CN102035625A - Method for optimizing precoder of communication system, base station and terminal - Google Patents

Abstract

The invention provides a method for optimizing a precoder in a communication system. The communication system comprises a terminal and a base station for implementing the precoder. The method comprises the following steps of: sending a reference signal from the base station to the terminal; acquiring channel transmission matrixes on different time frequency resources through the reference signal; calculating a sending end covariance matrix of a channel and a receiving end covariance matrix for different preset pre-coding vectors according to the channel transmission matrixes; feeding a base station end covariance matrix and the receiving end covariance matrix for different preset pre-coding vectors to the base station; and optimizing the precoder based on the base station end covariance matrix and the receiving end covariance matrix for different preset pre-coding vectors so as to form a pre-coding vector. According to the precoder, average throughput of the communication system is considered, the practical channel relativity of a receiving terminal is considered and the performance of the entire communication system is enhanced.

Description

Optimize method, base station and the terminal of the precoder of communication system

Technical field

The present invention relates to multi-antenna signal transmission and beam-forming technology in the communications field.

Background technology

Utilizing multi-antenna signal transmission (MIMO) and wave beam to form in the communication system of (Beamforming), the precoder extensive use in the standard of wireless communication system based on the channel prompting message that cell base station uses also begins to come into question based on the precoder of channel statistical characteristic.Precoder based on statistical property has two advantages, and the firstth, the time delay of background information exchange between the time delay of terminal to base station feedback and the base station is had robustness preferably, the secondth, the required Radio Resource expense of the feedback of terminal to base station is lower.

The optimization criterion of the most frequently used precoder based on statistical property is the maximization average throughput.When the order of transmission was one (only transmitting a data flow), the maximization average throughput was equal to maximization average signal received power, and its mathematic(al) representation is:

$\underset{\stackrel{\→}{w}}{\arg \max }{\stackrel{\→}{w}}^H{R}_T\stackrel{\→}{w}---\left(1\right)$

The implication of following formula is: seek Value, make expression formula

The value maximum.Herein, Be precoding vectors, (.) ^HRepresent conjugate transpose, R _TBe transmitting terminal (cell base station) covariance matrix of channel, it is defined as:

E[. herein] to represent mathematic expectaion, H then be the Channel Transmission matrix.Therefore, when the maximization average throughput, the transmitting terminal covariance matrix need be fed back to this transmitting terminal by the receiving terminal of signal.

Here, suppose that transmitting terminal has four transmitting antennas, receiving terminal has two reception antennas, and transmitting terminal sends reference signal (for example being sub-district special reference signal or customer special reference signal) at a certain specific running time-frequency resource to receiving terminal.First reception antenna of receiving terminal receives this reference signal from first transmitting antenna of transmitting terminal, can calculate channel parameter h1, and second reception antenna of receiving terminal receives reference signal from first transmitting antenna of transmitting terminal, can calculate channel parameter h2.Equally, at another running time-frequency resource, first and second reception antennas of receiving terminal receive reference signal from second transmitting antenna of transmitting terminal respectively, can calculate channel parameter h3 and h4 respectively, at the 3rd running time-frequency resource, first and second reception antennas of receiving terminal receive reference signal from the 3rd transmitting antenna of transmitting terminal respectively, can calculate channel parameter h5 and h6 respectively, at the 4th running time-frequency resource, first and second reception antennas of receiving terminal receive reference signal from the 4th transmitting antenna of transmitting terminal respectively, can calculate channel parameter h7 and h8 respectively.

Like this, the antenna by receiving terminal receives reference signal from the transmitting antenna of transmitting terminal, can obtain the Channel Transmission matrix H 1 that this transmission system is carved at a time, and it can be expressed as:

$H1=\left[\begin{array}{cccc}h1& h3& h5& h7\\ h2& h4& h6& h8\end{array}\right]---\left(3\right)$

The transmitting terminal of channel (cell base station) covariance matrix R _TTherefore utilize channel statistic property, also need to obtain constantly and/or the Channel Transmission matrix of different frequency bands in difference.Here suppose at moment t1, t2, t3 ..., tn, the Channel Transmission matrix that obtains from different transmitting antennas (with different frequency bands) transmitted reference signal are respectively H1, H2, H3, ..., Hn (the Channel Transmission matrixes on different running time-frequency resources), then the transmitting terminal covariance matrix R of channel _TFor example can be expressed as:

R _T＝(H ₁ ^HH ₁+H ₂ ^HH ₂+H ₃ ^HH ₃...+H _n ^HH _n)/n (4)

Thereby can obtain transmitting terminal covariance matrix R _T, and and then the precoding vectors when obtaining the maximization average throughput

Come the precoder in the design of communications system thus.

But average throughput is not performance index unique when designing precoder.

Fig. 1 shows the relation between average signal received power and the actual signal received power.As shown in Figure 1, although the average signal received power is on the received power threshold, if the signal fluctuation amplitude is bigger, some code word (subframe) may be among the deep fading, promptly be in below the signal received power threshold, thereby make the actual performance of communication system descend.

Summary of the invention

The present invention proposes precoder based on channel statistical information.Be different from the precoder that maximizes average throughput in the prior art based on statistical information, the precoder that the present invention proposes is not only considered average throughput, also consider the amplitude fluctuation size of signal received power, promptly the actual channel correlation of receiving terminal has been made the present invention thus.

According to an aspect of the present invention, provide the method for the precoder that is used for optimizing communication system, described communication system comprises terminal and the base station of realizing described precoder, and described method comprises: send reference signal from described base station to described terminal; Obtain Channel Transmission matrix on the different running time-frequency resources by described reference signal; According to the transmitting terminal covariance matrix of described Channel Transmission matrix computations channel and receiving terminal covariance matrix at different default precoding vectors; To the described base station of described base station feedback end covariance matrix and receiving terminal covariance matrix at different default precoding vectors; And based on described base station end covariance matrix with optimize described precoder at the receiving terminal covariance matrix of different default precoding vectors, to form precoding vectors.

According to another aspect of the present invention, provide the base station that is used for communication system, described base station comprises: transmitting element sends reference signal to terminal; Receiving element receives based on described reference signal base station end covariance matrix that obtain, channel and receiving terminal covariance matrix corresponding to different default precoding vectors from described terminal; And precoder, based on described base station end covariance matrix and receiving terminal covariance matrix, form precoding vectors corresponding to different default precoding vectors.

