptation value of the current individual; obtaining a target individual; taking the target individual as a new current individual; and when the adaptive value of the target individual is smaller than that of the optimal individual, the target individual is taken as the optimal individual.
2. The optimization method of claim 1,
the total swallowingThe discharge amount is satisfied
Figure FDA0002608790410000037
The throughput of the ith logical node and the jth controller are respectively
Figure FDA0002608790410000038
And
Figure FDA0002608790410000039
3. the optimization method of claim 1,
the transmission energy of the ith logic node and the jth controller is respectively ELi＝Pt,LinLiAnd ECj＝Pt,CjnCjWherein E isLiAnd ECjAll have the unit of W.Hz.s, and the total number of the used channels satisfies
Figure FDA00026087904100000310
4. The optimization method of claim 1,
an adaptation value λ of each of the componentsijSatisfy the requirement of
Figure FDA0002608790410000041
Wherein, Δ xijIs the offset between the current position and the new position of the ith frog.
5. The optimization method of claim 1,
the running time of the mixed frog leap-extremum optimization algorithm meets the complexity of
Figure FDA0002608790410000042
Wherein, the extremum optimizes the process jump condition NEOThe value is set to any value between 1 and 100，NiteIs the number of iterations of the extremum optimization, and NiteIs set to be 2, lmaxRepresents the maximum iteration number of the mixed frog leaping algorithm, and t is a dimension.
CN201811537839.3A 2018-12-15 2018-12-15 Resource optimization method for maximizing total throughput of network function virtualization Active CN109362093B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811537839.3A CN109362093B (en) 2018-12-15 2018-12-15 Resource optimization method for maximizing total throughput of network function virtualization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811537839.3A CN109362093B (en) 2018-12-15 2018-12-15 Resource optimization method for maximizing total throughput of network function virtualization

Publications (2)

Publication Number Publication Date
CN109362093A CN109362093A (en) 2019-02-19
CN109362093B true CN109362093B (en) 2020-11-20

Family

ID=65328947

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811537839.3A Active CN109362093B (en) 2018-12-15 2018-12-15 Resource optimization method for maximizing total throughput of network function virtualization

Country Status (1)

Country Link
CN (1) CN109362093B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109743731B (en) * 2019-03-05 2020-01-10 深圳大学 Performance evaluation method and system of URLLC system based on physical layer authentication
CN109699039B (en) * 2019-03-05 2020-01-17 深圳大学 Heuristic algorithm-based equal frame length perception network resource optimization method
CN109788569B (en) * 2019-03-05 2020-01-07 深圳大学 Perceptual network resource optimization method based on heuristic algorithm
CN111818453B (en) * 2020-07-13 2021-10-15 深圳大学 Method and system for sharing frequency spectrum of millimeter wave mobile base station based on clustering algorithm

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104618912A (en) * 2015-01-29 2015-05-13 南京理工大学 Spectrum sensing-based heterogeneous cognitive wireless network resource distributing method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102497643B (en) * 2011-12-13 2014-08-06 东南大学 Cognitive ratio power control method
CN104573369A (en) * 2015-01-14 2015-04-29 天津大学 Shuffled frog-leaping based division method of software and hardware
US10200463B2 (en) * 2017-05-22 2019-02-05 At&T Intellectual Property I, L.P. Systems and methods to improve the performance of a network by more efficient virtual network resource allocation
CN108040018B (en) * 2017-11-13 2020-02-14 华中科技大学 Fine-grained network flow scheduling method and system under network function virtualization

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104618912A (en) * 2015-01-29 2015-05-13 南京理工大学 Spectrum sensing-based heterogeneous cognitive wireless network resource distributing method

Also Published As

Publication number Publication date
CN109362093A (en) 2019-02-19

Similar Documents

Publication Publication Date Title
CN109362093B (en) Resource optimization method for maximizing total throughput of network function virtualization
JP5351241B2 (en) Apparatus and method for allocating resources to nodes in a communication system using repetitive resource weight update
US9723572B2 (en) Systems and methods for uplink power control and scheduling in a wireless network
CN104509019B (en) Method and apparatus for provided for radio resources management
Mismar et al. A framework for automated cellular network tuning with reinforcement learning
CN110233653A (en) Blind multipath recognition methods and system based on the mimo system for weighting integrated clustering algorithm
CN109996248A (en) Electronic equipment and method and computer readable storage medium for wireless communication
EP3669566B1 (en) Community detection in radio access networks with constraints
CN111181792A (en) SDN controller deployment method and device based on network topology and electronic equipment
US20130337818A1 (en) Apparatus and method for determining a control unit using feasibility requests and feasibility responses
CN109788485A (en) The method for being used to help solve the cluster optimization of the border issue in communication system
Perlaza et al. On the base station selection and base station sharing in self-configuring networks
CN110809893A (en) Electronic device and method for wireless communication
Min et al. Meta-scheduling framework with cooperative learning towards beyond 5G
Liu et al. DAFEE: A decomposed approach for energy efficient networking in multi-radio multi-channel wireless networks
CN109495402B (en) Resource optimization method for minimizing physical layer resources of network function virtualization
CN109526006B (en) Resource optimization method for maximizing URLLC service quantity of network function virtualization
CN110224861A (en) The implementation method of adaptive dynamic heterogeneous network selection policies based on study
CN112911674A (en) Data transmission method, device, equipment and storage medium
Vijayaraghavan et al. Delay-aware Wireless Resource Allocation and User Association in LiFi-WiFi Heterogeneous Networks
KR101537891B1 (en) Energy-efficient transmission power management scheme for multicast service over wireless access network
KR101671635B1 (en) Method and system for selecting mobile relay based on dynamic programming in millimeter-wave broadband 5g network
Chen et al. Foresighted resource scheduling in software-defined radio access networks
CN113630906B (en) Method and device for compensating interruption of wireless self-organizing network
US11502793B2 (en) Cooperative radio resource allocation in a wireless communication network and methods for use therewith

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