CN104702325A - MSE-based (mean square error-based) virtual MIMO (multiple input multiple output) user pairing method - Google Patents

Abstract

Disclosed is an MSE-based virtual MIMO user pairing method in the field of wireless communication. The MSE-based virtual MIMO user pairing method discloses three formulas for computing the MSE of every user pair for solving the problem of incapability of predicting the MSSE of paired users and unsuitability for processing pairing of more different users in the prior art. The method comprises computing the MSE of every user pair through the formulas, selecting the user pairs with the MSE not greater than a system-set MSE threshold as selectable paired user pairs, selecting the paired users with the largest number of users in the set of the selectable paired user pairs as optimal system user pairs. The MSE-based virtual MIMO user pairing method overcomes the problem that system MSE performance cannot be ensured in the prior art and accordingly has the advantages of ensuring the system MSE performance and system communication quality and meanwhile achieving dynamic multi-user pairing.

Description

Based on the virtual MIMO user matching method of MSE

Technical field

The invention belongs to communication technical field, further relate to the virtual multiple-input and multiple-output based on mean square error (Mean Square Error) MSE (Multiple Input Multiple Output) the MIMO user's matching method in wireless communication technology field.The present invention can realize in virtual multiple-input and multiple-output mimo system, while dynamically adjusting pairing number of users, under ensureing that system mean square error MSE remains on a specific thresholding.

Background technology

Multiple-input and multiple-output MIMO technology has been widely used in improving spectrum efficiency (Spectral efficiency) SE in various wireless communication system.But due to cost and the size of subscriber equipment, the application of the up link of multiple-input and multiple-output MIMO is limited to due to the difficulty in customer-side practical operation.In order to address this problem, multi-user's multiple-input and multiple-output MIMO transmission, is also referred to as virtual multiple-input and multiple-output MIMO transmission, proposes up link and distributes two or more user, disposes single transmitting antenna at identical frequency band and each user of time slot.Compared with the multiple-input and multiple-output mimo system of routine, virtual multiple-input and multiple-output MIMO is by designing good user grouping strategy to obtain extra multi-user diversity gain.How selecting grouping user to form virtual multiple-input and multiple-output MIMO is the key issue directly affecting its performance.

Patent application document " multiple input/output system MU-MIMO user pairing implementation method " (the publication number CN 102957468A of CHINA PUTIAN INSTITUTE OF TECHNOLOGY Co., Ltd, application number CN201110247744.X, applying date 2011.8.24) in disclose one and realize multi-input multi-output system MIMO user matching method.This invention adopts chooses pairing user collection by maximum feedback channel quality information (Channel Quality Information) CQI, singular value decomposition is done to the channel matrix of pairing user collection and user to be selected collection, obtain corresponding null matrix, the method then concentrating each subscriber channel matrix and null matrix to select new user to enter pairing user collection according to pairing user collection and user to be selected realizes MIMO user's pairing.The deficiency that this invention exists is, cannot predict pairing user to mean square error MSE value, be not suitable for the distribution of process more different user.

Patent application document " a kind of virtual MIMO matching method considering user QoS " (the publication number CN102291842A of University Of Xiangtan, application number CN201110274457.8, applying date 2011.9.16) in disclose one and realize multi-input multi-output system MIMO user matching method.This invention adopts according to user service information, channel condition information, calculates the method for the user data rate in different antennae group after user's pairing to realize multiple-input and multiple-output MIMO user pairing.The deficiency that this invention exists is, two antenna multiple-input and multiple-output MIMO user pairings are only carried out, pairing number of users dynamically cannot be selected to carry out user's pairing, can only throughput of system be ensured when antenna number increases, the mean square error MSE performance of system cannot be ensured.

Summary of the invention

The object of the invention is to the deficiency for above-mentioned prior art, propose a kind of virtual MIMO user matching method based on MSE, while guarantee system mean square error MSE performance, guarantee that throughput of system is maximum.

The concrete thought that the present invention realizes above-mentioned purpose is, first its mean square error MSE value is calculated according to the channel matrix that the user of base station selection is right, then be less than user's centering of threshold value in mean square error MSE value, choose the maximum user of pairing number of users to matching user couple as optimum.Consider that in a time slot, channel matrix is constant, adopts squeeze theorem algorithm to carry out input in signal receiver.

