CN105188144B - Stream allocation proportion equity dispatching method based on MU-MIMO - Google Patents
Stream allocation proportion equity dispatching method based on MU-MIMO Download PDFInfo
- Publication number
- CN105188144B CN105188144B CN201510495986.9A CN201510495986A CN105188144B CN 105188144 B CN105188144 B CN 105188144B CN 201510495986 A CN201510495986 A CN 201510495986A CN 105188144 B CN105188144 B CN 105188144B
- Authority
- CN
- China
- Prior art keywords
- value
- user
- normal distribution
- iteration
- emission probability
- 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
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a kind of stream allocation proportion equity dispatching method based on MU MIMO, mainly solves the problems, such as the Proportional Fair for ignoring QoS of customer in the prior art.Implementation step is：1, corresponding parameter is arranged according to actual schedule problem；2, meet user's weights in the mean value computation utility function of normal distribution according to emission probability；3, the sample value of emission probability is generated and according to the utility function value of user's weight computing sample；4, the good sample value of performance is filtered out to update the mean value of normal distribution；5, it records the situation of change of mean value and switches the variance of different phase according to corresponding rule；6, iterative process is terminated according to end condition, obtains the scheduling result for meeting equitable proportion.The present invention in the case that usable radio resources abundance and it is insufficient can obtain satisfied scheduling result, and computation complexity is low, iterations are few, improves the dispatching efficiency of base station, can be used for single base station system using MIMO technology.
Description
Technical field
The invention belongs to wireless communication technology fields, are related to a kind of stream allocation proportion fair method based on MU-MIMO, can
For single base station system using MIMO technology.
Background technology
With the active demand now to high data rate system, 802.11ac is as new after 802.11n
Generation wireless local area network technology standard is got the attention.802.11ac is made of several important technologies, including being claimed
Be multi-user's multiple-input and multiple-output MU-MIMO technologies.MU-MIMO refer to multiple users in same running time-frequency resource with base station into
Row communication, base station end and user terminal are multiple antennas here.MU-MIMO not only realizes spatial division multiplexing in the communication of single user
With, and space division multiple access is realized between multiple users, so MU-MIMO can not only increase the rate of single user, more attach most importance to
What is wanted is the handling capacity and the availability of frequency spectrum for improving whole system.
It is most of in relation to MU-MIMO resource allocation algorithms be concentrated mainly on to radio resource it is efficient utilization and be seldom concerned about
The quality of service requirement of user, the simple utilization rate for pursuing radio resource can cause the unfairness of QoS of customer.Work as user
Between channel gain when differing greatly, the preferable user of channel condition can obtain excessive radio resource, and channel condition is poor
Therefore user service request may be not being met.
In view of the above problems, existing distribution method (1：V.Valls and D.J.Leith,”Proportional Fair
MU-MIMO in802.11WLANs,”IEEE Wireless Communications Letters,vol.3,no.2,
pp.221-224,2014.2：A.Checco and D.J Leiith,”Proportional Fairness in
802.11Wireless LANs,”IEEE Communication Letters,vol.15,no.8,pp.807-809,2011)
The proportional fair algorithm about the MU-MIMO stream distribution in 802.11ac WLANs has been obtained using convex optimization method.Side
Method (3：D.J.Leith,Q.Cao,V.G.Subramanian,“Max-Min Fairness in 802.11Mesh
Networks, " IEEE/ACM Transaction on Netwoking, vol.20, no.3, pp.756-769,2012.) it will be another
A kind of fairness policy --- minimax is fair --- is applied in Mesh network, and Mesh network is realized most using logarithmic convexity
It is big minimum fair.
Above method only has studied the case where resource mean allocation, but has ignored the quality of service requirement of each user to money
The influence of source distribution, especially when available resources deficiency, these mean allocations are not best allocation plan, this is because
In inadequate resource, the high user of quality of service requirement is assigned relatively small number of resource, and quality of service requirement it is low user it is anti-
And it is assigned relatively large number of resource, obviously unfair.
Invention content
It is an object of the invention to the deficiencies for the above method, propose a kind of resource allocation methods based on MU-MIMO,
Under the premise of meeting QoS of customer, equitable proportion and the maximum good folding of throughput of system between user are obtained in system
In.
