CN102300269A - Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method - Google Patents
Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method Download PDFInfo
- Publication number
- CN102300269A CN102300269A CN2011102409230A CN201110240923A CN102300269A CN 102300269 A CN102300269 A CN 102300269A CN 2011102409230 A CN2011102409230 A CN 2011102409230A CN 201110240923 A CN201110240923 A CN 201110240923A CN 102300269 A CN102300269 A CN 102300269A
- Authority
- CN
- China
- Prior art keywords
- chromosome
- chromosomal
- qos
- value
- genetic algorithm
- 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
Images
Abstract
The invention relates to a genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method, which adopts the wireless recognition network technology, and mainly comprises quality of service (QoS) index, intelligent decision and network reconfiguration in perception service running process. The QoS index in the perception service running process is used for evaluating the current service running state and for providing decision foundation to the intelligent decision process; the intelligent decision process adopts adjustable parameters in a network protocol stack as input, the genetic algorithm capable of solving multipurpose optimization problem is adopted, and an adjustable parameter value which can optimize the multiple multi QoS indexes is searched in a value space of each adjustable parameter; and the action process adopts the result of the intelligent decision as input, and the network is reconfigured according to the solved adjustable parameter value so as to guarantee the service end-to-end QoS and optimize the network end-to-end performance.
Description
Technical field
The present invention relates to a kind of mobile ad-hoc network, belong to the technical field that end-to-end quality of service is guaranteed in the mobile ad-hoc network with cognitive function.
Background technology
In recent years, the end-to-end quality of service in the mobile ad-hoc network (QoS) guarantees that problem has caused the extensive concern of domestic and international experts and scholars and scientific research institution.End-to-end QoS refers to all nodes that relate in the data transmission procedure, total effect of link service performance, it has determined the satisfaction of a user to serving, usually describe by some basic performance index, comprise link availability, throughput, time delay, shake, packet loss rate, connection setup time etc.But mobile ad-hoc network is widely used in each field of society with its stronger flexibility survivability and self organization ability.Yet, also have some problems in this network, such as resource-constrained, Radio Link off and on, intermittently available, central controlled infrastructure shortage etc., these make the end-to-end QoS in the mobile ad-hoc network guarantee that problem is faced with a lot of challenges.
Mainly there is following problem in the existing research of guaranteeing at end-to-end QoS in the mobile ad-hoc network at present: the one, and, the end-to-end QoS method of assuring does not have versatility.Because it is different that the mechanism of different application causes influencing the key factor of its end-to-end QoS, thereby, adopt different end-to-end QoS method of assuring according to the characteristics of different application.Such as, the coded system of video, audio frequency is one of key factor that influences multimedia service QoS, thereby much the research of its QoS security problem is all concentrated on coding method, other a lot of end-to-end QoSs of using are ensured the consideration that does not then have this respect factor.The 2nd,, exist some associations between the QoS index, to the guarantee of some of them index, may have influence on the performance of other indexs.Present most research can only be guaranteed single QoS index, guarantees a normally np problem of a plurality of QoS indexs simultaneously, need solve by means of some intelligent algorithms.The 3rd,, the ability of present most of method aspect the response environment variation is more weak, this environment of the mobile ad-hoc network network that can change at any time especially, the adaptive ability of Enhancement Method, make it when environmental change, it is very necessary can making response fast.
Meanwhile, wireless cognition network is emerging in recent years a kind of network technology with self-management, self-optimization, self-healing ability.It can be by the parameter value of approach perception such as measurement, estimation, induction reflection current network state; Then, take certain intelligent method,,, make a strategic decision out and can improve the allocation plan of network end-to-end performance according to certain end-to-end target as genetic algorithm, game theory etc.; At last, according to the result who makes a strategic decision out, carry out the reconstruct of network; In said process, network also has the ability of study, accumulates certain reconstruct case, therefrom extracts knowledge, to instruct follow-up decision process.