According to a further aspect of the invention, provide the terminal that is used for communication system, described terminal comprises: receiving element receives reference signal from the base station; Computing unit obtains Channel Transmission matrix on the different running time-frequency resources by described reference signal, according to the transmitting terminal covariance matrix of described Channel Transmission matrix computations channel and receiving terminal covariance matrix at different default precoding vectors; And transmitting element, to the described base station of described base station feedback end covariance matrix and described receiving terminal covariance matrix at different default precoding vectors.

Precoder of the present invention has not only been considered the average throughput of communication system, has also considered the amplitude fluctuation size of signal received power, i.e. the actual channel correlation of receiving terminal has improved the performance of whole communication system thus.

Description of drawings

From below in conjunction with the detailed description of accompanying drawing to embodiments of the invention, these and/or others of the present invention and advantage will become clear more and be more readily understood, wherein:

Fig. 1 illustrates the diagram that concerns between average signal received power and the actual signal received power.

Fig. 2 is the diagram that the simple communication system of being made up of cell base station and receiving terminal is shown.

Fig. 3 is the diagram that the basic structure of the cell base station among Fig. 2 is shown.

Fig. 4 is the diagram that the basic structure of the receiving terminal among Fig. 2 is shown.

Fig. 5 is the schematic diagram that the characteristic of channel that cell base station in the communication system of Fig. 2 transmits to receiving terminal is shown.

Fig. 6 illustrates the diagram of optimizing the result of precoder according to the channel relevancy of receiving terminal.

Fig. 7 is another schematic diagram that the characteristic of channel that cell base station in the communication system of Fig. 2 transmits to receiving terminal is shown.

Fig. 8 is the flow chart of method that is used for optimizing the precoder of communication system according to the present invention.

Fig. 9 illustrates two cell base stations to form the diagram of coming transmitting-receiving data flow by the collaboration type wave beam.

Embodiment

Describe specific embodiments of the invention in detail below in conjunction with accompanying drawing.If consider and to obscure main points of the present invention, then can not provide its detailed description here to the detailed description of some related art.In each embodiment, identical Reference numeral is used to represent to carry out the element or the unit of identical function.

First embodiment

The first embodiment of the present invention is that example describes with the single subdistrict in the communication system.

Fig. 2 is the diagram that the simple communication system of being made up of cell base station and receiving terminal is shown.

As shown in Figure 2, communication system 200 comprises cell base station 201 and receiving terminal 202.Cell base station 201 has a plurality of (for example four) transmitting antenna, receiving terminal 202 has one or more (for example two) reception antenna, wherein send for example proprietary reference signal in sub-district to receiving terminal 202 by cell base station 201, reception forms precoding vectors from the channel information that calculates according to above-mentioned reference signal of receiving terminal 202 feedback (for example be the base station end covariance matrix of channel with at different default precoding vectors receiving terminal covariance matrix) according to the channel information that receives

, and utilize precoding vectors

Send data flow to receiving terminal 202.Here, the order of supposing the transmission of communication system is one, is a signal or a data flow from cell base station 201 to the signal or the data flow of receiving terminal 202 transmission promptly.The number of the reception antenna that the number of the transmitting antenna that cell base station 201 has and receiving terminal 202 have is not construed as limiting scope of the present invention, and they can be the actual any numbers that is disposed in the communication system 200.

Even in the data of as shown in Figure 2 single subdistrict sent, as previously mentioned, average throughput neither design unique performance index of precoder.If the fluctuating range of signal received power is bigger, some code word (subframe) may be among the deep fading, promptly falls within below the signal received power threshold, makes the actual performance of this communication system descend.Therefore, the fluctuating range of signal received power equally also needs to be considered.

An important indicator of the fluctuating range of gauge signal received power is the channel relevancy of receiving terminal (for example receiving terminal 202).Channel relevancy is high more, and the possibility that the fluctuating range of expression signal received power is big is big more.

Fig. 3 is the diagram that the basic structure of the cell base station among Fig. 2 is shown.

Shown in Figure 3 is is used for optimizing the cell base station 201 of the precoder of communication system, and precoder can be realized that cell base station 201 comprises by this base station: precoder 301, transmitting element 302, receiving element 303, and transmitting antenna 304.Precoder 301, transmitting element 302, receiving element 303 and transmitting antenna 304 etc. are connected by internal bus 305 or other necessary cable of cell base station 201.Can also comprise in the structure of cell base station 201 for needed other unit of the operation of base station or parts, but other the included unit or the concrete structure of parts are not construed as limiting scope of the present invention.Precoder 301 can be implemented in the control unit of CPU etc. for example, also can be embodied as same unit with transmitting element 302 and receiving element 303.

Transmitting element 302 can send various signals or data flow to the receiving terminal 202 of this sub-district, for example sends the proprietary reference signal in sub-district.Receiving element 303 can receive various signals and data flow from receiving terminal 202, for example according to base station end covariance matrix that above-mentioned reference signal obtained and receiving terminal covariance matrix at different default precoding vectors.In precoder 301,, form precoding vectors according to base station end covariance matrix that is received and described receiving terminal covariance matrix.

More specifically, the base station 201 that is used for communication system according to the present invention comprises: transmitting element 302, be used for sending reference signal to receiving terminal 202, and for example be the proprietary reference signal in sub-district; Receiving element 302 is used for receiving based on above-mentioned reference signal base station end covariance matrix that obtain, channel and receiving terminal covariance matrix corresponding to different default precoding vectors from receiving terminal 202; And precoder 301, based on base station end covariance matrix that is obtained and described receiving terminal covariance matrix, form precoding vectors.

Fig. 4 is the diagram that the basic structure of the receiving terminal among Fig. 2 is shown.

Shown in Figure 4 is the receiving terminal 202 that is used for communication system, and this receiving terminal 202 comprises: receiving element 401, transmitting element 402, computing unit 403, reception antenna 404.Receiving element 401, transmitting element 402, computing unit 403 and reception antenna 404 etc. are connected by internal bus 405 or other necessary cable of receiving terminal 202.Can also comprise in the structure of receiving terminal 202 for needed other unit of the operation of receiving terminal 202 or parts, but other the included unit or the concrete structure of parts are not construed as limiting scope of the present invention.For example, can comprise control unit control receiving element 401, transmitting element 402 and computing unit 403, that be implemented as CPU etc. in the receiving terminal 202.Computing unit 403, transmitting element 402, receiving element 401 may be implemented as same unit.

Receiving element 401 receives various signals or the data flow that sends from cell base station 201, for example reference signal.Computing unit 403 is according to the reference signal that is received, the base station end covariance matrix of calculating channel and receiving terminal covariance matrix corresponding to different default precoding vectors, this receiving terminal covariance matrix has the value of the channel relevancy of representing receiving terminal 202.Base station end covariance matrix that transmitting element 402 is obtained to cell base station 201 feedback and receiving terminal covariance matrix corresponding to different default precoding vectors.