To achieve the above object, technical method of the invention includes the following steps：
(1) initiation parameter is set：
If Base Transmitter pattern count is F, it is M to need the number of users serviced, and the mean vector of emission probability is gt=
(g1,t,...,gk,t,...,gF,t), wherein gk,tIndicate the sending probability Normal Distribution of the kth kind emission mode of base station
Mean value, initial valueThe equal Normal Distribution of emission probability；
The standard deviation for being located at the iteratively faster stage is σ1, it is σ in the standard deviation of search stabilization sub stage2, and 0 ＜ σ2＜
σ1＜ 0.05；
Practical business demand according to user sets the service quality Q=[Q of user1,...,Qm,...,QM], wherein Qm
Indicate the quality of service requirement of m-th of user, m=1 ..., M；
Iteration ends parameter d=5 is set；
T indicates iterations, initializes t=1.
(2) t=t+1 is enabled, user's weights are calculated：
After (2a) calculates t-1 iteration, the spatial flow that each user gets is：
[S1,t-1,...,Sm,t-1,...,SM,t-1]=[g1,t-1,...,gk,t-1,...,gF,t-1]V,
Wherein Sm,t-1Indicate the space that m-th of user obtains according to the emission probability of each emission mode after the t-1 times iteration
Stream, V are Base Transmitter mode matrix；
The spatial flow that (2b) gets according to user after t-1 iteration calculates the service miss rate of the user in the t times iteration
cm,t：
Work as Sm,t-1≥QmWhen, cm,t=0；
(2c) calculates user's weights ω according to the service miss rate of obtained userm,t,
(3) normal distribution obeyed according to sending probabilityN group sample values are randomly generated, normalization is calculated
Sample value π ' afterwardsn,k,t, and according to the sample value π ' after normalizationn,k,tCalculate sample utility function value H (π 'n,t)：
(4) by the utility function value H (π ' of all samplesn,t) ascending arrangement, the λ value is chosen as the t times iteration
Utility function value γ at ρ quantilest=H(λ), whereinExpression rounds up to ρ N, and 0 ＜ ρ≤
0.1；
(6) according to (5) newer mean value gk,tSituation of change in each iterative process, by normal distribution
Variance by the iteratively faster stageIt is switched to the search stabilization sub stage
(7) judged whether to terminate iterative process according to end condition：
If meeting t >=d, and γt=γt-1=...=γt-d, then iterative process is terminated, at this time minimum effect in the t times iteration
With the corresponding sampled value of functional value, as meet the optimal emission probability of the Base Transmitter pattern of stream allocation proportion justice, base station
Emission mode is selected according to the optimal emission probability, realizes the equity dispatching of base station spatial flow；
Otherwise, it goes to step (2) and continues next iteration, until meeting end condition.
The present invention has the following advantages：
1, the present invention in the distribution of the stream of base station due to having taken into account two factors of QoS of customer and equitable proportion so that
Reach equitable proportion and the maximized good compromise of throughput of system between user under the premise of guaranteeing QoS of customer and requiring；
2, since the present invention is using the important of user's weights characterization allowable resource Different Optimization target in varied situations
Property, therefore different scheduling results can be obtained according to allowable resource abundance or deficiency：
When allowable resource abundance, the service miss rate of user is zero, the user's weights obtained by service miss rate
It is 1, so that realizing importance of the maximized importance of throughput of system more than equitable proportion between user, ensure that
QoS of customer reaches throughput of system maximum under the premise of requiring；
When allowable resource deficiency, the service miss rate of user is there are nonzero value, the user obtained by service miss rate
There is the numerical value more than 1 in weights, so that it is maximized to realize that the importance of equitable proportion between user is more than throughput of system
Importance ensures that user obtains fair service；
3, computation complexity of the present invention is low, iterations are few, improves the efficiency of base station resource scheduling.
Description of the drawings
Fig. 1 is the implementation flow chart of the present invention；
Fig. 2 is in usable radio resources deficiency with the present invention and conventional method to different service quality demand user
Scheduling result comparison diagram；
Fig. 3 is in usable radio resources deficiency with the present invention and conventional method to different service quality demand user
Scheduling fairness comparison diagram；
Fig. 4 is in usable radio resources abundance with the present invention and conventional method to different service quality demand user
Scheduling result comparison diagram；
Fig. 5 is in usable radio resources abundance with the present invention and conventional method to different service quality demand user
Scheduling fairness comparison diagram.
