CN102185683A - Signal-to-leakage-and-noise ratio (SLNR) rule statistic-based MIMO (Multiple Input Multiple Output) multi-user downlink transmission method - Google Patents

Abstract

The invention provides a signal-to-leakage-and-noise ratio (SLNR) rule statistic-based MIMO (Multiple Input Multiple Output) multi-user downlink transmission method which comprises the following steps of: acquiring a sending correlation matrix between a base station and a user by the base station in a feedback or hidden feedback mode according to a channel estimated result; calculating a downlink precoding vector space to be formed by a dispersion Fourier matrix by utilizing the information according to the lower bound of the maximum statistics of the SLNR; and further dispatching all users suitable for downlink transmission and carrying out power allocation among users by the base station through diagonal matrix features of feature values of the sending correlation matrix according to the principle of mutual coupling of diagonal matrix energy of different feature values, thereby completing the MIMO multi-user downlink transmission method on the basis. The MIMO multi-user downlink transmission method improves the frequency spectrum utilization ratio and power efficiency of a communication system, and greatly reduces the system cost needed by the base station for carrying out downlink precoding transmission; and the base station adjusts the sending parameters according to the channel statistic channel state information so as to obtain a higher system transmission rate with lower system cost.

Description

Leakage is made an uproar than the MIMO multiuser downstream transmission method of criterion based on the statistics letter

Technical field

The invention belongs to wireless communication technology field, relate to the multiple-input and multiple-output MIMO communication technology, be used to realize the multiuser mobile communication system of high speed data transfer, be specially a kind of the leakage especially and make an uproar than the MIMO multiuser downstream transmission method of criterion based on maximization statistics letter.

Background technology

Compare with 3-G (Generation Three mobile communication system), the future mobile communication system requirement can be supported the high-speed data service speed up to per second hundreds of million even 1G bit, under situation about being becoming tight Radio Resource day, adopt the new network framework, utilize MIMO multi-user radio transmission technology, fully the excavated space resource improves the availability of frequency spectrum and power efficiency to greatest extent, becomes the key point of future mobile communications research.

The MIMO transmission technology that adopts many antenna transmission and many antennas to receive is digging utilization Spatial Dimension Radio Resource, the Basic Ways that improves the availability of frequency spectrum and power efficiency, be the main research point of moving communicating field over year surplus in the of ten always, and become the core technology of future mobile communication system.The main characteristic of MIMO technology is all to adopt many antenna elements to finish the transmission and the reception of information at transmitting terminal and receiving terminal, this technology is directly applied to multi-user system, can cause comparatively serious inter-user interference, thereby greatly restrict the raising of system spectrum utilance.Theoretical research shows: adopt MIMO multiuser downstream transmission technology can effectively utilize space resources, break through the restriction of traditional mobile cellular transmission technology to the availability of frequency spectrum, significantly the power efficiency of elevator system.At present, this technology has caused extensive concern, also becomes simultaneously three main standard tissues and realizes that its technology is to one of mainstream technology of IMT-Advanced standard evolution.The research of MIMO multiuser downstream transmission technology in recent years mainly concentrates on the situation of the known instantaneous channel condition information in base station, these researchs are only applicable to fixing and low speed mobile communication scene, MIMO multiuser downstream transmission technology research at the high speed mobile communication environment is less relatively, and the rarely seen report of multiuser downstream transmission method of comprehensive utilization channel statistical state information.On the other hand, even in low speed mobile communication scene, under based on feedback link capacity confined condition, the descending multi-user transmission is comparatively responsive to the precision of the channel parameter of feedback, utilizing the channel statistical state information to carry out the multiuser downstream transmission and whether can obtain close or more performance, is the focal issue that merits attention too.

Summary of the invention

The problem to be solved in the present invention is: existing research to the transmission of MIMO multiuser downstream is only applicable to fixing and low speed mobile communication scene, MIMO multiuser downstream transmission technology research at the high speed mobile communication environment is less relatively, and how utilizing the channel statistical state information to carry out the transmission of MIMO multiuser downstream is the problem to be solved in the present invention.