More specifically, the receiving terminal 202 that is used for communication system according to the present invention comprises: receiving element 401, be used for receiving reference signal from cell base station 201, and for example be the proprietary reference signal in sub-district; Computing unit 403 is used for obtaining Channel Transmission matrix on the different running time-frequency resources by above-mentioned reference signal, with the transmitting terminal covariance matrix of calculating channel and receiving terminal covariance matrix at different default precoding vectors; And transmitting element 402, be used for to the above-mentioned base station of cell base station 201 feedbacks end covariance matrix and above-mentioned receiving terminal covariance matrix at different default precoding vectors.

The value of the channel relevancy of the representative receiving terminal 202 that above-mentioned receiving terminal covariance matrix has can be the characteristic value of receiving terminal covariance matrix.

Owing to the acquisition mode of base station end covariance matrix has been discussed in front, the computational methods of base station end covariance matrix and receiving terminal covariance matrix will specifically be discussed below.

Be under one the situation, can see that from following expression the actual channel correlation of receiving terminal is not only relevant with the statistical property of actual channel, and is also relevant with the precoding vectors that is adopted in transmission rank.Also promptly, the design of precoding vectors is relevant with the actual channel correlation of receiving terminal.

$\stackrel{\→}{y}=H\stackrel{\→}{w}x+\stackrel{\→}{n}=\stackrel{\→}{h}x+\stackrel{\→}{n}---\left(5\right)$

Here,

The signal that expression receives; H has represented to reflect the Channel Transmission matrix of channel statistical characteristic, can represent for example characteristics such as decline of channel;

The expression precoding vectors; X represents the signal that transmits;

The expression noise;

Expression actual channel characteristic.Here, owing to will utilize channel statistic property, therefore need to obtain to send with different frequency bands constantly the Channel Transmission matrix of reference signal in difference.Here suppose at moment t1, t2, t3 ..., tn, be respectively H1 from different transmitting antennas (with different frequency bands) the Channel Transmission matrix that transmitted reference signal was obtained, H2, H3 ..., Hn.The Channel Transmission matrix H that has reflected the channel statistical characteristic can be the set of the above-mentioned Channel Transmission matrix that obtains on different running time-frequency resources, shown in formula (4).

The fluctuating range of signal received power depends on actual channel

The receiving terminal covariance matrix, it can be used for representing the channel relevancy of receiving terminal.Compare with formula (2), the receiving terminal covariance matrix can be expressed as:

$R=E\left[\stackrel{\→}{h}{\stackrel{\→}{h}}^H\right]=E\left[H\stackrel{\→}{w}{\stackrel{\→}{w}}^H{H}^H\right]---\left(6\right)$

Show that as following formula this receiving terminal covariance matrix (expression channel relevancy) was both relevant with the statistical property of Channel Transmission matrix H, also with precoding vectors Relevant, and different precoding vectors is corresponding to different receiving terminal covariance matrixes.Therefore, when the optimization precoder, not only on the basis of transmitting terminal covariance matrix, consider the average throughput of communication system, also need to consider at a certain given precoding vectors

The channel relevancy of receiving terminal, promptly above-mentioned actual channel

The receiving terminal covariance matrix.According to an aspect of the present invention, above-mentioned a certain given precoding vectors

Can obtain by a plurality of different default precoding vectors.

Thus, the present invention proposes the precoder of the channel relevancy of considering receiving terminal.Precoder according to the present invention is optimised under the prerequisite of channel relevancy less than certain predetermined threshold of receiving terminal, the maximization communication system average throughput, its expression formula can for:

$\underset{\stackrel{\&RightArrow;}{w}}{\arg \max }{\stackrel{\&RightArrow;}{w}}^H{R}_T\stackrel{\&RightArrow;}{w}$ Condition: $\frac{d_{\max }}{d_{\min }}<\mathrm{\&alpha;},---\left(7\right)$

Here,

Be a certain given precoding vectors, (.) ^HRepresent conjugate transpose, R _TFor the base station end covariance matrix of channel, d _MaxWith d _MinBe respectively eigenvalue of maximum and the minimal eigenvalue of (defining in the formula (6)) receiving terminal covariance matrix R, α is the default above-mentioned predetermined threshold of cell base station.

The precoder of design according to the present invention need obtain channel relevancy at the receiving terminal (for example being receiving terminal 202) of any above-mentioned a certain given precoding vectors at base station end (for example being cell base station 201).For obtaining the receiving terminal channel relevancy of any above-mentioned a certain given precoding vectors in the base station, not only will be from receiving terminal 202 to the base station of cell base station 201 feedback channels end covariance matrix R _T, also need feed back corresponding receiving terminal covariance matrix R at the precoding vectors of a plurality of (for example being Nt) mutually orthogonal.

Fig. 5 shows the schematic diagram of the characteristic of channel that cell base station in the communication system of Fig. 2 transmits to receiving terminal.

As shown in Figure 5, cell base station 201 has four transmitting antennas, and receiving terminal 202 has two reception antennas, supposes that here cell base station 202 sends the proprietary reference signal in sub-district to receiving terminal 202.

In Fig. 5, suppose

Corresponding to first default precoding vectors of cell base station 201, Corresponding to second default precoding vectors,

Corresponding to the 3rd default precoding vectors,

Corresponding to the 4th default precoding vectors.Here

Be different default precoding vectors.An example of default precoding vectors is

Four characteristic vectors corresponding to the transmitting terminal covariance matrix.First channel direction (corresponding to first characteristic vector of receiving terminal covariance matrix) of 1 ' expression receiving terminal 202, second channel direction (corresponding to second characteristic vector of receiving terminal covariance matrix) of 2 ' expression receiving terminal 202.

Receiving terminal 202 is according to the reference signal that is received, calculating channel at described different default precoding vectors receiving terminal covariance matrix.