Specific implementation mode
Referring to Fig.1, steps are as follows for specific implementation of the invention：
Step 1, initiation parameter is set.
If Base Transmitter pattern count is F, it is M to need the number of users serviced, and the mean vector of emission probability is gt=
(g1,t,...,gk,t,...,gF,t), wherein gk,tIndicate the sending probability Normal Distribution of the kth kind emission mode of base station
Mean value, initial valueThe equal Normal Distribution of emission probability；
Since the size of the standard deviation of normal distribution determines the sample range of emission probability, when standard deviation is larger, sampling
Range is larger, and iterations are few, but last result can be caused inaccurate；And when standard deviation is smaller, sample range is smaller,
Although can obtain more accurately solving, iterations can be caused excessive, it accordingly can be in the iteratively faster stage using larger
Standard deviation with iteratively faster to optimal emission probability near, be then changed to smaller standard deviation again, obtain more accurate result.
Therefore standard deviation is set as two, that is, the standard deviation for being located at the iteratively faster stage is σ1, equal in the standard deviation of search stabilization sub stage
For σ2, and 0 ＜ σ2＜ σ1＜ 0.05；
Practical business demand according to user sets the service quality Q=[Q of user1,...,Qm,...,QM], wherein Qm
Indicate the quality of service requirement of m-th of user, m=1 ..., M；
Iteration ends parameter d=5 is set；
T indicates iterations, initializes t=1.
Step 2, t=t+1 is enabled, user's weights are calculated.
The space that can be got by each user after the t-1 times iteration of mean value computation of the newer normal distribution of the t-1 times iteration
Stream calculates user's weights in current iteration according to the quality of service requirement of user and the spatial flow that can get；
After (2a) calculates t-1 iteration, the spatial flow that each user gets is：
[S1,t-1,...,Sm,t-1,...,SM,t-1]=[g1,t-1,...,gk,t-1,...,gF,t-1]V,
Wherein Sm,t-1Indicate the space that m-th of user obtains according to the emission probability of each emission mode after the t-1 times iteration
Stream, V are Base Transmitter mode matrix；
The spatial flow that (2b) gets according to user after t-1 iteration calculates the service miss rate of the user in the t times iteration
cm,t：
Work as Sm,t-1＜ QmWhen,
Work as Sm,t-1≥QmWhen, cm,t=0；
(2c) calculates user's weights ω according to the service miss rate of obtained userm,t,
WhenWhen, ωm,t=1,
WhereinIndicate that the average value of each user service miss rate, α expressions meet QoS of customer
Importance index, β indicates to meet the importance index of equitable proportion, and α, β are positive integer；
In above-mentioned (2a)-(2c) step, when allowable resource deficiency, all QoS of customer cannot be met and wanted
It asks, therefore services miss rate there are nonzero value, cause to service miss rate difference due to the difference of service routine, pass through each use
The comparison of the service miss rate and average service miss rate at family, obtains corresponding user's weights：When the service miss rate of user is high
When average service miss rate, corresponding user's weights are bigger, and more resources will be distributed to the user in scheduling of resource；
When the service miss rate of user is less than average service miss rate, corresponding user's weights are smaller, will be in scheduling of resource
The user distributes less resource, to reach the equitable proportion between user；
In usable radio resources abundance, since the quality of service requirement of user can be met, so service miss rate is equal
It is 0, weights 1, equitable proportion becomes by-end between the user of user's weights characterization at this time, and the main target of scheduling of resource becomes
To maximize the handling capacity of system, to realize throughput of system maximum.
Step 3, normalization random sample value is generated.
The normal distribution obeyed according to sending probabilityN group sample values are randomly generated, and are calculated by following formula
Sample value π ' after normalizationn,k,t：
Wherein πn,k,tIndicate the emission probability of kth kind emission mode in the t times iteration n-th set of samples value, k=1 ..., F,
N=1 ..., N, N >=100.
Step 4, according to the sample value π ' after normalizationn,k,tCalculate sample utility function value H (π 'n,t)。
WhereinFor indicative function, whenWhen,Value is 1, no
It is then 0；vk,mFor the element in Base Transmitter matrix, indicate that the kth kind emission mode of base station distributes to the space of m-th of user
Stream.
Step 5, the utility function value at ρ quantiles is calculated.