The concrete steps that the present invention realizes above-mentioned purpose are as follows:

(1) current time slots user to be paired is obtained:

From the arrival user stream after the start of base station, obtain current time slots user to be paired;

(2) determine that selectable user is to set:

Permutation and combination, to current time slots total number of users to be paired and the number of users that can match, is carried out in base station, forms selectable user to set;

(3) adopt mean square error MSE value computing formula, calculate the mean square error MSE value that each user is right;

(4) new user pair set is formed:

From selectable user to set, choose and meet M≤M ₀the user couple of condition, forms new user pair set, and wherein, M represents the value of the mean square error MSE that each user is right, M ₀represent the threshold value of mean square error MSE;

(5) user couple that pairing number of users is maximum is chosen:

From new user pair set, choose the pairing user couple that pairing number of users is maximum, as the best pairing user couple in current time slots;

(6) data-signal is sent:

Best pairing user is sent to signal receiver to entrained data-signal by base station;

(7) judged whether that new user flows to reach, if so, performed step (8), otherwise, perform step (9);

(8) base station obtains next time slot user to be paired from newly arrived user's stream, performs step (2);

(9) pairing is completed.

The present invention compared with prior art has the following advantages:

First, because the present invention adopts the method calculating the right mean square error MSE value of each user, overcome in prior art and cannot predict pairing user to mean square error MSE value, the problem being not suitable for processing the pairing of more different users, the present invention is had and can predict the advantage of user to mean square error MSE value.

Second, the user couple meeting mean square error MSE value and be not more than the mean square error MSE threshold condition of default is chosen because the present invention adopts, as the method that optional pairing user is right, overcome the problem that cannot ensure system mean square error MSE performance in prior art, make the present invention have the advantage of guarantee system mean square error MSE performance and system communication quality.

3rd, because the present invention adopts from new user pair set, choose the maximum pairing user of pairing number of users to as the right method of system best pairing user, overcome in prior art and dynamically cannot select to match number of users and carry out the problem of user's pairing, the present invention is had realize the advantage of dynamically multi-user's pairing.

Accompanying drawing explanation

Fig. 1 is flow chart of the present invention;

Fig. 2 is for using completing user of the present invention to match and after data-signal transmission, adopting the analogous diagram of the system Averaged Square Error of Multivariate MSE performance of squeeze theorem;

Fig. 3 is for using completing user of the present invention to match and after data-signal transmission, adopting the analogous diagram of the throughput of system of squeeze theorem.

Embodiment:

Below in conjunction with accompanying drawing, the present invention will be further described.

Embodiments of the invention are that channel matrix carries out under constant communication environment in single time slot.

With reference to accompanying drawing 1, concrete steps of the present invention are as follows:

Step 1, base station obtains current time slots user to be paired.

From the arrival user stream after the start of base station, obtain current time slots user to be paired.

Step 2, base station determines that selectable user is to set.

Permutation and combination, to current time slots total number of users to be paired and the number of users that can match, is carried out in base station, forms selectable user to set.Number of users can be matched at [1, N _r] interior selection, wherein, N _rrepresent the sum of reception antenna.When can match number of users be 1 time, choose from user to be paired a user formed pairing user couple; When can match number of users be 2 time, choose from user to be paired two users formed pairing user couple; Pushing away class successively, is N when matching number of users _rtime, from user to be paired, choose N _rindividual user forms pairing user couple, wherein, and N _rrepresent the sum of reception antenna.

Step 3, adopts mean square error MSE value computing formula, calculates the mean square error MSE value that each user is right.

From three kinds of mean square error MSE value computing formula of following computing formula 1, computing formula 2, computing formula 3, choose one arbitrarily, object of the present invention can be realized in systems in practice.

Computing formula 1, base station according to the following formula, calculates the mean square error MSE value that each user is right

$M=\frac{{\mathrm{\σ}}^2}{Q_i}\underset{k=1}{\overset{Q_i}{\mathrm{\Σ}}}\frac{1}{{\mathrm{\λ}}_k}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, σ ²represent the channel noise power of current time slots, Q _irepresent that user is to U _iin total number of users, Σ represents sum operation, λ _krepresent that user is to U _ithe kth characteristic value of corresponding channel matrix.

Computing formula 2, base station according to the following formula, calculates the mean square error MSE value that each user is right

$M=\frac{{\mathrm{\σ}}^2}{\mathrm{\λ}}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, σ ²represent the channel noise power of current time slots, λ represents that user is to U _ithe minimal eigenvalue of corresponding channel matrix.

Computing formula 3, base station according to the following formula, calculates the mean square error MSE value that each user is right

$M=\frac{{\left|\right|{H_i}^H{H}_i\left|\right|}_F^{Q_i-1}\×{\mathrm{\σ}}^2}{\left|\det \left({H_i}^H{H}_i\right)\right|\×{(Q_i-1)}^{\frac{Q_i-1}{2}}}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, H _irepresent that user is to U _iat the channel matrix of current time slots, H _i ^hrepresent H _ihermitian transpose matrix, represent and ask Q after asking F norm to operate _i-1 power operation, σ ²represent the channel noise power of current time slots, || represent the operation that takes absolute value, det () expression asks determinant to operate, Q _irepresent that user is to U _iin total number of users, expression is asked power operates.

Step 4, forms new user pair set.

From selectable user to set, choose and meet M≤M ₀the user couple of condition, forms new user pair set, and wherein, M represents the value of the mean square error MSE that each user is right, M ₀represent the threshold value of mean square error MSE, the threshold value M of mean square error MSE ₀setting be that mean square error MSE performance needed for system determines.Such as, if the mean square error MSE value after system requirements detection is not more than 0.06, then we arrange M ₀be 0.06.

Step 5, chooses the user couple that pairing number of users is maximum.

From new user pair set, choose the pairing user couple that pairing number of users is maximum, as the best pairing user couple in current time slots.

Step 6, best pairing user is sent to signal receiver to entrained data-signal by base station.

Step 7, has judged whether that new user flows to and has reached, reached if there is new user to flow to, and performs step (8), otherwise, perform step (9).

Step 8, base station obtains next time slot user to be paired from newly arrived user's stream, performs step (2).

Step 9, completes pairing.

Below in conjunction with accompanying drawing 2 and accompanying drawing 3, effect of the present invention is further described.

1. simulated conditions:

Emulation of the present invention is carried out in the radio communication scene of single base station, and the number of users to be paired that base station obtains at every turn is set to 8, the total N of reception antenna _rbe set to 4, modulation system is 16-QAM, detection mode is squeeze theorem, and emulation experiment of the present invention adopts the computing formula 1 in mean square error MSE value three kinds of computing formula, computing formula 2, computing formula 3 to carry out the calculating of mean square error MSE value, the threshold value M of initialization system mean square error MSE respectively ₀be 0.06.

2. analysis of simulation result:

Simulation result of the present invention is as shown in Fig. 2, Fig. 3.Fig. 2 is for using completing user of the present invention to match and after data-signal transmission, adopting the analogous diagram of the system Averaged Square Error of Multivariate MSE performance of squeeze theorem.Fig. 3 is for using completing user of the present invention to match and after data-signal transmission, adopting the analogous diagram of the throughput of system of squeeze theorem.Representing with the curve of prismatic mark in Fig. 2 adopts computing formula 1 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the system Averaged Square Error of Multivariate performance after data-signal transmission, detection.Representing with the curve of circular indicia in Fig. 2 adopts computing formula 2 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the system Averaged Square Error of Multivariate performance after data-signal transmission, detection.Representing with the curve of square mark in Fig. 2 adopts computing formula 3 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the system Averaged Square Error of Multivariate performance after data-signal transmission, detection.Representing with the curve of prismatic mark in Fig. 3 adopts computing formula 1 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the throughput of system after data-signal transmission, detection.Representing with the curve of circular indicia in Fig. 3 adopts computing formula 2 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the throughput of system after data-signal transmission, detection.Representing with the curve of square mark in Fig. 3 adopts computing formula 3 to calculate mean square error MSE value, carries out optimal user pairing and chooses, the system Averaged Square Error of Multivariate throughput after data-signal transmission, detection.We can be as seen from Figure 2, the computing formula of the three kinds of mean square error MSE adopting the present invention to mention, calculate each user to mean square error MSE value, and then choose best pairing user couple, carried the method that data-signal sends to signal receiver, the all successful Averaged Square Error of Multivariate MSE value by the system of system signal noise ratio between [4,30] dB has been limited under the threshold value of system mean square error MSE.Our composition graphs 2, can be as seen from Figure 3, the computing formula of the three kinds of mean square error MSE adopting the present invention to mention, calculate each user to mean square error MSE value, and then choose the right method of best pairing user, under different system signal to noise ratio environment, all there is different throughput of system, thus can illustrate that the present invention has the function realizing dynamic subscriber's pairing.

Claims (7)

1., based on the virtual MIMO user matching method of MSE, its step comprises as follows:

(1) current time slots user to be paired is obtained:

From the arrival user stream after the start of base station, obtain current time slots user to be paired;

(2) determine that selectable user is to set:

Permutation and combination, to current time slots total number of users to be paired and the number of users that can match, is carried out in base station, forms selectable user to set;

(3) adopt mean square error MSE value computing formula, calculate the mean square error MSE value that each user is right;

(4) new user pair set is formed:

From selectable user to set, choose and meet M≤M ₀the user couple of condition, forms new user pair set, and wherein, M represents the value of the mean square error MSE that each user is right, M ₀represent the threshold value of mean square error MSE;

(5) user couple that pairing number of users is maximum is chosen:

From new user pair set, choose the pairing user couple that pairing number of users is maximum, as the best pairing user couple in current time slots;

(6) data-signal is sent:

Best pairing user is sent to signal receiver to entrained data-signal by base station;

(7) judged whether that new user flows to reach, if so, performed step (8), otherwise, perform step (9);

(8) base station obtains next time slot user to be paired from newly arrived user's stream, performs step (2);

(9) pairing is completed.

2. the virtual MIMO user matching method based on MSE according to claim 1, it is characterized in that, the matched number of users described in step (2) is at [1, N _r] interior selection, wherein, N _rrepresent the sum of base station receive antenna.

3. the virtual MIMO user matching method based on MSE according to claim 1, it is characterized in that, the mode of step (2) described permutation and combination is, when can match number of users be 1 time, choose from user to be paired a user formed pairing user couple; When can match number of users be 2 time, choose from user to be paired two users formed pairing user couple; Pushing away class successively, is N when matching number of users _rtime, from user to be paired, choose N _rindividual user forms pairing user couple, wherein, and N _rrepresent the sum of base station receive antenna.

4. the virtual MIMO user matching method based on MSE according to claim 1, is characterized in that, the mean square error MSE value computing formula described in step (3) is as follows:

$M=\frac{{\mathrm{\σ}}^2}{Q_i}\underset{k=1}{\overset{Q_i}{\mathrm{\Σ}}}\frac{1}{{\mathrm{\λ}}_k}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, σ ²represent the channel noise power of current time slots, Q _irepresent that user is to U _iin total number of users, Σ represents sum operation, λ _krepresent that user is to U _ia kth characteristic value of corresponding channel matrix.

5. the virtual MIMO user matching method based on MSE according to claim 1, is characterized in that, the mean square error MSE value computing formula described in step (3) is as follows:

$M=\frac{{\mathrm{\σ}}^2}{\mathrm{\λ}}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, σ ²represent the channel noise power of current time slots, λ represents that user is to U _ithe minimal eigenvalue of corresponding channel matrix.

6. the virtual MIMO user matching method based on MSE according to claim 1, is characterized in that, the mean square error MSE value computing formula described in step (3) is as follows:

$M=\frac{{\left|\right|{H_i}^H{H}_i\left|\right|}_F^{Q_i-1}\×{\mathrm{\σ}}^2}{\left|\det \left({H_i}^H{H}_i\right)\right|{(Q_i-1)}^{\frac{Q_i-1}{2}}}$

Wherein, M represents that user is to U _ithe value of mean square error MSE, U _irepresent that selectable user is to i-th user couple in set, H _irepresent that user is to U _iat the channel matrix of current time slots, H _i ^hrepresent H _ihermitian transpose matrix, represent and ask Q after asking F norm to operate _i-1 power operation, σ ²represent the channel noise power of current time slots, || represent the operation that takes absolute value, det () represents the operation asking determinant, Q _irepresent that user is to U _iin total number of users, expression is asked power operates.

7. the virtual MIMO user matching method based on MSE according to claim 1, is characterized in that, the threshold value M of step (4) described mean square error MSE ₀setting be that mean square error MSE performance needed for system determines.