For the situation that has four transmitting antennas, two reception antennas in the cell base station 201 as shown in Figure 5, a certain arbitrarily given precoding vectors can be represented as the linear combination of the default precoding vectors of above-mentioned four mutually orthogonals:

$\stackrel{\&RightArrow;}{w}=U^H\stackrel{\&RightArrow;}{a}=a_1{\stackrel{\&RightArrow;}{u}}_1^H+a_2{\stackrel{\&RightArrow;}{u}}_2^H+a_3{\stackrel{\&RightArrow;}{u}}_3^H+a_4{\stackrel{\&RightArrow;}{u}}_4^H---\left(8\right)$

At this a certain given precoding vectors, receiving terminal covariance (cross-correlation) matrix of receiving terminal 202 is:

$E\left[\stackrel{\&RightArrow;}{h}{\stackrel{\&RightArrow;}{h}}^H\right]=E\left[H\stackrel{\&RightArrow;}{w}{\stackrel{\&RightArrow;}{w}}^H{H}^H\right]$

$=E\left[U_R(H_{\mathrm{ud}}\&CircleTimes;M)U_T{U}_T^H\stackrel{\&RightArrow;}{a}{\stackrel{\&RightArrow;}{a}}^T{U}_T{U}_T^H{(H_{\mathrm{ud}}\&CircleTimes;M)}^H{U}_R^H\right]$

$=U_{R}E\left[(H_{\mathrm{ud}}\&CircleTimes;M)\stackrel{\&RightArrow;}{a}{\stackrel{\&RightArrow;}{a}}^T{(H_{\mathrm{ud}}\&CircleTimes;M)}^H\right]U_R^H---\left(9\right)$

$=U_{R}E\left[\begin{array}{cc}\left[\begin{array}{cccc}{a}_1{h}_{11}{m}_{11}& a_2{h}_{12}{m}_{12}& a_3{h}_{13}{m}_{13}& a_3{h}_{14}{m}_{14}\\ a_1{h}_{21}{m}_{21}& a_2{h}_{22}{m}_{22}& a_3{h}_{23}{m}_{23}& a_3{h}_{24}{m}_{24}\end{array}\right]& \left[\begin{array}{cc}{a}_1{h}_{11}^H{m}_{11}& a_1{h}_{21}^H{m}_{21}\\ a_2{h}_{12}^H{m}_{12}& a_2{h}_{22}^H{m}_{22}\\ a_3{h}_{13}^H{m}_{13}& a_3{h}_{23}^H{m}_{23}\\ a_3{h}_{14}^H{m}_{14}& a_3{h}_{24}^H{m}_{24}\end{array}\right]\end{array}\right]U_R^H$

$=U_R(a_1^2\left[\begin{array}{cc}{m}_{11}& 0\\ 0& m_{21}\end{array}\right]+a_2^2\left[\begin{array}{cc}{m}_{12}& 0\\ 0& m_{22}\end{array}\right]+a_3^2\left[\begin{array}{cc}{m}_{13}& 0\\ 0& m_{23}\end{array}\right]+a_4^2\left[\begin{array}{cc}{m}_{14}& 0\\ 0& m_{24}\end{array}\right])U_R^H$

Herein, U _RThe eigenvectors matrix of expression receiving terminal covariance matrix,

Be corresponding to first precoding vectors

The receiving terminal covariance matrix, m ₁₁And m ₂₁Be its characteristic value, at first precoding vectors

Direction on, m ₁₁And m ₂₁And the eigenvectors matrix U of receiving terminal covariance matrix _RIt is the value that need feed back to cell base station 201 from receiving terminal 202.Here, under the situation of having only a receiving terminal 202 (single user's multi-input multi-output system), might ignore the eigenvectors matrix U of receiving terminal covariance matrix _RAnd do not feed back.Therefore, only need feedback characteristic value m ₁₁And m ₂₁Get final product.

Equally,

Be corresponding to second precoding vectors

The receiving terminal covariance matrix, m ₁₂And m ₂₂Be its characteristic value, at second precoding vectors

Direction on, m ₁₂And m ₂₂It is the value that need feed back to cell base station 201 from receiving terminal 202.

Be corresponding to the 3rd precoding vectors The receiving terminal covariance matrix, m ₁₃And m ₂₃Be its characteristic value, at the 3rd precoding vectors

Direction on, m ₁₃And m ₂₃It is the value that need feed back to cell base station 201 from receiving terminal 202.

Be corresponding to the 4th precoding vectors The receiving terminal covariance matrix, m ₁₄And m ₂₄Be its characteristic value, at the 4th precoding vectors Direction on, m ₁₄And m ₂₄It is the value that need feed back to cell base station 201 from receiving terminal 202.

According to one embodiment of present invention, can obtain:

d _max＝max[(a ₁ ²m ₁₁+a ₂ ²m ₁₂+a ₃ ²m ₁₃+a ₄ ²m ₁₄)，(a ₁ ²m ₂₁+a ₂ ²m ₂₂+a ₃ ²m ₂₃+a ₄ ²m ₂₄)] (10)

d _min＝min[(a ₁ ²m ₁₁+a ₂ ²m ₁₂+a ₃ ²m ₁₃+a ₄ ²m ₁₄)，(a ₁ ²m ₂₁+a ₂ ²m ₂₂+a ₃ ²m ₂₃+a ₄ ²m ₂₄)] (11)

Here, d _MaxAnd d _MinAs eigenvalue of maximum and the minimal eigenvalue shown in for example formula (6) at the receiving terminal covariance matrix R of a certain given precoding vectors, can be the combination of the characteristic value of above-mentioned each receiving terminal covariance matrix corresponding to different default precoding vectors, also can be called as " maximum weighted characteristic value " and " minimum weight characteristic value ".

Simultaneously, in these cases, transmitting terminal covariance matrix R _TCan be expressed as:

$R_T=U_T{D}_T{U_T}^H$

$=U_T\left[\begin{array}{cccc}{m}_{11}{m}_{21}& 0& 0& 0\\ 0& m_{12}+m_{22}& 0& 0\\ 0& 0& m_{13}+m_{23}& 0\\ 0& 0& 0& m_{14}+m_{24}\end{array}\right]{U_T}^H---\left(12\right)$

Here, U _TThe characteristic vector of expression base station end covariance matrix.

Thus, can design precoder,, satisfy promptly according to formula (7) according to the present invention

Condition under, seek a certain given precoding vectors

Value, make expression formula

The value maximum.

From the above, receiving terminal can be preset a plurality of precoding vectors, each precoding vectors corresponds respectively to a receiving terminal covariance matrix, receiving terminal 202 has a plurality of reception antennas (for example two), and wherein each receiving terminal covariance matrix has the characteristic value with the number similar number of reception antenna.According to the abovementioned embodiments of the present invention, described terminal forms four default precoding vectors, corresponding to four receiving terminal covariance matrixes.

According to an aspect of the present invention, be that the quantity of the characteristic value in above-mentioned each correlation matrix is two under two the situation at the number of the reception antenna of receiving terminal 202, at this moment, the characteristic value that needs in each correlation matrix to feed back is two.The number of the reception antenna of receiving terminal 202 greater than two situation under, the quantity of the characteristic value in each correlation matrix will be greater than two, for example under situation with three reception antennas, three characteristic values are arranged in each correlation matrix, at this moment, the characteristic value that needs in each correlation matrix to feed back is three.