Utility function is worth that smaller sample value its performance is better, and ρ quantiles are by sampled value by performance quality in the present invention
It is divided into two-part separation.The sample that utility function value is less than utility function value at ρ quantiles is the good sample of performance, and will
These screening samples come out the mean value for updating normal distribution in next step；Utility function value is more than utility function at ρ quantiles
The sample of value is the sample of poor performance, is abandoned.
By the utility function value H (π ' of all samplesn,t) ascending arrangement, it chooses the λ value and is used as the t times iteration in ρ
Utility function value γ at quantilet=H(λ), whereinExpression rounds up to ρ N, 0 ρ≤0.1 ＜.
Step 6, the mean value of normal distribution is updated.
It is less than γ using utility function in N number of sampletSample value come update emission probability obedience normal distributionMean value gk,t-1, i.e., by normal distributionMean value gk,t-1It is updated to：
WhereinFor indicative function, as H (π 'n,t)≤γtWhen, indicative function value is 1, as H (π 'n,t) ＞ γt
When, indicative function value is 0, k=1 ..., F；
It is gradually close to the optimal emission probability search in base station for meeting equitable proportion according to this mean value update method.
Step 7, switch the variance yields of different phase normal distribution.
According to newer mean value g in step 6k,tSituation of change in each iterative process, according to following regular by normal state
DistributionVariance by the iteratively faster stageIt is switched to the search stabilization sub stage
As t=2, with vectorial R=[R1,...,Rk,...,RF] record mean vector gtSituation of change, wherein RkFor
Record gk,tThe mark value of situation of change：If gk,tRelative to its initial valueBecome smaller, then enables Rk=-1；If gk,tRelative at the beginning of it
Initial valueBecome larger, then enables Rk=1；If gk,tRelative to its initial valueDo not change, then enables Rk=0；
As t ＞ 2, with vectorial P=[P1,...,Pk,...,PF] record mean vector gtSituation of change, wherein PkFor
Record gk,tThe mark value of situation of change：If gk,tRelative to gk,t-1Become smaller, then enables Pk=-1；If gk,tRelative to gk,t-1Become larger, then
Enable Pk=1；If gk,tRelative to gk,t-1Do not change, then enables Pk=0；
Y=P-R is calculated, if number of the absolute value equal to 2 is more than F/2 in the vector element of Y, by normal distributionVariance by the iteratively faster stageIt is switched to the search stabilization sub stageOtherwise, do not switch normal distribution
Variance.
Step 8, judged whether to terminate iterative process according to end condition.
If meeting t >=d, and γt=γt-1=...=γt-d, then iterative process is terminated, at this time minimum effect in the t times iteration
With the corresponding sampled value of functional value, as meet the optimal emission probability of the Base Transmitter pattern of stream allocation proportion justice, base station
Emission mode is selected according to the optimal emission probability, realizes the equity dispatching of base station spatial flow；
Otherwise, it goes to step 2 and continues next iteration, until meeting end condition.
The impact of performance of the present invention can be further illustrated by following emulation：
A, simulated conditions
In simulations, by taking the downlink of the access point AP of a support MU-MIMO in WLAN as an example, the AP is total
4 users need to be serviced.Assuming that sharing 4 kinds of emission modes in the AP, emission mode matrix is
Simultaneously in order to which the fair degree of the spatial flow of user is distributed in more objective appraisal, we use and reach herein
To evaluation criterion of the Jain justices index as simulation result of the ratio of service quality, Jain justice Index Definitions are as follows：
Wherein [x1,...,xm,...,xM] indicate that scheduling result, the value of Jain justice sex index indicate to get over closer to 1
Justice is equal to 1 and indicates absolutely fair.
The hits N=200 of each iteration, quantile ρ=0.1, calculate two positive integer constants α=8, β of weights=
4。
B, emulation content
Emulation 1：Insufficient in usable radio resources, the standard deviation for choosing the iteratively faster stage is σ1=0.01,
The standard deviation for searching for the stabilization sub stage is σ2=0.005, the spatial flow quality of service requirement number of each user be Q=[1.2,2.2,
2.8,3.6], dispatching method using the present invention carries out resource allocation with conventional method to different service quality demand user, point
With the results are shown in Figure 2, the scheduling fairness comparison of two methods is as shown in Figure 3.