Technical scheme of the present invention is: leak based on the statistics letter and make an uproar than the MIMO multiuser downstream transmission method of criterion, each user who participates in downlink transfer is connected with the base station by Radio Link, the antenna number of base station is Nt, each user's antenna number is 1, the base station utilizes the channel statistical state information between itself and each user to carry out the precoding design, may further comprise the steps:

1) base station obtains the relevant battle array of transmission and the characteristic value diagonal matrix thereof of each subscriber channel in base station to the sub-district by the channel statistical state information, the channel matrix H between base station and j the user _jBe 1 * N _tMatrix:

Then the relevant battle array of the transmission of user j is:

Wherein T is N _t* N _tThe discrete Fourier matrix of dimension,

Be H _jThe virtual representation form,

For sending the characteristic value diagonal matrix of relevant battle array;

2) base station utilize obtain in the step 1) with the user between the relevant battle array of transmission, making an uproar than SLNR lower bound with the maximization statistics letter leakage of user j is target, calculate the precoding vector of MIMO multiuser downstream transmission, i.e. the feature transmission direction of user j is by the discrete Fourier matrix

Column vector constitute;

3) scheduling is fit to the user of MIMO multiuser downstream transmission, and concrete steps are as follows:

3.1) the feature transmission direction of calculating each user;

3.2) user that the feature transmission direction is intercoupled is divided in the same transmission group, carries out the transmission of MIMO multiuser downstream;

3.3) be no more than antenna for base station at scheduled user's sum and count under the prerequisite of Nt repeating step 3.1) and step 3.2), travel through all users, dispatch out the user who is suitable for the transmission of MIMO multiuser downstream;

4) base station is carried out power division according to sending relevant battle array to sending to each user's data stream;

5) transmission of MIMO multiuser downstream is finished according to feature transmission direction and power distribution method in the base station.

In the step 1), obtaining of channel statistical state information is divided into feedback and two kinds of patterns of latent feedback: when adopting feedback model, the base station obtains the channel statistical state information by user's feedback; Adopt when concealing feedback model, directly in base station calculating channel statistic behavior information.

Further, step 2) be specially:

Total number of users is K, establish user j (signal indication of K) receiving is for j=1, L:

$r_j=\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(3\right)$

N wherein _jBe to be zero by average, variance is j user's constituting of 1 independent identically distributed multiple gaussian variable an additive Gaussian noise vector, s _kFor the base station sends to k (k=1, L, K) individual user's power normalization symbolic information, w _kBe the N of base station to k user _tThe transmission beam forming vector of * 1 dimension and having

p _kFor the base station sends to k user's power, the signal that user j is received is equivalently represented to be:

$r_j=\sqrt{p_j}{H}_j{w}_j{s}_j+\underset{k=1,k\≠j}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(4\right)$

Send to the symbol s of user j for the base station _j, the effective information power that user j receives is Other user k receives the equivalent power of this information, and promptly leakage power is:

Wherein

The letter leakage that obtains user j is thus made an uproar as follows than SLNR:

Utilize the statistics letter to leak to make an uproar and compare E{SLNR _jTo w _jBe optimized design, maximization statistics letter leak make an uproar than target under the transmission beam forming vector of each base station of design:

Molecule in the formula (6) and denominator are to add up independently, according to the Mullen inequality, obtain adding up letter leak make an uproar than lower bound:

With the expectation computing expansion of molecule and denominator in (7) formula and to H _kAsk expectation to get:

(8) formula will maximize statistics letter and leak and make an uproar than being converted into maximization lower bound problem, in the formula

A _jT and

Be the Hermite matrix, can get w by the Rayleigh-Ritz theorem _jAs follows:

The characteristic vector of eigenvalue of maximum is asked in Q in the following formula (g) expression, will With T from

Extract and do further simplification promptly:

Diagonal matrix wherein

w _jBe

In corresponding to the column vector of Ω diagonal greatest member, i.e. the feature transmission direction of user j.