From as can be known above-mentioned, base station end covariance matrix that can obtain channel at receiving terminal 202 places and receiving terminal covariance matrix at different default precoding vectors, and fed back base station end covariance matrix to cell base station 201, and whole characteristic value m of described receiving terminal covariance matrix from receiving terminal 202 ₁₁, m ₂₁, m ₁₂, m ₂₂, m ₁₃, m ₂₃, m ₁₄And m ₂₄But, traditional when utilizing in the communication system that wave beam forms the design precoding vectors, fed back the value (characteristic value) of base station end covariance matrix to cell base station 201 from receiving terminal 202, for example be (m ₁₁+ m ₂₁) and the value, (m ₁₁+ m ₂₁) and the value, (m ₁₁+ m ₂₁) with the value and (m ₁₁+ m ₂₁) and the value.Therefore, feeding back on the basis of base station end covariance matrix, feeding back whole characteristic values of receiving terminal covariance matrix again, increase feedback overhead that can be unnecessary to cell base station 201.

According to another aspect of the present invention, in order to reduce feedback overhead, at each mutually orthogonal precoding vectors, receiving terminal 202 only needs to feed back the partial feature value in the receiving terminal covariance matrix.For example, can set the characteristic vector of the precoding vectors of a plurality of (for example being Nt) mutually orthogonal corresponding to base station end covariance matrix.In the case, if the quantity of the reception antenna of receiving terminal is n, then only need be on the basis of feedback base station end covariance matrix, many feedback individual characteristic values of Nt (n-1) (arithmetic number) get final product again.For example in Fig. 5, the number N t of the precoding vectors of mutually orthogonal is four, and reception antenna only has two, can make

Be base station end covariance matrix R _TCharacteristic vector U _T, then because corresponding to first precoding vectors

The characteristic value m of receiving terminal covariance matrix ₁₁With m ₂₁Be base station end covariance matrix R with value _TCorresponding to precoding vectors

Characteristic value (m ₁₁+ m ₂₁), therefore only need be from receiving terminal 202 to cell base station 201 feedback m ₁₁And m ₂₁In one, and do not need two values all to feed back, because another can be according to this base station end covariance matrix R _TCorresponding to precoding vectors

Characteristic value and the amount (m that is fed ₁₁And m ₂₁In one) between calculate with difference relation, thereby reduced feedback overhead.Covariance (being correlated with) matrix corresponding to other precoding vectors also has same characteristic.Therefore in situation shown in Figure 5, only need on the basis of feedback base station end covariance matrix, feed back four characteristic values (arithmetic number) more again and get final product, and not need to feed back eight whole characteristic values.

Promptly, in the scheme of cell base station feedback base station end covariance matrix and receiving terminal covariance matrix, both can feed back whole characteristic values of receiving terminal covariance matrix to cell base station, also can feed back the partial feature value of receiving terminal covariance matrix to cell base station.

According to the foregoing description, can design precoder, make the data flow that on the direction of precoding vectors, sends, experience is lower than signal received power threshold, it is less than the possibility of great fluctuation process amplitude to have.Avoided the fluctuating range of received power of data flow excessive thus.Like this, can optimize the precoder of cell base station 201.

Fig. 6 illustrates the diagram of optimizing the result of precoder according to receiving terminal covariance matrix of the present invention (channel relevancy).

In Fig. 6, straight dashed line is illustrated in a precoding vectors

Average signal received power on the direction, imaginary curve is illustrated in this precoding vectors Actual signal received power on the direction, number line is illustrated in another precoding vectors

Average signal received power on the direction, solid-line curve are represented this another precoding vectors

Actual signal received power on the direction.Dot-dash straight line below among Fig. 6 is represented signal received power threshold s, and the data flow that the signal received power is lower than this threshold value can not normally be received.As can be seen from Figure 6, according to traditional method, the higher precoding vectors of average signal received power will be selected to utilize

Design precoder, but the actual signal received power of this direction is lower than signal received power threshold on some position, for example from an a1 to putting a2, and from an a3 to the position of putting a4, thereby cause signal normally not received.According to embodiments of the invention, another lower precoding vectors of average signal received power will be selected to utilize

Design precoder, on this direction, the actual signal received power of each position all is higher than signal received power threshold, makes signal can be in good accepting state.

Fig. 7 is another schematic diagram of the characteristic of channel that cell base station transmits to receiving terminal in the communication system of Fig. 2.

As shown in Figure 7, in order to reduce feedback overhead, can also feed back the value of receiving terminal covariance matrix roughly.In Fig. 7, will be corresponding to first precoding vectors

With second precoding vectors

Receiving terminal covariance matrix combined statement be shown

With m ₁₁And m ₂₁Feed back to cell base station 201 from receiving terminal 202, to be illustrated in first precoding vectors

With second precoding vectors

The correlation of the receiving terminal on this both direction.Will be corresponding to the 3rd precoding vectors With the 4th precoding vectors Receiving terminal covariance matrix combined statement be shown

With m ₁₂And m ₂₂Feed back to cell base station 201 from receiving terminal 202, to be illustrated in the 3rd precoding vectors

With the 4th precoding vectors

The correlation of the receiving terminal on this both direction.Thus, can further reduce the value that feeds back to cell base station 201, thereby reduce feedback overhead.

And, can also only feed back the partial feature value in each receiving terminal covariance matrix according to above stated specification, further reduce the value that feeds back to cell base station 201, thereby reduce feedback overhead.

According to embodiments of the invention, formula (7) is an example of precoder of considering the channel relevancy of receiving terminal.According to other embodiments of the invention, the channel relevancy of receiving terminal can be represented as other arithmetic operator value of the characteristic value of each receiving terminal covariance matrix.Here, the channel relevancy of receiving terminal can be represented as the maximum weighted characteristic value of each receiving terminal covariance matrix and the arithmetic operator value of minimum weight characteristic value, as,

f(d _max，d _min)＜α (13)

Correlation between the value representation variable wherein of the function f () in the formula (13), as shown in formula (7), the value of this function can be the merchant of maximum weighted characteristic value and minimum weight characteristic value.Above-mentioned expression formula for example can also be expressed as:

d _max ²/d _min ²＜α；

The value of above-mentioned function also can be absolute value poor of maximum weighted characteristic value and minimum weight characteristic value, as,

|d _max|-|d _min|＜α；

|d _max| ²-|d _min| ²＜α；

The value of above-mentioned function can also be the merchant's of maximum weighted characteristic value and minimum weight characteristic value a inverse, for example,

d _min/d _max＞α；

d _Min ²/ d _Max ²＞α etc.