Emulation 2：In the case of usable radio resources abundance, the standard deviation for choosing the iteratively faster stage is σ1=0.015,
The standard deviation for searching for the stabilization sub stage is σ2=0.008, the spatial flow quality of service requirement number of each user be Q=[0.5,1,1.5,
2], dispatching method using the present invention carries out resource allocation, allocation result with conventional method to different service quality demand user
As shown in figure 4, the scheduling fairness comparison of two methods is as shown in Figure 5.
C, analysis of simulation result
The sending probability that dispatching method using the present invention obtains each sending mode in base station in Fig. 2 be π=[0.3653,
0.1202,0.0471,0.4674], the quantity for the spatial flow that user 1 divides to user 4 be S=[0.8020,1.5554,
1.9638,2.516], the total space fluxion that base station is sent is 6.8374；And conventional allocation algorithm is not due to considering service quality
The probability of obtained each emission mode is π=[1/3,0,1/3,1/3], the quantity of the spatial flow that each user assigns to be S=[1,
2,2,2], the total space fluxion that base station is sent is 7.By comparing the scheduling result of two methods as it can be seen that in inadequate resource
In the case of, can be different clothes though the present invention has certain loss of throughput compared to the indiscriminate scheduling mode of conventional method
The user of business quality requirement provides the service of same services miss rate.
The ratio that the spatial flow that distributes of the present invention reaches user 1 to 4 service quality of user in Fig. 3 is respectively 66.83%,
70.7%, 70.14%, 69.89%, Jain justice indexes are 0.9995；Conventional proportional fair allocat method reaches user 1 to use
The ratio of the service quality at family 4 is respectively 83.33%, 90.91%, 71.43%, 55.56%, Jain justice indexes be
0.9695.By comparing the Jain justices index of two methods as it can be seen that the present invention is adjusted compared to tradition under out-of-resource condition
Degree method, scheduling of resource result have reached higher fairness.
The sending probability that dispatching method using the present invention obtains each sending mode in base station in Fig. 4 be π=[0.5774,
0.0046,0.0783,0.3398], the quantity for the spatial flow that user 1 assigns to user 4 be S=[0.5056,2.4662,
1.5158,2.9938], the total space fluxion that base station is sent is 7.4814；And conventional allocation algorithm is not due to considering Service Quality
The probability of each emission mode measured is π=[1/3,0,1/3,1/3], and the quantity for the spatial flow that each user assigns to is S=
[1,2,2,2], the total space fluxion that base station is sent are 7.By comparing known to the scheduling result of two methods in resource abundance
In the case of, compared to conventional scheduling method, the present invention can improve system under the premise of ensureing that each QoS of customer requires
The handling capacity of system.
The ratio that the spatial flow that distributes of the present invention reaches user 1 to the service quality of user 4 in Fig. 5 is respectively
101.12%, 246.62%, 101.05%, 149.69%, Jain justice indexes are 0.8638；Conventional proportional fair allocat method
The ratio for reaching user 1 to the service quality of user 4 is respectively 200%, 200%, 133.33%, 100%, Jain justice indexes
It is 0.9304.By comparing the Jain justices index contrast of two methods it is found that in the case of resource abundance, compared to tradition
Dispatching method, the present invention can improve the handling capacity of system using smaller scheduling fairness as cost.
By the above simulation result and analysis as can be seen that in the present invention, when radio resource deficiency, meeting ratio public affairs
Flat is most important target, and handling capacity maximum is then by-end, and higher tune is obtained by cost of smaller throughput of system
Spend fairness；When radio resource abundance, to improve based on throughput of system after the quality of service requirement for meeting each user
Target is wanted, though the fairness dispatched at this time is declined, but increases the handling capacity of system.
Claims (2)
1. the stream allocation proportion equity dispatching method based on MU-MIMO, includes the following steps：
(1) initiation parameter is set：
If Base Transmitter pattern count is F, it is M to need the number of users serviced, and the mean vector of emission probability is gt=(g1,t,...,
gk,t,...,gF,t), wherein gk,tIndicate the mean value of the emission probability of kth kind emission mode Normal Distribution, initial valueThe equal Normal Distribution of emission probability；
The standard deviation for being located at the iteratively faster stage is σ1, it is σ in the standard deviation of search stabilization sub stage2, and 0 ＜ σ2＜ σ1＜
0.05；
Practical business demand according to user sets the service quality Q=[Q of user1,...,Qm,...,QM], wherein QmIt indicates
The quality of service requirement of m-th of user, m=1 ..., M；
Iteration ends parameter d=5 is set；
T indicates iterations, initializes t=1.