Step 4) is specially: according to the result of downlink multi-user scheduling, the average transmission power that calculates S the user characteristics transmission direction of having dispatched is as follows:

$\stackrel{\‾}{\mathrm{\α}}=\frac{1}{S}\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{\mathrm{\α}}_i---\left(12\right)$

Wherein

Each user's feature transmission direction power among S the user who refers to dispatch, the gross power of establishing the base station is P, the user relatively poor according to channel quality distributes the more multiple principle of sending power, the distribution power p of user i _iFor:

$p_i=\frac{P(\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i)}{\underset{i=1}{\overset{S}{\mathrm{\Σ}}}\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i}---\left(13\right).$

The invention provides in a kind of MIMO transmission system utilize maximization statistics letter to leak to make an uproar than downlink transmission method, this method can send parameter according to the characteristic adjustment of channel statistical state information, the system and the speed of the more efficient that obtains with less overhead.

Beneficial effect: the research of existing MIMO multiuser downstream transmission technology mainly concentrates on the situation of the known instantaneous channel condition information in base station, these researchs are only applicable to fixing and low speed mobile communication scene, the present invention is based on the channel statistical state information, be suitable for the MIMO multiuser downstream transmission technology of high-speed mobile communications environment.The invention provides a kind of utilize maximization statistics letter to leak to make an uproar than MIMO multiuser downstream pre-coding transmission method, this method has following advantage:

1, this method has been utilized the channel statistical state information, and the channel statistical state information is the relevant battle array of the transmission of channel, obtains from actual channel easilier;

2, the base station utilizes and sends relevant battle array in this method, dispatches out space isolation user preferably, implements the transmission of MIMO multiuser downstream;

3, the base station only needs the relevant battle array of transmission of channel between base station and the user in this method, and the mutual overhead of base station and each user-to-user information is very low;

4, in this method, the user relatively poor to channel quality distributes the more multiple power that send, and guarantees each telex network performance,

The power allocation scheme of customer traffic is finished by the base station, and only utilizes the relevant battle array of transmission, and computation complexity is low.

The present invention utilizes the statistical channel state to carry out MIMO multiuser downstream transmission design, main characteristic is to utilize the channel statistical state information to carry out each user's feature transmission direction calculating, multi-subscriber dispatching, and power distribution design, owing to only utilize the statistical channel state information at the base station end, can significantly reduce the base station and carry out the required overhead of downlink precoding transmission, simultaneously owing to utilize the space separability of statistical information, the higher user of space isolation is implemented the transmission of MIMO multiuser downstream, therefore can improve the availability of frequency spectrum and the power efficiency of communication system to a great extent, obtain the higher system transmission rate.

Description of drawings

Fig. 1 is a specific embodiment of the present invention, is used to support that MIMO multi-user makes the wireless communication system schematic diagram of downlink transfer, and it is by the base station, and several users constitute, and with 1 base station, 6 users are example among the figure.

Fig. 2 is the flow chart of a specific embodiment of the present invention.

Embodiment

Among the present invention, the base station utilizes channel estimation results to calculate the relevant battle array of transmission of channel matrix by the mode of feedback or latent feedback.The base station is according to sending relevant battle array and dispatching out each user who is fit to downlink transfer.Corresponding different user, maximization statistics letter leak make an uproar than principle under, utilize to send the transmission direction of relevant battle array calculated characteristics, meanwhile, the base station is carried out power division according to sending relevant battle array to sending to each user's data stream, finishes MIMO multiuser downstream transmission method on this basis.Mainly may further comprise the steps:

1) base station obtains the relevant battle array of transmission of each subscriber channel to the sub-district;

2) base station is according to sending the feature transmission direction that relevant battle array calculates downlink transfer;

3) user who is fit to the transmission of MIMO multiuser downstream is dispatched out according to sending relevant battle array in the base station;

4) power of each customer traffic is determined to send to according to sending relevant battle array in the base station;

5) transmission of MIMO multiuser downstream is finished according to feature transmission direction and power distribution method in the base station.

Consider a base station services in a plurality of users' scene, the antenna for base station number is N _t, the user antenna number is 1, each user links to each other with the base station by Radio Link, supposes each with receiving from signal of base station per family, and can estimate the channel information of each user to the base station.

For making the technical scheme among the present invention clearer, below this programme is specifically described:

One, sends obtaining of relevant battle array

Under feedback model, the relevant battle array of channel transmission can be calculated according to channel estimation results by the user in the described scheme.We use N _r* N _tMatrix H _jChannel matrix between expression base station and j the user, channel matrix H _jBe modeled as:

Wherein T is N _t* N _tThe discrete Fourier matrix of dimension, Be H _jThe virtual representation form,

Component be independent non-with distributing, each element representative be corresponding to every pair of virtual transmitting-receiving angle.Utilize the estimated value of channel parameter, it is as follows to calculate the relevant battle array of transmission:

Wherein Refer to the characteristic value diagonal matrix that sends relevant battle array.Observe (2) formula, can with Regard the special battle array of an ell rice as, A _jIt is the diagonal matrix of the characteristic value formation of non-ordering.When employing concealed feedback model, the base station utilized up reception link

Channel estimation results and the reciprocity of channel

Adopt with feedback model under identical method calculate and send relevant gust and characteristic value diagonal matrix.

Two, the calculating of feature transmission direction

Total number of users is K, establish user j (signal indication of K) receiving is for j=1, L:

$r_j=\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(3\right)$

N wherein _jBe to be zero by average, variance is j user's constituting of 1 independent identically distributed multiple gaussian variable an additive Gaussian noise vector, s _kFor the base station sends to k (k=1, L, K) individual user's power normalization symbolic information, w _kBe the N of base station to k user _tThe transmission beam forming vector of * 1 dimension and having

p _kFor the base station sends to k user's power, the signal that user j is received is equivalently represented to be:

$r_j=\sqrt{p_j}{H}_j{w}_j{s}_j+\underset{k=1,k\≠j}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(4\right)$

Send to the symbol s of user j for the base station _j, the effective information power that user j receives is

Other user k receives the equivalent power of this information, and promptly leakage power is:

Wherein

The letter leakage that obtains user j is thus made an uproar as follows than SLNR:

The present invention utilizes statistics letter to leak to make an uproar to compare E{SLNR _jTo w _jBe optimized design, maximization statistics letter leak make an uproar than target under the transmission beam forming vector of each base station of design, this moment, problem can be described as:

Molecule in the formula (6) and denominator are to add up independently, according to the Mullen inequality, obtain adding up letter leak make an uproar than lower bound:

With the expectation computing expansion of molecule and denominator in (7) formula and to H _kAsk expectation to get:

(8) formula will maximize statistics letter and leak and make an uproar than being converted into maximization lower bound problem.Because in (8) formula

A _jT and

Be the Hermite matrix, can get w by the Rayleigh-Ritz theorem _jAs follows:

The Q of following formula (g) refers to ask for the characteristic vector of eigenvalue of maximum, will

With T from

Extract and do further simplification promptly:

Diagonal matrix wherein

w _jBe

In corresponding to the column vector of Ω diagonal greatest member, i.e. the feature transmission direction of user j.。

Three, MIMO multiuser downstream dispatching algorithm

(8) letter of the statistics in the formula leaks to make an uproar and can be expressed as than lower bound:

(11) formula shows that it is exactly according to the statistical information dispatched users that the leakage of maximization statistics letter is made an uproar than lower bound, thus the eigenvalue of maximum of maximization diagonal matrix Ω.At this, the present invention proposes a kind of dispatching method that utilizes channel to send relevant battle array fully, and details are as follows:

The channel that the base station at first obtains each user feedback sends relevant gust characteristic value diagonal matrix A _j, then according to A _jThe size distribution of diagonal entry find out the user who is fit to downlink transfer, concrete steps are as follows:

1) observes A _jDiagonal entry, greatest member characteristic of correspondence value vector is decided to be the feature transmission direction.In the following Example, by A ₁, A ₂, A ₃, A ₄Selecting 7.3,11.2,12.8,9.5 characteristics of correspondence vector respectively is the feature transmission direction;

2) user that the feature transmission direction is intercoupled, the nonoverlapping user of feature transmission direction just forms one group, carries out the multiuser downstream transmission.In the following example, A ₁, A ₂And A ₄Can be divided in and carry out multiuser downstream transmission, A in same group ₂, A ₃And A ₄Also can be divided in and carry out the multiuser downstream transmission in same group.

3) sum the scheduled user is no more than N _tPrerequisite under, repeating step 1) and step 2), the traversal all

The user dispatches out the user who is suitable for the transmission of MIMO multiuser downstream;

Four, power distribution method

The characteristic value diagonal matrix that the channel of feature transmission direction correspondence sends relevant battle array has determined propagation channel quality between them, in order to improve systematic function, the characteristic value diagonal matrix that the present invention proposes to utilize channel to send relevant battle array carries out power division to the base station to each user's data stream, and details are as follows for method:

According to the result of downlink multi-user scheduling, the average transmission power that calculates S the user characteristics transmission direction of having dispatched is as follows:

$\stackrel{\‾}{\mathrm{\α}}=\frac{1}{S}\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{\mathrm{\α}}_i---\left(12\right)$

Wherein Each user's feature transmission direction power among S the user who refers to dispatch, the gross power of establishing the base station is P, the user relatively poor according to channel quality distributes the more multiple principle of sending power, the distribution power p of user i _iFor:

$p_i=\frac{P(\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i)}{\underset{i=1}{\overset{S}{\mathrm{\Σ}}}\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i}---\left(13\right).$

Claims (5)

1. leak based on the statistics letter and make an uproar than the MIMO multiuser downstream transmission method of criterion, it is characterized in that each user who participates in downlink transfer is connected with the base station by Radio Link, the antenna number of base station is N _t, each user's antenna number is 1, the base station utilizes the channel statistical state information between itself and each user to carry out the precoding design, may further comprise the steps:

1) base station obtains the relevant battle array of transmission and the characteristic value diagonal matrix thereof of each subscriber channel in base station to the sub-district by the channel statistical state information, the channel matrix H between base station and j the user _jBe 1 * N _tMatrix:

Then the relevant battle array of the transmission of user j is:

Wherein T is N _t* N _tThe discrete Fourier matrix of dimension,

Be H _jThe virtual representation form,

For sending the characteristic value diagonal matrix of relevant battle array;

2) base station utilize obtain in the step 1) with the user between the relevant battle array of transmission, making an uproar than SLNR lower bound with the maximization statistics letter leakage of user j is target, calculate the precoding vector of MIMO multiuser downstream transmission, i.e. the feature transmission direction of user j is by the discrete Fourier matrix Column vector constitute;

3) scheduling is fit to the user of MIMO multiuser downstream transmission, and concrete steps are as follows:

3.1) the feature transmission direction of calculating each user;

3.2) user that the feature transmission direction is intercoupled is divided in the same transmission group, carries out the transmission of MIMO multiuser downstream;

3.3) be no more than antenna for base station at scheduled user's sum and count N _tPrerequisite under, repeating step 3.1) and step 3.2), travel through all users, dispatch out the user who is suitable for MIMO multiuser downstream transmission;

4) base station is carried out power division according to sending relevant battle array to sending to each user's data stream;

5) transmission of MIMO multiuser downstream is finished according to feature transmission direction and power distribution method in the base station.

2. according to claim 1 the leakage based on the statistics letter made an uproar than the MIMO multiuser downstream transmission method of criterion, it is characterized in that in the step 1), obtaining of channel statistical state information is divided into feedback and two kinds of patterns of latent feedback: when adopting feedback model, the base station obtains the channel statistical state information by user's feedback; Adopt when concealing feedback model, directly in base station calculating channel statistic behavior information.

3. according to claim 1 and 2 the leakage based on the statistics letter made an uproar than the MIMO multiuser downstream transmission method of criterion, it is characterized in that step 2) be specially:

Total number of users is K, establish user j (signal indication of K) receiving is for j=1, L:

$r_j=\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(3\right)$

N wherein _jBe to be zero by average, variance is j user's constituting of 1 independent identically distributed multiple gaussian variable an additive Gaussian noise vector, s _kFor the base station sends to k (k=1, L, K) individual user's power normalization symbolic information, w _kBe the N of base station to k user _tThe transmission beam forming vector of * 1 dimension and having

p _kFor the base station sends to k user's power, the signal that user j is received is equivalently represented to be:

$r_j=\sqrt{p_j}{H}_j{w}_j{s}_j+\underset{k=1,k\≠j}{\overset{K}{\mathrm{\Σ}}}\sqrt{p_k}{H}_j{w}_k{s}_k+n_j---\left(4\right)$

Send to the symbol s of user j for the base station _j, the effective information power that user j receives is

Other user k receives the equivalent power of this information, and promptly leakage power is:

Wherein

The letter leakage that obtains user j is thus made an uproar as follows than SLNR:

Utilize the statistics letter to leak to make an uproar and compare E{SLNR _jTo w _jBe optimized design, maximization statistics letter leak make an uproar than target under the transmission beam forming vector of each base station of design:

Molecule in the formula (6) and denominator are to add up independently, according to the Mullen inequality, obtain adding up letter leak make an uproar than lower bound:

With the expectation computing expansion of molecule and denominator in (7) formula and to H _kAsk expectation to get:

(8) formula will maximize statistics letter and leak and make an uproar than being converted into maximization lower bound problem, in the formula

A _jT and

Be the Hermite matrix, can get w by the Rayleigh-Ritz theorem _jAs follows:

The characteristic vector of eigenvalue of maximum is asked in Q in the following formula (g) expression, will

With T from

Extract and do further simplification promptly:

Diagonal matrix wherein

w _jBe

In corresponding to the column vector of Ω diagonal greatest member, be the feature transmission direction of user j.

4. according to claim 1 and 2 the leakage based on the statistics letter made an uproar than the MIMO multiuser downstream transmission method of criterion, it is characterized in that step 4) is specially: according to the result of downlink multi-user scheduling, the average transmission power that calculates S the user characteristics transmission direction of having dispatched is as follows:

$\stackrel{\‾}{\mathrm{\α}}=\frac{1}{S}\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{\mathrm{\α}}_i---\left(12\right)$

Wherein

Each user's feature transmission direction power among S the user who refers to dispatch, the gross power of establishing the base station is P, the user relatively poor according to channel quality distributes the more multiple principle of sending power, the distribution power p of user i _jFor:

$p_i=\frac{P(\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i)}{\underset{i=1}{\overset{S}{\mathrm{\Σ}}}\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i}---\left(13\right).$

5. according to claim 3 the leakage based on the statistics letter made an uproar than the MIMO multiuser downstream transmission method of criterion, it is characterized in that step 4) is specially: according to the result of downlink multi-user scheduling, the average transmission power that calculates S the user characteristics transmission direction of having dispatched is as follows:

$\stackrel{\‾}{\mathrm{\α}}=\frac{1}{S}\underset{i=1}{\overset{S}{\mathrm{\Σ}}}{\mathrm{\α}}_i---\left(12\right)$

α wherein _i(i=1, L, S) each user's feature transmission direction power among S the user who refers to dispatch, the gross power of establishing the base station is P, the user relatively poor according to channel quality distributes the more multiple principle of sending power, the distribution power p of user i _iFor:

$p_i=\frac{P(\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i)}{\underset{i=1}{\overset{S}{\mathrm{\Σ}}}\stackrel{\‾}{\mathrm{\α}}/{\mathrm{\α}}_i}---\left(13\right).$

Images