At present several problems in the mobile ad-hoc network end-to-end QoS is guaranteed to study, the method for introducing wireless cognition network is very necessary.A key issue in the wireless cognition network is the Using Intelligent Decision-making Method that decision process adopts.Genetic algorithm is to have used for reference some phenomenons in the evolution biology, as variation, intersection etc., grows up, and be a kind of searching algorithm that is used to solve optimization problem.Its feature has: the trail of separating from problem begins search, and wide coverage is beneficial to the overall situation according to qualifications; Handle a plurality of individualities simultaneously, reduced the risk that is absorbed in locally optimal solution, algorithm itself is easy to realize parallelization; Have self-organizing, adaptivity and self-study habit; Can solve multi-objective optimization question etc.These characteristics make it become the optimization searching method that is fit to guarantee professional a plurality of QoS indexs.
Summary of the invention
The present invention is to guarantee that a plurality of end-to-end QoS indexs professional in the mobile ad-hoc network are target, thought according to wireless cognition network, set up professional end-to-end QoS in the mobile ad-hoc network and guarantee the wireless cognition network model of problem, and with genetic algorithm as the Using Intelligent Decision-making Method in the wireless cognition network model, specifically comprise:
1. wireless cognition network model
The basic composition of wireless cognition network has these three parts of perception, decision-making and action.The present invention guarantees that with end-to-end QoS professional in the mobile ad-hoc network problem is established as the wireless cognition network model.With the QoS index of business, as the end-to-end time delay of business, shake, packet loss etc., as the perceptual parameters of reflection network state; Can solve the method for the genetic algorithm of multi-objective optimization question as the intelligent decision process; With the adjustable parameter in the network protocol stack, as the transmitting power of radio node, the transmission rate of data etc., as the action scheme configuration parameter of finishing network reconfiguration.
2. adopt the cognitive Decision process of genetic algorithm
Genetic algorithm is as the Using Intelligent Decision-making Method in the wireless cognition network model, and it mainly acts on is the adjustable parameter value of searching in the network protocol stack that can guarantee professional a plurality of end-to-end QoS indexs.To make up the chromosome as Attended Operation in the genetic algorithm through the adjustable parameter of coding, the number of adjustable parameter is chromosomal length; The QoS index is a plurality of QoS indexs as fitness function owing to what the present invention is directed to, and fitness function represents that with the form of vector each element in the vector is corresponding to each treated QoS index.The computing that chromosome participates in is operations such as the existing algorithm of tournament selection of genetic algorithm, intersection, variation, by these operations, chromosome can constantly be transformed into different chromosome, by non-domination sort algorithm these chromosomes are carried out the multiple target ordering again, select equal better chromosome on each sub-goal, as new population, continue to participate in genetic manipulation.After algorithm finished, the chromosome decoding with solving can obtain required adjustable parameter value.
Compared with prior art, innovation part of the present invention is: this method is a kind of more common end-to-end QoS method of assuring, is not subjected to concrete professional limitation; The technology of employing wireless cognition network has strengthened the flexibility and the adaptivity of this method; Genetic algorithm can solve multi-objective optimization question, makes this method can ensure a plurality of QoS indexs simultaneously.Be embodied in:
1) this method is in the process that searching is distributed rationally, and the standard of judge is exactly professional end-to-end QoS index, and these indexs generally are the general end-to-end time delay of miscellaneous service, packet loss, shake etc.; When improving the QoS situation, the parameter of adjustment is the general parameter in the network protocol stack, as the transmitting power of radio node, the transmission rate of data etc.; Complicated mechanism in the professional concrete running need not too much consideration.This design makes this method in the process of using, and can not be subjected to the limitation of concrete business characteristic, has versatility.