(2) t=t+1 is enabled, user's weights are calculated：
After (2a) calculates t-1 iteration, the spatial flow that each user gets is：
[S1,t-1,...,Sm,t-1,...,SM,t-1]=[g1,t-1,...,gk,t-1,...,gF,t-1]V,
Wherein Sm,t-1Indicate the spatial flow that m-th of user obtains according to the emission probability of each emission mode after the t-1 times iteration, V
For Base Transmitter mode matrix；
The spatial flow that (2b) gets according to user after t-1 iteration calculates the service miss rate c of the user in the t times iterationm,t：
Work as Sm,t-1＜ QmWhen,
Work as Sm,t-1≥QmWhen, cm,t=0；
(2c) calculates user's weights ω according to the service miss rate of obtained userm,t,
WhenWhen,
WhenWhen, ωm,t=1,
WhereinIndicate that the average value of each user service miss rate, α expressions meet the important of QoS of customer
Property index, β indicates to meet the importance index of equitable proportion, and α, β are positive integer；
(3) normal distribution obeyed according to emission probabilityN group sample values are randomly generated, after calculating normalization
Sample value π 'n,k,t, and according to the sample value π ' after normalizationn,k,tCalculate sample utility function value H (π 'n,t)：
WhereinFor indicative function, whenWhen,Value is 1, is otherwise 0；
vk,mIndicate that the kth kind emission mode of base station distributes to the spatial flow of m-th of user；
(4) by the utility function value H (π ' of all samplesn,t) ascending arrangement, it chooses the λ value and is used as the t times iteration in ρ
Utility function value γ at quantilet=H(λ), wherein Expression rounds up to ρ N, 0 ρ≤0.1 ＜；
(5) by normal distributionMean value gk,t-1It is updated to：
WhereinFor indicative function, as H (π 'n,t)≤γtWhen, indicative function value is 1, as H (π 'n,t) ＞ γtWhen, show
Property functional value be 0, k=1 ..., F；
(6) according to (5) newer mean value gk,tSituation of change in each iterative process, by normal distributionSide
Difference is by the iteratively faster stageIt is switched to the search stabilization sub stage
As t=2, with vectorial R=[R1,...,Rk,...,RF] record mean vector gtSituation of change, wherein RkFor recording
gk,tThe mark value of situation of change：If gk,tRelative to its initial valueBecome smaller, then enables Rk=-1；If gk,tRelative to its initial valueBecome larger, then enables Rk=1；If gk,tRelative to its initial valueDo not change, then enables Rk=0；
As t ＞ 2, with vectorial P=[P1,...,Pk,...,PF] record mean vector gtSituation of change, wherein PkFor recording
gk,tThe mark value of situation of change：If gk,tRelative to gk,t-1Become smaller, then enables Pk=-1；If gk,tRelative to gk,t-1Become larger, then enables Pk
=1；If gk,tRelative to gk,t-1Do not change, then enables Pk=0；
Y=P-R is calculated, if number of the absolute value equal to 2 is more than F/2 in the vector element of Y, by normal distributionVariance by the iteratively faster stageIt is switched to the search stabilization sub stageOtherwise, do not switch normal distribution
Variance；
(7) judged whether to terminate iterative process according to end condition：
If meeting t >=d, and γt=γt-1=...=γt-d, then iterative process is terminated, at this time minimum effectiveness letter in the t times iteration
The corresponding sampled value of numerical value as meets the optimal emission probability of the Base Transmitter pattern of stream allocation proportion justice, base station foundation
The optimal emission probability selects emission mode, realizes the equity dispatching of base station spatial flow；
Otherwise, it goes to step (2) and continues next iteration, until meeting end condition.