Also this function might be directly embedded in the optimization criterion of precoder, for example in addition

$\underset{\stackrel{\&RightArrow;}{w}}{\arg \max }({\stackrel{\&RightArrow;}{w}}^H{R}_T\stackrel{\&RightArrow;}{w}-\frac{d_{\max }}{d_{\min }})$

Those skilled in the art should be able to know that other embodiment that considers the precoder of channel relevancy still may exist, and still falls within the scope of the invention.

Fig. 8 is the flow chart of method that is used for optimizing the precoder of communication system according to the present invention.

As shown in Figure 8, at step S801, sending reference signal from cell base station to receiving terminal, for example is the sub-district special reference signal.At step S802, receiving terminal obtains Channel Transmission matrix on different running time-frequency resources by the reference signal that receives, and according to the transmitting terminal covariance matrix of described Channel Transmission matrix computations channel and receiving terminal covariance matrix at different default precoding vectors.At step S803, base station end covariance matrix that receiving terminal is calculated to cell base station feedback and receiving terminal covariance matrix at different default precoding vectors.At step S804, cell base station forms corresponding precoding vectors based on the base station end covariance matrix and the above-mentioned receiving terminal covariance matrix that receive.

According to an aspect of the present invention, above-mentioned steps S801 can be realized by receiving element 401 in transmitting element in the cell base station 201 302 and the receiving terminal 202 and antenna 304 and 404.Above-mentioned steps S802 can be realized by the computing unit in the receiving terminal 202 403.Above-mentioned steps S803 can be realized by receiving element 302 in transmitting element in the receiving terminal 202 402 and the cell base station 201 and antenna 304 and 404.Above-mentioned steps S804 can be realized by the precoder in the cell base station 201 301.

In the above embodiment of the present invention, the precoder of cell base station has not only been considered the average throughput of communication system, also considered the amplitude fluctuation size of signal received power, i.e. the actual channel correlation of receiving terminal has improved the performance of whole communication system thus.

Second embodiment

In actual communication systems, exist a plurality of base stations to carry out the situation of transceive data simultaneously, existing many base stations mode comprises joint transmission, the formation of collaboration type wave beam and dynamic cell selection etc.

Fig. 9 illustrates two cell base stations to form the diagram that sends signal by the collaboration type wave beam.

In Fig. 9, the cell base station 901 in the communication system 900 sends signals to receiving terminal 903, and the cell base station 902 adjacent with cell base station 901 sends signals to receiving terminal 904.Because cell base station 901 is adjacent with cell base station 902, the part signal power (leakage) that cell base station 901 takes place easily flows to the situation of receiving terminal 904, causes the interference to receiving terminal 904.Equally, the part signal power of cell base station 902 also may (leak) and flow to receiving terminal 903, causes the interference to receiving terminal 903.Therefore, during precoding vectors (precoder) in as shown in Figure 9 collaboration type wave beam of design forms, need consider signal strength signal intensity simultaneously to the receiving terminal (for example receiving terminal 903) of this sub-district (for example cell base station 901), and to the interference strength of the receiving terminal (for example receiving terminal 904) of adjacent sub-district (for example cell base station 902).

The precoder that a kind of known collaboration type wave beam forms is designed to:

Wherein the letter of the signal (for example data flow) that sends for cell base station 901 of SLNR (signal-leakage-noise ratio) leak make an uproar than value, this value is big more, the average throughput of the data flow of expression cell base station 901 transmissions is high more; R _SBe the base station end covariance matrix (be similar to the situation of precoder in shown in Figure 2 single subdistrict) of cell base station 901 at the receiving terminal 903 of this sub-district (cell base station 901); R _LBe the base station end covariance matrix of cell base station 901 at the receiving terminal 904 of adjacent sub-district (cell base station 902).

According to embodiments of the invention, form for multi-base station cooperative formula wave beam, also the correlation of receiving terminal can be joined in the design of precoder.

Particularly, can comprise a plurality of sub-districts (being not limited to 2 sub-districts shown in Figure 9) according to communication system of the present invention, the receiving terminal (receiving terminal 903) of base station (cell base station 901) to this sub-district that is positioned at this sub-district sends reference signal, and receive from the base station end covariance matrix about this cell base station (cell base station 901) of receiving terminal (receiving terminal 903) feedback of this sub-district, and the receiving terminal covariance matrix at different default precoding vectors of the terminal of this sub-district (receiving terminal 903), the base station (cell base station 901) of this sub-district also receives the base station end covariance matrix about this cell base station (cell base station 901) of terminal (receiving terminal 904) feedback from neighbor cell simultaneously, and the receiving terminal covariance matrix at different default precoding vectors of the terminal of this neighbor cell (receiving terminal 904).

According to another aspect of the present invention, the terminal of this neighbor cell (receiving terminal 904) is at first to the base station end covariance matrix of another base station that is positioned at this neighbor cell (cell base station 902) feedback about this cell base station (cell base station 901), receiving terminal covariance matrix with the terminal (receiving terminal 904) of this neighbor cell at different default precoding vectors, the base station of this sub-district (cell base station 901) connects (wired or wireless) by the backstage between the base station, feeds back from the terminal (receiving terminal 904) of this neighbor cell from this another base station (cell base station 902) reception of this neighbor cell, receiving terminal covariance matrix about the terminal (receiving terminal 904) of the base station end covariance matrix of this cell base station (cell base station 901) and this neighbor cell at different default precoding vectors.

According to an aspect of the present invention, above-mentioned value can be the characteristic value of receiving terminal covariance matrix of representing the channel relevancy of each terminal.

Thus, can be optimised for according to precoder of the present invention: under channel relevancy the prerequisite greater than second threshold value of channel relevancy less than the receiving terminal of first threshold and this neighbor cell of the receiving terminal of this sub-district, the average throughput of maximization communication system.According to an aspect of the present invention, at a certain given precoding vectors, the channel relevancy of receiving terminal can be represented as at the eigenvalue of maximum of the receiving terminal covariance matrix of this a certain given precoding vectors and the arithmetic operator value of minimal eigenvalue.For example, this arithmetic operator value can be the merchant of above-mentioned eigenvalue of maximum and minimal eigenvalue, also can be the poor of above-mentioned eigenvalue of maximum and minimal eigenvalue, or the like.

Can represent according to precoder of the present invention by following formula (15):

$\underset{\stackrel{\&RightArrow;}{w}}{\arg \max }\frac{{\stackrel{\&RightArrow;}{w}}^H{R}_S\stackrel{\&RightArrow;}{w}}{{\stackrel{\&RightArrow;}{w}}^H{R}_L\stackrel{\&RightArrow;}{w}+{\mathrm{\&sigma;}}^2}$ Condition: $\frac{d_{s,\max }}{d_{s,\min }}<\mathrm{\&alpha;};\frac{d_{L,\max }}{d_{L,\min }}>\mathrm{\&beta;}---\left(15\right)$

Wherein,

Be a certain given precoding vectors, (.) ^HRepresent conjugate transpose, R _SBe the base station end covariance matrix of this cell terminal, R _LBe the base station end covariance matrix of adjacent cell terminal, σ is a noise component(s), d _{S, max}And d _{S, min}Be respectively the eigenvalue of maximum and the minimal eigenvalue at the receiving terminal covariance matrix of a certain given precoding vectors of the receiving terminal 603 of this sub-district (cell base station 901), d _{L, max}And d _{L, min}Be respectively the eigenvalue of maximum and the minimal eigenvalue at the receiving terminal covariance matrix of a certain given precoding vectors of the receiving terminal 904 of adjacent sub-district (cell base station 902), α, β are respectively the base station end presetting first threshold and second threshold value.

Above-mentioned condition

Implication be: the eigenvalue of maximum and the ratio between the minimal eigenvalue at the receiving terminal covariance matrix of a certain given precoding vectors of the receiving terminal 903 of this sub-district (cell base station 901) should be less than predetermined thresholds, condition

Implication be: the receiving terminal 904 of adjacent sub-district (cell base station 902), about a certain given precoding vectors of cell base station 901 the eigenvalue of maximum and the ratio between the minimal eigenvalue of receiving terminal covariance matrix should be greater than another predetermined threshold.

Under this form, each receiving terminal not only needs respectively at this sub-district and adjacent sub-district feedback relevant base station end covariance matrix, for example the base station end covariance matrix R of the receiving terminal 903 of this sub-district (cell base station 901) _SBase station end covariance matrix R with the receiving terminal 904 of adjacent sub-district (cell base station 902) _L, but also will be respectively at this sub-district and adjacent sub-district, at a plurality of quadrature precoding vectors, feed back corresponding receiving terminal covariance matrix.

In example as shown in Figure 9, receiving terminal 903 directly feeds back the base station end covariance matrix at receiving terminals 903 places to cell base station 901, and the receiving terminal covariance matrix at different default precoding vectors of receiving terminal 903, receiving terminal 904 then elder generation to (about cell base station 901) at cell base station 902 feedback receiving terminals 904 places base station end covariance matrix, and at the receiving terminal covariance matrix of different default precoding vectors, cell base station 902 is by background process then, and the base station end covariance matrix and the receiving terminal covariance matrix at receiving terminal 904 places that received sent to cell base station 901.

Thus, cell base station 901 according to respectively from above-mentioned each values of receiving terminal 903 and receiving terminal 904 feedbacks, in the condition that satisfies formula (15):

Prerequisite under, determine suitable a certain given precoding vectors

Make the letter of the signal that sends from cell base station 901 leak and make an uproar, can realize the raising of the performance of whole communication system thus than the value maximum of (SLNR).

The present embodiment supposition is used letter to let out and is made an uproar than (SLNR) as optimizing performance index.Those skilled in the art should be able to understand that other optimization performance index are Signal to Interference plus Noise Ratio (SINR) for example, and what channel capacity etc. also can be same is applied among the present invention.

When regarding cell base station 902 as this sub-district, and when regarding cell base station 901 as adjacent sub-district, can design the precoder of cell base station 902 with identical method.

The 3rd embodiment

Situation when transmission rank is one (cell base station only sends a data flow) has been described in aforementioned two embodiment.Present embodiment will be described the situation when transmission rank is not one (cell base station sends a plurality of data flow simultaneously).

When the order of transmission be for the moment, can be expressed as from the signal of cell base station transmission:

$\stackrel{\&RightArrow;}{z}=\stackrel{\&RightArrow;}{w_1}{x}_1+\stackrel{\&RightArrow;}{w_2}{x}_2\&CenterDot;\&CenterDot;\&CenterDot;\stackrel{\&RightArrow;}{w_n}{x}_n---\left(16\right)$

Here, The signal that expression sends; x ₁, x ₂... x _nRespectively first signal that sends of expression, secondary signal ..., the n signal;

Respectively expression send first signal, secondary signal ..., each precoding vectors of n signal.As seen, the signal of transmission is the stack of each signal of sending with different precoding vectors respectively.

Corresponding with above-mentioned transmission signal, can be expressed as in the received signal of receiving terminal:

$\stackrel{\&RightArrow;}{y}=H\stackrel{\&RightArrow;}{w_1}{x}_1+H\stackrel{\&RightArrow;}{w_2}{x}_2\&CenterDot;\&CenterDot;\&CenterDot;H\stackrel{\&RightArrow;}{w_n}{x}_n+\stackrel{\&RightArrow;}{n}=\stackrel{\&RightArrow;}{h_1}{x}_1+\stackrel{\&RightArrow;}{h_2}{x}_2\&CenterDot;\&CenterDot;\&CenterDot;\stackrel{\&RightArrow;}{h_n}{x}_n+\stackrel{\&RightArrow;}{n}---\left(17\right)$

Here,

The signal that expression receives; x ₁, x ₂... x _nRespectively first signal that sends of expression, secondary signal ..., the n signal;

Respectively expression send first signal, secondary signal ..., the precoding vectors of n signal; H represents the Channel Transmission matrix;

The expression noise component(s).

By formula (17) is compared as can be known with formula (5), for different signals or data flow, can be respectively at each precoding vectors carry out as first embodiment and second embodiment described in optimization process.Thus, can realize not being the optimal design of the precoder of cell base station for the moment when transmission rank.

Therefore, when the transmission rank of communication system was not one (perhaps greater than one), promptly signal of Fa Songing or data flow were under a plurality of situation, receiving terminal identical feedback method when still adopting and being with transmission rank.In the precoder of base station end, can design a plurality of different a certain given precoding vectors, the correlation that all satisfies its corresponding receiving terminal covariance matrix is lower than the condition of a certain threshold value.When these a plurality of different a certain given precoding vectors have preferably orthogonality, described a plurality of a certain given precoding vectors can be used for transmitting a plurality of data flow (order is greater than) by above-mentioned each embodiment of the present invention, even the order of transmission is not one, still the maximum average throughput of taking into account system and the channel relevancy of receiving terminal design precoder simultaneously, make it possible to significantly improve the overall performance of communication system.

Above-mentioned each embodiment among the application only is exemplary description, their concrete structure and operation are not construed as limiting scope of the present invention, those skilled in the art can make up different piece and operation among above-mentioned each embodiment, produce new execution mode, same design according to the invention.

Embodiments of the invention can by hardware, software, firmware or between them the mode of combination realize that its implementation is not construed as limiting scope of the present invention.

Each function element (unit) annexation each other in the embodiments of the invention is not construed as limiting scope of the present invention, and one or more function element can comprise or be connected in other function element arbitrarily.

Though illustrated and described some embodiments of the present invention above in conjunction with the accompanying drawings, but those skilled in the art is to be understood that, under the situation that does not depart from principle of the present invention and spirit, can change and revise these embodiment, but they still drop within the scope of claim of the present invention and equivalent thereof.

Claims (15)

1. method that is used for optimizing the precoder of communication system, described communication system comprises terminal and the base station of realizing described precoder, described method comprises step:

Send reference signal from described base station to described terminal;

Obtain Channel Transmission matrix on the different running time-frequency resources by described reference signal;

According to the transmitting terminal covariance matrix of described Channel Transmission matrix computations channel and receiving terminal covariance matrix at different default precoding vectors;

To the described base station of described base station feedback end covariance matrix and receiving terminal covariance matrix at different default precoding vectors; And

Optimize described precoder based on described base station end covariance matrix and receiving terminal covariance matrix, to form precoding vectors.

2. the method for claim 1 comprises with receiving terminal covariance matrix at different default precoding vectors to the described base station of described base station feedback end covariance matrix: to the described base station of described base station feedback end covariance matrix and all or part of eigen vector at the receiving terminal covariance matrix of different default precoding vectors.

3. the method for claim 1, described terminal pins is calculated corresponding receiving terminal covariance matrix to different default precoding vectors, and each described receiving terminal covariance matrix has the characteristic value with the number similar number of the reception antenna of described terminal.

4. method as claimed in claim 3, different default precoding vectors are the characteristic vector of transmitting terminal covariance matrix.

5. the method for claim 1, the optimization criterion of described precoder is: under the prerequisite of channel relevancy less than predetermined threshold of described terminal, the average throughput of maximization communication system.

6. method as claimed in claim 5, the channel relevancy of described terminal are represented as the arithmetic operator value at the characteristic value of the receiving terminal covariance matrix of a certain given precoding vectors.

7. method as claimed in claim 6, the optimization criterion of described precoder is:

$\underset{\stackrel{\&RightArrow;}{w}}{\arg \max }{\stackrel{\&RightArrow;}{w}}^H{R}_T\stackrel{\&RightArrow;}{w}$ Condition: $\frac{d_{\max }}{d_{\min }}<\mathrm{\&alpha;},$

Wherein,

Be a certain given precoding vectors, (.) ^HRepresent conjugate transpose, R _TBe the base station end covariance matrix of channel, d _MaxWith d _MinBe respectively eigenvalue of maximum and minimal eigenvalue at the receiving terminal covariance matrix of a certain given precoding vectors, α is the default described predetermined threshold in base station.

8. the method for claim 1, described communication system comprises a plurality of sub-districts, the terminal of described base station to this sub-district that is positioned at this sub-district sends described reference signal, and receive from the base station end covariance matrix of the terminal feedback of this sub-district and receiving terminal covariance matrix at different default precoding vectors, receive simultaneously from the terminal feedback of neighbor cell about the base station end covariance matrix of described base station and the receiving terminal covariance matrix at different default precoding vectors of described neighbor cell.

9. method as claimed in claim 8, the terminal of described neighbor cell is at first to the receiving terminal covariance matrix at different default precoding vectors of the described base station of another base station feedback end covariance matrix that is positioned at described neighbor cell and described neighbor cell, and the described base station of this sub-district receives from described another base station of described neighbor cell from the described base station end covariance matrix of the terminal feedback of neighbor cell and the receiving terminal covariance matrix at different default precoding vectors of described neighbor cell.

10. method as claimed in claim 8, the optimization criterion of described precoder is, under channel relevancy the prerequisite greater than second threshold value of channel relevancy less than the terminal of first threshold and described neighbor cell of the terminal of this sub-district, maximize the average throughput of described communication system.

11. method as claimed in claim 10, the optimization criterion of described precoder is:

$\underset{\stackrel{\&RightArrow;}{w}}{\arg \max }\frac{{\stackrel{\&RightArrow;}{w}}^H{R}_S\stackrel{\&RightArrow;}{w}}{{\stackrel{\&RightArrow;}{w}}^H{R}_L\stackrel{\&RightArrow;}{w}+{\mathrm{\&sigma;}}^2}$ Condition: $\frac{d_{s,\max }}{d_{s,\min }}<\mathrm{\&alpha;};\frac{d_{L,\max }}{d_{L,\min }}>\mathrm{\&beta;}$

Wherein,

Be a certain given precoding vectors, (.) ^HRepresent conjugate transpose, R _SBe the base station end covariance matrix of this cell terminal, R _LBe the base station end covariance matrix of adjacent cell terminal, σ is a noise component(s), d _{S, max}And d _{S, min}Be respectively eigenvalue of maximum and minimal eigenvalue, d at the receiving terminal covariance matrix of this cell terminal of a certain given precoding vectors _{L, max}And d _{L, min}Be respectively eigenvalue of maximum and minimal eigenvalue at the receiving terminal covariance matrix of the adjacent cell terminal of a certain given precoding vectors, α, β are respectively the default described first threshold and second threshold values of base station.

12. as each described method of claim 1-11, the transmission rank of described communication system is one.

13. as each described method of claim 1-11, when the transmission rank of described communication system greater than for the moment, described terminal pins is to each signal or data flow, respectively the corresponding optimum precoding vectors of design.

14. a base station that is used for communication system, described base station comprises:

Transmitting element sends reference signal to terminal;

Receiving element receives based on described reference signal base station end covariance matrix that obtain, channel and receiving terminal covariance matrix corresponding to different default precoding vectors from described terminal; And

Precoder based on described base station end covariance matrix and receiving terminal covariance matrix corresponding to different default precoding vectors, forms precoding vectors.

15. a terminal that is used for communication system, described terminal comprises:

Receiving element receives reference signal from the base station;

Computing unit obtains Channel Transmission matrix on the different running time-frequency resources by described reference signal, according to the transmitting terminal covariance matrix of described Channel Transmission matrix computations channel and receiving terminal covariance matrix at different default precoding vectors; And

Transmitting element is to the described base station of described base station feedback end covariance matrix and described receiving terminal covariance matrix at different default precoding vectors.

Images