2) this method has adopted the technology of wireless cognition network.It on this technological essence the circulation of a band feedback, each sensing network environment, the scheme of reshuffling of making a strategic decision out and finish after the network reconfiguration, cognitive process does not finish, and node is perception primary network environment at set intervals, when environment changed, decision-making and action can be carried out once more.This process of going round and beginning again makes that this method can be made in time the variation of environment, response exactly, has strengthened the adaptivity and the flexibility of network.
3) the non-domination sort method in the genetic algorithm is to design at the multiple target ordering specially.When solving multi-objective optimization question, general method is to utilize means such as weight with the synthetic integration objective of a plurality of target group, and it is crucial often that combined method wherein and weight are selected, if design badly, will have a strong impact on decision-making, result calculated.Utilize non-domination sort method, do not need can directly carry out multiobject ordering, comparison to the synthetic integration objective of a plurality of target group, convenience is easy-to-use and accuracy is higher.
Description of drawings
Fig. 1 is based on the wireless cognition network end-to-end QoS method of assuring schematic diagram of genetic algorithm.
Fig. 2 perception flow chart.
Fig. 3 genetic algorithm flow chart.
Embodiment
Based on the basic principle of the wireless cognition network end-to-end QoS method of assuring of genetic algorithm as shown in Figure 1, mainly comprise perception QoS index, intelligent decision and three parts of action.
Perception QoS indexing section is the end-to-end QoS information of obtaining in the reflection service operation process, comprises information such as end-to-end time delay, shake, packet loss, and this process is to realize by the mode that sends QoS Information Statistics packet.The flow process of this process is referring to Fig. 2, and concrete steps are as follows:
1) global variable QoS_ID of definition, this variable is used to identify QoS Information Statistics packet, and it is encapsulated in the QoS Information Statistics packet, makes that this packet can be identified;
2) set detecting period Sense_time, QoS Information Statistics time QoS_period and timer t three variablees, wherein Sense_time＞QoS_period at the destination of business;
3) when t=0, begin to add up the QoS indication informations such as time delay of the business data packet of sending from the source end;
4) when t=QoS_period, stop to add up QoS information, and the QoS information that will add up during this period of time is packaged into QoS Information Statistics packet, QoS_ID inserted the stem of packet, corresponding field name sends to this packet the node that need reshuffle adjustment " command ";
5) when t=Sense_time, t is clear 0, gets back to step 3).
Perception is carried out according to above-mentioned steps circulation, the influence that professional end-to-end QoS is caused with reasons such as monitoring of environmental variations.
The intelligent decision part mainly is the adjustable parameter value that can make a plurality of QoS index optimizations by the genetic algorithm removal search, and its flow process is referring to Fig. 3, and concrete steps are as follows:
1) adjustable parameter refers to can adjust automatically in the network protocol stack or by the parameter of manual configuration, as the transmitting power of radio node, the transmission rate of data etc.Span according to each adjustable parameter, be encoded into (0,1] an interval interior real number, the adjustable parameter value behind each coding is as a gene in the genetic algorithm, all genomic constitutions participate in the chromosome of computing in genetic algorithm, the number of adjustable parameter is chromosomal length.
2) QoS index, as end-to-end time delay of business etc., after treatment as the fitness function in the genetic algorithm, at arbitrary chromosome i, its fitness function is defined as:
U
i(u
1i，u
2i，...，u
mi)
Wherein, U
iBe the fitness function of chromosome i correspondence, by u
1i, u
2i..., u
MiThis m sub-target formed arbitrary sub-goal u
JiBe that QoS index j (function of 1≤j≤m), obtained by statistics QoS information and the method that sends the QoS information packet by the perception link by the QoS index.
3) set maximum genetic algebra,, form the parent population for chromosome is composed n (n is an even number) class value at random as initial solution.
4) calculate every chromosomal fitness function value in the father population.
5) chromosome in the parent colony is adopted the algorithm of tournament selection method select the filial generation that makes new advances.Concrete grammar is: select n in the uncle population at random
1(n
1＜n) bar chromosome carries out the comparison of fitness value size, relatively adopts the method for non-domination ordering, as described in step 6), arrives the chromosomal inheritance that wherein the fitness function value is the highest of future generation; If chromosomal fitness function value is identical, then by chromosomal density compare operation, select a chromosome of density maximum, the computational methods of chromosome density are as described in the step 7); Repeat said process n/2 time, just can obtain comprising the chromosomal new population of n/2 bar.
6) non-domination ordering is a kind of at multiobject sort method, domination is defined as: chromosome p domination chromosome q, all targets of expression chromosome p correspondence are all unlike the goal discrepancy of chromosome q correspondence, and have at least the target than chromosome q correspondence good in the target of chromosome p correspondence, its detailed process is as follows:
6a) chromosome of every in population p safeguards two data structure n
pAnd S
p, n wherein
pThe chromosome quantity that can arrange p in the expression population, S
pThe chromosome congression of being arranged by p in the expression population;
6b) in dual circulation, every in population chromosome and other chromosomes are compared, if this chromosome is arranged other chromosome, then the chromosome that will be arranged is added into this chromosomal S set
pIf this chromosome is arranged by other chromosomes, this chromosomal n then
pAdd one; After the dual loop ends, all chromosomal n have just been calculated
pAnd S
p
6c) with n
pValue is that 0 chromosome is kept at set F
1In, the expression grade is 1 chromosome;
6d) at F
1In each element p, at its S
pIn find out all and satisfy n
qThe chromosome q of-1=0 preserves these chromosomes into set F
2In, the expression grade is 2 chromosome;
6e) to F
2In chromosome execution in step 6d), find out grade and be 3 chromosome;
6f) repetitive process 6e), all be assigned with non-dominance hierarchy up to all chromosome.
7) introducing of chromosome density notion is identical and can't discharge the situation of sequencing in order to solve some chromosomal non-dominance hierarchies, and the computational process of chromosome density is as follows:
7a) every chromosomal density is initialized as 0;
7b) at chromosomal j sub-goal, wherein (j=1,2 ..., m), m is the quantity of chromosome sub-goal, when j≤m, and circulation execution in step 7c) to 7f);
7c) chromosome being carried out ascending order according to j sub-goal arranges;
7d) in this minor sort, will be positioned at first two chromosome density assignment with last is a bigger value;
7e) to the 2nd to n-1 bar chromosome execution in step (7f);
7f) density value of chromosome i by two parts and form, a part is the current density value of chromosome i, another part is in this minor sort, is positioned at the absolute value of the difference of two chromosomal j sub-desired values before and after the chromosome i, wherein 2≤i≤n-1.
8) child chromosome that step 5) is generated is carried out the new child chromosome of interlace operation generation, and detailed process is as follows:
8a) select two different chromosomes at random in population, certain gene position of selective staining body at random again exchanges the gene information of corresponding positions on these two chromosomes;
8b) repeat step 8a n/2 time), can produce the new chromosome of n/2 bar;
N/2 bar chromosome that 8c) will newly produce and original n/2 bar chromosome merge, and obtain comprising the chromosomal progeny population of n bar.
9) child chromosome that generates in the step 8) is carried out mutation operation, obtain new progeny population once more, detailed process is as follows:
9a) set a variation Probability p
0
9b) at every chromosome, generate a Probability p at random, if p＞p0, certain on selective staining body position at random changes this gene information;
9c) every chromosome execution in step 9b), generate a new chromosomal progeny population of n bar that comprises.
10) the fitness function value of calculating child chromosome.
11) parent chromosome and child chromosome are merged.
12) adopt the method for non-domination ordering to select preceding n bar fitness function value better chromosome, as new parent chromosome.If some chromosomal fitness function value is in identical non-dominance hierarchy in non-domination ordering, make and only adopt non-domination ordering can't just in time select n bar chromosome, then adopt the method for density calculation from the identical chromosome of non-dominance hierarchy, to select the bigger chromosome of density.
13) when hereditary number of times does not reach the maximum algebraically of setting, repeated execution of steps 5) to 12).
14) when hereditary number of times reaches the maximum algebraically of setting, the n bar chromosome of obtaining is more excellent in each sub-goal, the last the highest chromosome of comprehensive evaluation value of in this n bar chromosome, selecting, with its decoding, just can obtain required adjustable parameter value, the formula that the chromosome value is carried out overall merit is:
Wherein, b
1, b
2..., b
nBe the chromosomal comprehensive evaluation value of n bar, a
1, a
2..., a
mM sub-objective weight of expression, these weights are distributed according to the importance of sub-goal, u
JiBe j sub-goal of chromosome i correspondence value (j=1,2 ..., m; I=1,2 ..., n).
What undercarriage was mainly finished is the work of network reconfiguration.In this part, corresponding adjustable parameter in the protocol stack of configuration network as a result that node goes out according to the intelligent decision process decision-making of last step.
After the adjustable parameter adjustment, just can guarantee professional end-to-end QoS, because the mobility of mobile ad-hoc network, the variation of network environment may cause the decline of professional end-to-end QoS once more, node regularly carries out perception, when finding that QoS can't guarantee, can start intelligent decision and action once more.The present invention adopts this wireless cognition network technology of going round and beginning again to guarantee professional end-to-end QoS, and strengthens the adaptibility to response to environmental change.
Claims (7)
1. the wireless cognition network end-to-end quality of service method of assuring based on genetic algorithm comprises the steps:
1) the partial parameters value in the network protocol stack is adjustable, as the transmitting power of radio node, the transmission rate of data etc., the present invention calls adjustable parameter to this class parameter, span according to each adjustable parameter, be encoded into (0,1] an interval interior real number, the adjustable parameter value behind each coding is as the gene in the genetic algorithm, all genomic constitutions participate in the chromosome of computing in genetic algorithm, the number of adjustable parameter is chromosomal length;
2) can reflect the parameter of network service quality (QoS) situation, as time delay of business etc., the present invention calls the QoS index to this class parameter, these QoS indexs after treatment, as the fitness function in the genetic algorithm, at chromosome i, its fitness function is defined as:
U
i(u
1i，u
2i，...，u
mi)
Wherein, U
iBe the fitness function of chromosome i correspondence, by u
1i, u
2i..., u
MiThis m sub-target formed arbitrary sub-goal u
JiBe QoS index j (function of 1≤j≤m), the QoS index by the node in the network by service operation in statistics a period of time QoS information and obtain to the method that corresponding adjustment node sends the QoS information packet;
3) set maximum genetic algebra,, form the parent population for chromosome is composed n (n is an even number) class value at random as initial solution;
4) calculate every chromosomal fitness function value in the father population, the parameter of fitness function is the QoS index;
5) chromosome in the parent colony is carried out these three kinds of traditional genetic manipulations of algorithm of tournament selection, intersection and variation, generate new child chromosome;
6) calculate chromosomal fitness function value in the new progeny population that produces;
7) child chromosome and the parent chromosome that produces is merged;
8) adopt the method for non-domination ordering to select preceding n bar fitness function value better chromosome, as new parent chromosome;
9) if some chromosomal fitness function value is in identical non-dominance hierarchy in non-domination ordering, make and only adopt non-domination ranking method can't just in time select n bar chromosome, then adopt the method for density calculation from the identical chromosome of non-dominance hierarchy, to select the more excellent chromosome of the bigger conduct of density;
10) when hereditary number of times does not reach the maximum genetic algebra of setting, repeated execution of steps 5) to 9);
11) when hereditary number of times reaches the maximum algebraically of setting, the n bar chromosome of obtaining is more excellent in each sub-goal, the last the highest chromosome value of comprehensive evaluation value of in this n bar chromosome, selecting, with its decoding, just can obtain required adjustable parameter value, the formula that the chromosome value is carried out overall merit is:
Wherein, b
1, b
2..., b
nBe the chromosomal comprehensive evaluation value of n bar, a
1, a
2..., a
mM sub-objective weight of expression, these weights are distributed according to the importance of sub-goal, u
JiBe chromosome i correspondence j sub-goal (j=1,2 ..., m; I=1,2 ..., n);
12) the adjustable parameter value that goes out according to genetic algorithm for solving is reconstructed network, and the end-to-end QoS of network can be protected.
2. according to the wireless cognition network end-to-end QoS method of assuring based on genetic algorithm described in the claim 1, the wherein operation of the algorithm of tournament selection shown in the step 5) is performed as follows:
2a) select n at random in the uncle population
1(n
1＜n) bar chromosome adopts non-domination sort method to carry out the comparison of fitness value size, with a chromosomal inheritance that wherein fitness is the highest to of future generation;
2b) if the non-dominance hierarchy of chromosomal fitness function value is identical, then by chromosomal density compare operation, a chromosomal inheritance selecting the density maximum is to of future generation;
2c) step 2a), 2b) repeat n/2 time, just can obtain comprising the chromosomal new population of n/2 bar.
3. according to the wireless cognition network end-to-end QoS method of assuring based on genetic algorithm described in the claim 1, the wherein interlace operation shown in the step 5) is performed as follows:
3a) select two different chromosomes at random in population, certain gene position of selective staining body at random again exchanges the gene information of corresponding positions on these two chromosomes;
3b) repeat said process n/2 time, obtain the new chromosome of n/2 bar;
N/2 bar chromosome that 3c) will newly produce and original n/2 bar chromosome merge, and obtain comprising the chromosomal progeny population of n bar.
4. according to the wireless cognition network end-to-end QoS method of assuring based on genetic algorithm described in the claim 1, the mutation operation shown in the step 5) wherein is performed as follows:
4a) set a variation Probability p
0
4b) at every chromosome, generate a Probability p at random, if p＞p
0, certain on selective staining body position at random changes this gene information, generates a new chromosome;
4c) every chromosome implementation 4b all), generate a new chromosomal progeny population of n bar that comprises.
5. according to the wireless cognition network end-to-end QoS method of assuring described in the claim 1 based on genetic algorithm, wherein the non-domination sorting operation shown in the step 8) is used to carry out the multiple target ordering, the implication of domination is: chromosome p domination chromosome q, all targets that are chromosome p correspondence are all unlike the goal discrepancy of chromosome q correspondence, and have at least the target than chromosome q correspondence good in the target of chromosome p correspondence; This process is performed as follows:
5a) chromosome of every in population p safeguards two data structure n
pAnd S
p, n wherein
pThe chromosome quantity that can arrange p in the expression population, S
pThe chromosome congression of being arranged by p in the expression population;
5b) in dual circulation, every in population chromosome and other chromosomes are compared, if this chromosome is arranged other chromosome, then the chromosome that will be arranged is added into this chromosomal S set
pIf this chromosome is arranged by other chromosomes, this chromosomal n then
pAdd one, after the dual loop ends, just calculated all chromosomal n
pAnd S
p
5c) with n
pValue is that 0 chromosome is kept at set F
1In, represent that non-dominance hierarchy is 1 chromosome;
5d) at F
1In each element p, at its S
pIn find out all and satisfy n
qThe chromosome q of-1=0 preserves these chromosomes into set F
2In, represent that non-dominance hierarchy is 2 chromosome;
5e) to F
2In chromosome execution in step 5d), find out non-dominance hierarchy and be 3 chromosome;
5f) repeat process 5e), all be assigned with non-dominance hierarchy up to all chromosome.
6. according to the wireless cognition network end-to-end QoS method of assuring based on genetic algorithm described in the claim 1, the wherein chromosomal density calculation operation shown in the step 9) is performed as follows:
6a) every chromosomal density is initialized as 0;
6b) at chromosomal j sub-goal, circulation execution in step 6c) to 6f), wherein (j=1,2 ..., m), m is the quantity of sub-goal in the fitness function of chromosome correspondence;
6c) chromosome being carried out ascending order according to j sub-goal arranges;
6d) in this minor sort, will be positioned at first two chromosome density assignment with last is a bigger numerical value;
6e) to the 2nd in this minor sort to n-1 bar chromosome execution in step 6f);
6f) density value of chromosome i by two parts and form, a part is the current density value of chromosome i, another part is in this minor sort, is positioned at the absolute value of the difference of two chromosomal j sub-desired values before and after the chromosome i, wherein 2≤i≤n-1.
7. computer program, it realizes the method for claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110240923 CN102300269B (en) | 2011-08-22 | 2011-08-22 | Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110240923 CN102300269B (en) | 2011-08-22 | 2011-08-22 | Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102300269A true CN102300269A (en) | 2011-12-28 |
| CN102300269B CN102300269B (en) | 2013-02-13 |
Family
ID=45360361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110240923 Expired - Fee Related CN102300269B (en) | 2011-08-22 | 2011-08-22 | Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102300269B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014139395A1 (en) * | 2013-03-12 | 2014-09-18 | International Business Machines Corporation | Early generation of individuals to accelerate genetic algorithms |
| CN104952001A (en) * | 2015-07-02 | 2015-09-30 | 华侨大学 | Method for performing power optimized scheduling on controllable loads comprising air conditioning loads |
| CN105281954A (en) * | 2015-10-21 | 2016-01-27 | 武汉大学 | Method for evaluating spatial information service quality and optimizing service chain |
| CN105634812A (en) * | 2015-12-31 | 2016-06-01 | 河南科技大学 | Network throughput optimization parameter acquiring method based on immunization algorithm |
| CN107743073A (en) * | 2017-10-17 | 2018-02-27 | 云南大学 | The service evolution selection strategy method and system that a kind of user's QoS demand is oriented to |
| CN111812977A (en) * | 2020-06-10 | 2020-10-23 | 北京宇航系统工程研究所 | GEO direct fixed-point launching orbit optimization method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060009209A1 (en) * | 2004-06-25 | 2006-01-12 | Rieser Christian J | Cognitive radio engine based on genetic algorithms in a network |
| CN101902747A (en) * | 2010-07-12 | 2010-12-01 | 西安电子科技大学 | Spectrum allocation method based on fuzzy logic genetic algorithm |
| CN102045775A (en) * | 2011-01-07 | 2011-05-04 | 哈尔滨工程大学 | Method for switching frequency spectrum of cognitive radio network |
-
2011
- 2011-08-22 CN CN 201110240923 patent/CN102300269B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060009209A1 (en) * | 2004-06-25 | 2006-01-12 | Rieser Christian J | Cognitive radio engine based on genetic algorithms in a network |
| CN101902747A (en) * | 2010-07-12 | 2010-12-01 | 西安电子科技大学 | Spectrum allocation method based on fuzzy logic genetic algorithm |
| CN102045775A (en) * | 2011-01-07 | 2011-05-04 | 哈尔滨工程大学 | Method for switching frequency spectrum of cognitive radio network |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014139395A1 (en) * | 2013-03-12 | 2014-09-18 | International Business Machines Corporation | Early generation of individuals to accelerate genetic algorithms |
| US9311597B2 (en) | 2013-03-12 | 2016-04-12 | International Business Machines Corporation | Early generation of individuals to accelerate genetic algorithms |
| US10755175B2 (en) | 2013-03-12 | 2020-08-25 | International Business Machines Corporation | Early generation of individuals to accelerate genetic algorithms |
| CN104952001A (en) * | 2015-07-02 | 2015-09-30 | 华侨大学 | Method for performing power optimized scheduling on controllable loads comprising air conditioning loads |
| CN105281954A (en) * | 2015-10-21 | 2016-01-27 | 武汉大学 | Method for evaluating spatial information service quality and optimizing service chain |
| CN105634812A (en) * | 2015-12-31 | 2016-06-01 | 河南科技大学 | Network throughput optimization parameter acquiring method based on immunization algorithm |
| CN105634812B (en) * | 2015-12-31 | 2019-05-10 | 河南科技大学 | A kind of network throughput Optimal Parameters acquisition methods based on immune algorithm |
| CN107743073A (en) * | 2017-10-17 | 2018-02-27 | 云南大学 | The service evolution selection strategy method and system that a kind of user's QoS demand is oriented to |
| CN111812977A (en) * | 2020-06-10 | 2020-10-23 | 北京宇航系统工程研究所 | GEO direct fixed-point launching orbit optimization method |
| CN111812977B (en) * | 2020-06-10 | 2022-07-29 | 北京宇航系统工程研究所 | GEO direct fixed-point launching orbit optimization method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102300269B (en) | 2013-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102300269B (en) | Genetic algorithm based antenna recognition network end-to-end service quality guaranteeing method | |
| CN102892188B (en) | Based on the ascending power control method of genetic algorithm and device in communication network | |
| CN110521234A (en) | The method and system of network self-organization is carried out using deep learning | |
| Kalathil et al. | Decentralized learning for multiplayer multiarmed bandits | |
| Hamdi et al. | Resource allocation based on hybrid genetic algorithm and particle swarm optimization for D2D multicast communications | |
| CN105430707A (en) | WSN (Wireless Sensor Networks) multi-objective optimization routing method based on genetic algorithm | |
| Rahmanian et al. | A novel genetic algorithm in LEACH-C routing protocol for sensor networks | |
| Deng et al. | A digital twin approach for self-optimization of mobile networks | |
| Romoozi et al. | A positioning method in wireless sensor networks using genetic algorithms | |
| Oliveira et al. | Handling network uncertainty in heterogeneous wireless networks | |
| CN105246097A (en) | Lifetime optimization method of mobile Sink-based wireless sensor network | |
| Xu et al. | A simulated annealing based genetic local search algorithm for multi-objective multicast routing problems | |
| CN105792218A (en) | Optimization method of cognitive radio network with radio frequency energy harvesting capability | |
| Sert et al. | Optimizing the performance of rule-based fuzzy routing algorithms in wireless sensor networks | |
| CN107343303B (en) | Based on the routing optimization method of Duality Decomposition in wireless Mesh netword | |
| Manalastas et al. | Where to go next?: A realistic evaluation of AI-assisted mobility predictors for HetNets | |
| Achankunju et al. | Particle swarm optimization based secure QoS clustering for mobile ad hoc network | |
| Baradaran et al. | NEECH: new energy-efficient algorithm based on the best cluster head in wireless sensor networks | |
| Wang et al. | Multi-objective joint optimization of communication-computation-caching resources in mobile edge computing | |
| CN105451293B (en) | Retransmission method in wireless network, the method and apparatus for determining forwarding strategy | |
| Prakash et al. | Best cluster head selection and route optimization for cluster based sensor network using (M-pso) and Ga algorithms | |
| CN110048945B (en) | Node mobility clustering method and system | |
| Alsharafat et al. | A New Crossover Methods and Fitness Scaling for Reducing Energy Consumption of Wireless Sensor Networks | |
| Mehta et al. | Adaptive Cross-Layer Optimization Using Mimo Fuzzy Control System in Ad-hoc Networks. | |
| Farahmand et al. | A Novel Energy-Efficient Weighted Multi-Level Clustering Protocol |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130213 Termination date: 20140822 |
|
| EXPY | Termination of patent right or utility model |