2. according to the method described in claim 1, calculating the sample value π ' after normalization in the wherein described step (3)n,k,t, pass through
Following formula calculates：
Wherein πn,k,tIndicate the emission probability of kth kind emission mode in the t times iteration n-th set of samples value, n=1 ..., N, k=
1 ..., F, N >=100.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510495986.9A CN105188144B (en) | 2015-08-12 | 2015-08-12 | Stream allocation proportion equity dispatching method based on MU-MIMO |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510495986.9A CN105188144B (en) | 2015-08-12 | 2015-08-12 | Stream allocation proportion equity dispatching method based on MU-MIMO |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105188144A CN105188144A (en) | 2015-12-23 |
| CN105188144B true CN105188144B (en) | 2018-08-21 |
Family
ID=54909922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510495986.9A Active CN105188144B (en) | 2015-08-12 | 2015-08-12 | Stream allocation proportion equity dispatching method based on MU-MIMO |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105188144B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102612092A (en) * | 2012-03-30 | 2012-07-25 | 西安交通大学 | Radio resource management function game-based method for optimizing radio resources |
| CN102811490A (en) * | 2012-07-06 | 2012-12-05 | 华中科技大学 | MISO-OFDM (Multiple-Input Single-Output-Orthogonal Frequency Division Multiplexing) downlink resource distribution method based on energy efficiency |
| CN103763086A (en) * | 2014-01-27 | 2014-04-30 | 湖北工业大学 | Multi-user multi-channel collaborative spectrum sensing method based on filter bank |
-
2015
- 2015-08-12 CN CN201510495986.9A patent/CN105188144B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102612092A (en) * | 2012-03-30 | 2012-07-25 | 西安交通大学 | Radio resource management function game-based method for optimizing radio resources |
| CN102811490A (en) * | 2012-07-06 | 2012-12-05 | 华中科技大学 | MISO-OFDM (Multiple-Input Single-Output-Orthogonal Frequency Division Multiplexing) downlink resource distribution method based on energy efficiency |
| CN103763086A (en) * | 2014-01-27 | 2014-04-30 | 湖北工业大学 | Multi-user multi-channel collaborative spectrum sensing method based on filter bank |
Non-Patent Citations (1)
| Title |
|---|
| Proportional Fair MU-MIMO in 802.11 WLANS;V.Valls， D.J.Leith;《IEEE WIRELESS COMMUNICATIONS LETTERS》;20140430;第3卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105188144A (en) | 2015-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yaffe et al. | Stable matching for channel access control in cognitive radio systems | |
| CN104410480A (en) | Large-scale fading based pilot frequency distribution method in large-scale MIMO (multiple input multiple output) system | |
| CN103228048B (en) | A kind of uplink multi-users frequency domain resource distribution method based on skew single-carrier modulated | |
| CN109922487A (en) | A kind of resource allocation methods under downlink MIMO-NOMA network | |
| WO2009083376A2 (en) | Virtual mimo system and apparatus thereof | |
| CN107172710B (en) | resource allocation and service access control method based on virtual subnet | |
| CN103249160A (en) | Resource allocation method under CoMP transfer mode in LTE-A system | |
| CN106027214A (en) | Pilot distribution method of multi-cell large-scale MIMO system | |
| CN104602353A (en) | Wireless resource allocation method for D2D links in cellular mobile communication system | |
| CN102056308B (en) | Resource allocation method and device | |
| CN102752763B (en) | Realistic type proportional fair scheduling metric calculation method and device in LTE (Long Time Evolution) system | |
| CN105611640B (en) | A kind of adjustable CoMP downlink user dispatching method of equitable degree | |
| CN110418360A (en) | Wirelessly taking can network multi-user subcarrier bit combined distributing method | |
| Chen et al. | Call blocking probability and bandwidth utilization of OFDM subcarrier allocation in next-generation wireless networks | |
| CN105188144B (en) | Stream allocation proportion equity dispatching method based on MU-MIMO | |
| CN107087309A (en) | Many base station system proportional fair dispatching methods under user QoS fixed rate constraint | |
| CN111711986A (en) | UC-UDN proportional fair resource allocation method in 5G communication system | |
| CN105873127A (en) | Heuristic user connection load balancing method based on random decision | |
| CN106209191A (en) | A kind of MU mimo system true environment low complex degree user choosing method | |
| CN107493606B (en) | A kind of the timesharing dispatching method and device of user terminal | |
| Srinivasan et al. | Maximizing throughput in MIMO networks with variable rate streams | |
| CN112996118B (en) | NOMA downlink user pairing method and storage medium | |
| CN105163392B (en) | Ensure the multi-BS (base station) multi-user proportional fair dispatching method of service quality | |
| CN109982336B (en) | Method for constructing virtual macro base station by ultrahigh frequency small base station in 5G network | |
| CN105960005A (en) | Power control method for ensuring user fairness in ultra-dense network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |