CN101227444A - Method and device for selecting precoding weight vectors - Google Patents

Abstract

The invention provides a selecting device of pre-code weight vector, which comprises an estimating unit which is used to estimate signal-to-noise ratio which is corresponded with each data flow, an assuring unit which is used to inquire modem coding scheme MCS to assure source rate which is corresponded with each data flow according to the signal-to-noise ratio, and a selecting unit which is used to select pre-code weight vector according to the source rate. The invention also provides a selecting method of pre-code weight vector. The example of the invention can improve the capacity of system channel in a certain extent. And meanwhile, the simple and effective weight vector selecting method of the example of the invention leads a UE end to be simple and feasible in the process of selecting feedback weight vector or matrix, and the manufacture and the production of UE terminal are simpler.

Description

The choice device of precoding weight vectors and method

Technical field

The present invention relates to the communications field, especially relate to a kind of choice device and method of precoding weight vectors.

Background technology

HSDPA (high speed downlink packet access) be to solve that the WCDMA system covers and capacity between contradiction, eliminate disturb, problems such as elevator system capacity and message transmission rate, satisfy the customer service demand and the new technology that produces.

In the prior art,, make the throughput of down channel can further rise to 20Mbit/s by in the HSDPA technology, introducing MIMO new technologies such as (multiple-input and multiple-outputs).Mimo system is to use many antennas simultaneously in transmission and receiving terminal, by effectively utilizing the multipath transmisstion effect to reach the purpose that improves spectrum efficiency.Mimo system can improve the capability of communication system and the availability of frequency spectrum exponentially under the situation that does not increase bandwidth.Channel capacity is linear the increase along with the increase of antenna amount, thereby under the situation that does not increase bandwidth and antenna transmission power, the availability of frequency spectrum can improve exponentially.Utilize the MIMO technology not only can improve channel capacity, also improved the reliability of channel and reduced the error rate simultaneously.

In existing 3GPP technology, the MIMO scheme will adopt the transmitter model based on D-TxAA (two transmitting antenna array), and its essence is exactly the MIMO of 2 * 2 precodings, as shown in Figure 1.In the D-TxAA model, when Node B (Node B) when being scheduling to double fluid, two data flow are called primary traffic and secondary data stream.The HS-DSCH of two data flow (high speed descending sharing channel) transmission block adds coefficient w by precoding module after handling (comprising operations such as chnnel coding, rate-matched, modulation) and spread spectrum and scrambling processing through transmission channel separately respectively ₁, w ₂, w ₃And w ₄Form two paths of data stream after the weight, add pilot signal CPICH then respectively ₁(Common Pilot Channel 1) and CPICH ₂After (Common Pilot Channel 2), launch through two different antennas.When Node B is scheduling to single data stream, have only the main transport block data stream of HS-DSCH to transmit, and by weight coefficient w ₁And w ₃Form two data flow after the weighting, add pilot signal CPICH equally ₁(Common Pilot Channel 1) and CPICH ₂(Common Pilot Channel 2) after two different antennas launch.Wherein, the main transport block data of HS-DSCH flows the output that always is scheduled, and time transmission of HS-DSCH then is that optionally promptly Node B both can be scheduling to single-stream transmission and also can be scheduling to stream transmission soon.

After introducing MIMO, control signaling link HS-SCCH (High-Speed Shared Control Channel) and HS-DPCCH (uplink high-speed special physical control channel) for two relevant with HS-PDSCH (high-speed physical downlink shared channel (HS-PDSCH)), carry more control information for supporting the MIMO technology.Under the MIMO pattern, user equipment (UE) is selected after the optimum pre-coding matrix, by HS-DPCCH channel notification Node B according to the situation of current channel.

HS-DPCCH has carried HARQ-ACK (automatic retransfer response) information and the PCI (precoding control indication) and corresponding C QI (channel quality indication) information of HAQR (the retransmitting automatically) operation that is respectively applied for two streams under the MIMO pattern, HARQ-ACK information and PCI/CQI information are encoded respectively, and are mapped in first time slot of HS-DPCCH subframe and latter two time slot respectively.UE is according to the pre-configured ratio of the upper-layer protocol CQI information of two types of reporting types A (Type A) and type B (Type B) simultaneously: during type A, UE selects the CQI information of the double-current and single current of report according to actual conditions; During type B, no matter actual conditions how UE be fixed for the CQI information of report flow transmission.Wherein, CQI information is always based on selected precoding vector or matrix, so PCI/CQI is corresponding.

The precoding weight selection course of D-TxAA:

Under the MIMO pattern, UE selects weight vectors according to certain rule (as the Signal to Interference plus Noise Ratio SINR of maximization UE end etc.), then by UE by the compound field notice of the PCI/CQI in HS-DPCCH subframe NodeB.Node B scheduler is reported the selection of the weight vectors that decides this transmission according to PCI/CQI.The precoding weight vectors that UE selects meets following rule:

■ precoding weight w ₁And w ₃Get constant real number value

Promptly $w_1={\mathrm{<figure-callout\; id="3"\; label="w"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">w</figure-callout>}}_3=1/\sqrt{2};$

■ precoding weight w ₂And w ₄Be variable complex values, satisfy w simultaneously ₄=-w ₂Promptly in case the needed information w of precoding weight of main transmission block ₂Determine w ₄Also determine thereupon.w ₂In four imaginary numbers, select, promptly $w_2\&Element;\{\frac{1+\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2},\frac{1-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2},\frac{-1+\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2},\frac{-1-j}{2}\}.$

Under the stream transmission situation, to shape as $W=\left[\begin{array}{cc}{w}_1& w_3\\ w_2& {\mathrm{<figure-callout\; id="4"\; label="w"\; filenames="A20071007286800141.png,A20071007286800161.png"\; state="\{\{state\}\}">w</figure-callout>}}_4\end{array}\right]$ Pre-coding matrix, precoding weight information and PCI can indicate following four kinds of precoding vectors (matrix) set:

${\mathrm{<figure-callout\; id="1"\; label="W"\; filenames="A20071007286800141.png,A20071007286800151.png"\; state="\{\{state\}\}">W</figure-callout>}}_1=\left[\begin{array}{cc}\frac{1}{\sqrt{2}}& \frac{1}{\sqrt{2}}\\ \frac{1+j}{2}& -\frac{1+\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2}\end{array}\right],$ ${\mathrm{<figure-callout\; id="2"\; label="W"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">W</figure-callout>}}_2=\left[\begin{array}{cc}\frac{1}{\sqrt{2}}& \frac{1}{\sqrt{2}}\\ \frac{-1+j}{2}& -\frac{-1+\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2}\end{array}\right]$

${\mathrm{<figure-callout\; id="3"\; label="W"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">W</figure-callout>}}_3=\left[\begin{array}{cc}\frac{1}{\sqrt{2}}& \frac{1}{\sqrt{2}}\\ \frac{-1-j}{2}& -\frac{-1-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2}\end{array}\right],$ ${\mathrm{<figure-callout\; id="4"\; label="W"\; filenames="A20071007286800141.png,A20071007286800161.png"\; state="\{\{state\}\}">W</figure-callout>}}_4=\left[\begin{array}{cc}\frac{1}{\sqrt{2}}& \frac{1}{\sqrt{2}}\\ \frac{1-j}{2}& -\frac{1-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">j</figure-callout>}}{2}\end{array}\right]$

Prove the whether accurate performance that affects system to a certain extent that weight vectors (matrix) W selects by computer simulation.

Existing technical scheme is: the UE end obtains the relevant information of channel matrix from the control channel of two reception antennas, thereby obtains the signal to noise ratio of two data flow correspondences.Further obtain and two corresponding throughput of system of signal to noise ratio by Shannon (Shannon) formula, calculate the total system capacity that two paths of data stream obtains then, determine the weight combination by the maximum of two data current capacity sums, feed back to the weight of Node B then as transmission next time.Claim that the method is the shannon formula method, as shown in Figure 2.

The concrete grammar step is as follows:

1, under the MIMO pattern, UE is according to two Common Pilot Channel (CPICH ₁And CPICH ₂) information estimate respectively channel matrix between two antennas to be called H.

2, UE is according to 4 kinds of (W ₁, W ₂, W ₃And W ₄) set of weight vectors and the channel matrix that in the last step, obtained, obtain gathering the signal to noise ratio that corresponding two paths of data flows with 4 kinds of weight vectors.For example for weight vectors W ₁, the signal to noise ratio δ of the primary traffic that obtains from reception antenna ₁ ²(H, W ₁) and the signal to noise ratio δ of secondary data stream ₂ ²(H, W ₁), and the like.See table shown in 1:

The noise of the different weight vectors of table 1 is channel capacity when

3, the signal to noise ratio that obtains according to shannon formula and the 2nd step is calculated the signal volume of two reception antennas under four kinds of weight vectors situations respectively.Now still with weight vectors W ₁Situation is the example explanation, weight vectors W ₁Corresponding channel capacity is that two antennas obtain the capacity sum respectively, and the channel capacity of two reception antenna acquisitions is respectively log ₂(1+ δ ₁ ²(H, W ₁)) and log ₂(1+ δ ₁ ²(H, W ₁)), then the total capacity of channel is ${\mathrm{<figure-callout\; id="1"\; label="C"\; filenames="A20071007286800141.png,A20071007286800151.png"\; state="\{\{state\}\}">C</figure-callout>}}_1={\log }_2(1+{\mathrm{\&delta;}}_1^2(H,W_1))+{\log }_2(1+{\mathrm{\&delta;}}_2^2(H,W_1)).$ The channel capacity of other weight situation correspondences, and the like, also as above shown in the table 1.

4, relatively obtain the maximum of 4 kinds of channel capacities under the situation, wherein the weight vectors of maximum correspondence is selected weight combination then.That is, make that overall system capacity is that maximum weight vectors or matrix is selected combination, is expressed as $W=\underset{W}{\arg \max }({\log }_2(1+{\mathrm{\&delta;}}_1^2(H,W))+{\log }_2(1+{\mathrm{\&delta;}}_2^2(H,W))).$ For example, suppose channel capacity C ₃Be 4 (C ₁, C ₂, C ₃And C ₄) maximum in the value, then weight vectors W ₃Be selected weight vectors combination.

There is following shortcoming in above-mentioned scheme:

There is bigger error in shortcoming one: UE end in the process of calculating channel capacity, caused accurately selecting precoding weight vectors (matrix).By formula

log ₁₀(1+x)＝log ₁₀(e)*ln(1+x)＝

log ₁₀(e)*(x-x ²/2+x ³/3-x ⁴/4+...+(-1) ^k·x ^k/k+...)

As can be seen, when actual treatment, certainly exist error.

Shortcoming two:, be unfavorable for the design and the realization of side circuit because UE end calculating channel capacity process need carry out logarithm operation.

Summary of the invention

The technical problem that the embodiment of the invention will solve is, overcomes the deficiency that above-mentioned prior art exists, and provides a kind of choice device and method of precoding weight vectors, fast and effeciently to select weight vectors.

For solving the problems of the technologies described above, the embodiment of the invention provides a kind of choice device of precoding weight vectors, comprising:

Estimation unit is used to estimate the signal to noise ratio of each data flow correspondence;

Determining unit is used for according to described signal to noise ratio inquiry modulation coding mode MCS table to determine the information source speed of each data flow correspondence; And

Selected cell is used for selecting precoding weight vectors according to described information source speed.

The embodiment of the invention also provides a kind of system of selection of precoding weight vectors, may further comprise the steps:

Estimate the signal to noise ratio of each data flow correspondence;

According to described signal to noise ratio inquiry modulation coding mode MCS table, determine the information source speed of each data flow correspondence; And

Select precoding weight vectors according to described information source speed.

The embodiment of the invention has following useful effect: the capacity that can improve system channel to a certain extent.Simultaneously, the simple and effective weight vectors system of selection of the embodiment of the invention makes the UE end become more simple and direct, feasible in the process of selecting feedback weight vectors or matrix.In addition, by looking into the mode of MCS table, only relate to simple plus and minus calculation, thereby make UE terminal corresponding circuit design more simple, manufacturing and production are also easier.

Description of drawings

Fig. 1 is the structure chart of HSDPA transmitting terminal D-TxAA.

Fig. 2 is that existing scheme adopts shannon formula ratio juris figure.

Fig. 3 is the structured flowchart of the choice device of the embodiment of the invention one precoding weight vectors.

Fig. 4 is the schematic flow sheet of the system of selection of the embodiment of the invention two precoding weight vectors.

Fig. 5 is the schematic diagram of the embodiment of the invention two under the D-TxAA pattern.

Fig. 6 is the schematic diagram of the embodiment of the invention three under general MIMO pattern.

Embodiment

Below in conjunction with accompanying drawing the embodiment of the invention is described in detail.

The embodiment of the invention is mainly selected weight vectors according to modulation coding mode (MCS:Modulation Code Scheme) table.Embodiments of the invention are applicable under all MIMO patterns, to the selection of precoding weight.Only be that example describes below with WCDMA, other system is suitable for embodiments of the invention equally as long as use the MIMO pattern to transmit.In WCDMA, UE estimates the signal to noise ratio of each data flow correspondence according to public guide frequency information, and searches MCS according to signal to noise ratio and show to obtain the corresponding information source speed of each data flow.Select precoding weight vectors according to described information source speed again.UE can give Node B by the HS-DPCCH channel feedback with this precoding weight vectors indication.This method is called the MCS look-up table.

Please refer to shown in Figure 3ly, the choice device of the embodiment of the invention one precoding weight vectors comprises estimation unit 1, determining unit 2 and selected cell 3, and wherein estimation unit 1 is used to estimate the signal to noise ratio of each data flow correspondence; Determining unit 2 is used for according to described signal to noise ratio inquiry modulation coding mode MCS table to determine the information source speed of each data flow correspondence; Selected cell 3 is used for selecting precoding weight vectors according to described information source speed.This estimation unit 1 further comprises computing unit, is used for gathering the signal to noise ratio of calculating each data flow correspondence according to estimated channel matrix and weight vectors.In addition, the choice device of embodiment of the invention precoding weight vectors also comprises receiving element and transmitting element, and receiving element is used for receiving data stream, and data flow is input to described estimation unit 1; Transmitting element is used for the weight vectors selection result of selected cell 3 outputs is outwards sent.

During work, data flow enters estimation unit 1, and estimation unit 1 is estimated channel matrix H according to the information of Common Pilot Channel, and further estimates the signal to noise ratio of each data flow correspondence again in conjunction with the set of precoding weight vectors; Then, signal to noise ratio information enters determining unit 2, and inquiry MCS shows, and determines the information source speed of each data flow correspondence; Select the optimal weights vector by selected cell 3 according to certain rule at last.

Please more simultaneously in conjunction with shown in Fig. 4-5, the concrete step of the selection of the embodiment of the invention two precoding weight vectors is as follows:

Step S1, estimate the signal to noise ratio of each data flow correspondence.

At first, UE estimates channel matrix information H according to the information of Common Pilot Channel.For example, for the two transmitting antenna arrays under the D-TxAA mode, UE is according to two Common Pilot Channel (CPICH ₁And CPICH ₂) information estimation channel matrix H.

Then, UE obtains the signal to noise ratio δ with corresponding each data flow of precoding weight vectors set according to the set of precoding weight vectors and the channel matrix that obtained in the last step ²For example, under D-TxAA, the precoding weight set has W ₁, W ₂, W ₃And W ₄Totally 4 kinds of situations, channel matrix H is by calculating the signal to noise ratio δ of two data flow correspondences ₁ ²(H, W _i) and δ ₂ ²(H ₂, W _i), i=1,2,3,4.For the signal to noise ratio of two paths of data stream under 4 kinds of situations, as shown in table 2.

The noise of the different weight vectors set of table 2 correspondence is information source speed when

Step S2, look into MCS table, obtain the information source speed of all data flow according to the resulting signal to noise ratio of step S1.

For example, under the R7MIMO pattern, weight set W ₁, corresponding two the signal to noise ratio δ of two paths of data stream (primary traffic and secondary data stream) ₁ ²(H, W ₁) and δ ₂ ²(H, W ₁), according to MCS table (reference table 3), search corresponding with it information source speed, be assumed to be V ₁₁And V ₁₂Further, obtain the throughput V of two antennas ₁=V ₁₁+ V ₁₂Under other weight set, and the like, as shown in table 2.

Table 3MCS table

The signal to noise ratio snr of symbol level [dB]	Information source speed [Mb/s]	Modulation system
			-11.5	0.23	QPSK
-10.5	0.27	QPSK
			-9.5	0.36	QPSK
-8.5	0.45	QPSK
			-7.5	0.55	QPSK
-6.5	0.71	QPSK
			-5.5	0.90	QPSK
-4.5	1.09	QPSK
			-3.5	1.42	QPSK
-2.5	1.80	QPSK
			-1.5	2.18	QPSK
-0.5	2.56	QPSK
			0.5	3.17	QPSK
1.5	3.75	QPSK
			2.5	4.33	QPSK
3.5	4.92	16-QAM
			4.5	5.50	16-QAM
5.5	6.08	16-QAM
			6.5	6.66	16-QAM
7.5	7.24	16-QAM
			8.5	9.05	16-QAM
9.5	9.96	16-QAM
			10.5	10.87	16-QAM
11.5	11.77	16-QAM
			12.5	12.96	64-QAM
13.5	13.82	64-QAM
			14.5	15.34	64-QAM
15.5	16.31	64-QAM
			16.5	17.50	64-QAM
17.5	18.58	64-QAM
			18.5	19.44	64-QAM
19.5	20.30	64-QAM
			20.5	20.74	64-QAM
21.5	21.17	64-QAM
			22.5	21.60	64-QAM

Information source speed under step S3, the different weight situations of comparison, wherein the weight vectors of maximum correspondence is selected.Select the mode of weight to mainly contain in the embodiment of the invention:

(1) selects weight according to the maximum of two (or a plurality of) data flow information source speed sums;

For example, under the R7MIMO pattern, obtain and 4 information source speed sum V that weight vectors is corresponding ₁, V ₂, V ₃And V ₄After, relatively obtain maximum wherein, be V as maximum ₃, then with V ₃Corresponding weight vectors W ₃Be selected.

(2) select weight according to the maximum of the mean value of two (or a plurality of) data flow information source speed;

For example, under the R7MIMO pattern, weight set W ₁, W ₂, W ₃And W ₄If, weight set W ₁The information source speed V of corresponding two paths of data stream (primary traffic and secondary data stream) ₁₁, V ₁₂The mean value maximum, W then ₁Be selected.

(3) select weight according to the maximum in the minimum value in two (or a plurality of) data flow information source speed;

For example, under the R7MIMO pattern, compare respectively and 4 two paths of data stream information source speed that weight vectors is corresponding, obtain minimum value and be respectively V ₁₂, V ₂₂, V ₃₂, V ₄₂If, V wherein ₂₂Maximum, then its corresponding weight vectors W ₂Be selected.

The embodiment of the invention only relates to simple plus and minus calculation, thereby makes UE terminal corresponding circuit design more simple by looking into the mode of MCS table when comparing each data flow information source speed, manufacturing and production are also easier.

In addition, in above-mentioned steps S2, according to the signal to noise ratio δ that calculates gained ²When searching information source speed, if in the MCS table signal to noise ratio snr that equates is with it arranged, directly inquiry gets final product.If in the MCS table, do not have and δ ²The signal to noise ratio that equates then can be at first to δ ²After handling, search corresponding with it information source speed again.Preferably, can adopt following method:

Method one:

Use approximation method to signal to noise ratio δ ²Handle.Wherein, a kind of method is the δ that will obtain ²Signal to noise ratio less in the MCS table is approximate.For example, obtaining a signal to noise ratio numerical value is ${\mathrm{\&delta;}}_1^2(H,W_1)=5.9\mathrm{dB},$ By approximate processing be ${\mathrm{\&delta;}}_1^2(H,W_1)=5.5\mathrm{dB}.$ According to the MCS table, searching the information source speed corresponding with 5.5dB is 6.08Mb/s.

The advantage of this method is: UE selects the more convenient, quick of precoding weight vectors change.

Method two:

Select for use interpolation method that signal to noise ratio and information source speed in the MCS table are handled.According to certain step-length signal to noise ratio and information source speed in the MCS table are carried out interpolation arithmetic,, use after the single order interpolation as adopting the step-length of 0.2dB, ${\mathrm{\&delta;}}_1^2({\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="A20071007286800151.png"\; state="\{\{state\}\}">H</figure-callout>}}_2,W_1)=5.9\mathrm{dB}$ Corresponding information source speed is 6.294Mb/s.

The advantage of this method is: UE shows according to MCS, and the error of searching the corresponding information source speed of signal to noise ratio is less.

Method three:

The first step all is converted into linear numerical with the smnr data in the MCS table, obtains the mapping table of new signal to noise ratio and information source speed.As in the table 2, SNR=2.5dB can be converted into SNR=10 ^2.5/ ₁₀=1.778 with the corresponding relation of information source speed 4.33Mb/s.SNR under other situations transforms, and the like.Thereby obtain a new SNR and information source speed correspondence table.

In second step, SNR and information source speed correspondence table that the last step obtains are carried out interpolation, interpolation method similar approach two.

In the 3rd step, the signal to noise ratio of MCS table will be desired to look into ${\mathrm{\&delta;}}_1^2(H,W_1)=5.9\mathrm{dB}$ Be converted to linear numerical, obtain ${\mathrm{\&delta;}}_1^2(H,W_1)={10}^{\frac{5.9}{10}}=3.89.$ Obtain the maximum information source speed corresponding according to the table of the MCS after the interpolation with 3.89.

The advantage of this method is: UE shows according to MCS, and the error of searching the corresponding information source speed of signal to noise ratio is less.

The embodiment of the invention not only is applicable to existing D-TxAA transmission plan, is equally applicable to general mimo system.Again as shown in Figure 6, when transmitting terminal has M antenna, when the UE receiving terminal had N reception antenna, UE estimated channel matrix H according to public guide frequency information; Then, UE is according to the signal to noise ratio of H and different M data flow of weight combination calculation; Afterwards, obtain the information source speed of M data flow according to signal to noise ratio inquiry MCS table.UE selects precoding weight vectors according to described information source speed again.UE can give Node B by the HS-DPCCH channel feedback with this precoding weight vectors indication.

The embodiment of the invention has proposed a kind of effective weight system of selection at the problem of feeding back based on HS-DPCCH weight in the MIMO transmission plan of D-TxAA among the present 3GPP, can improve the capacity of system channel to a certain extent.Simultaneously, the embodiment of the invention makes the UE end become simple and direct, feasible in the process of selecting feedback weight vectors or matrix, and the manufacturing of UE terminal and production are easier.

Claims (11)

1. the choice device of a precoding weight vectors is characterized in that: comprising:

Estimation unit is used to estimate the signal to noise ratio of each data flow correspondence;

Determining unit is used for showing to determine the information source speed of each data flow correspondence according to the signal to noise ratio inquiry modulation coding mode MCS of described estimation unit output; And

Selected cell is used for selecting precoding weight vectors according to the information source speed of described determining unit output.

2. device according to claim 1 is characterized in that: also comprise:

Receiving element is used for receiving data stream, and data flow is input to described estimation unit; And

Transmitting element is used for the weight vectors selection result of selected cell output is outwards sent.

3. the system of selection of a precoding weight vectors may further comprise the steps:

Estimate the signal to noise ratio of each data flow correspondence;

According to described signal to noise ratio inquiry modulation coding mode MCS table, determine the information source speed of each data flow correspondence; And

Select precoding weight vectors according to described information source speed.

4. method according to claim 3 is characterized in that: described estimating step specifically comprises:

Information according to Common Pilot Channel is estimated channel matrix H; And

According to the set W of precoding weight vectors and the channel matrix H that in steps A 1, obtains, obtain signal to noise ratio δ with corresponding each data flow of precoding weight vectors set W ²

5. method according to claim 3 is characterized in that: describedly further comprise according to described signal to noise ratio inquiry modulation coding mode MCS table: if having the signal to noise ratio snr that equates with described signal to noise ratio in the MCS table, then directly inquire about; Otherwise earlier to δ ²Handle, again inquiry.

6. method according to claim 3 is characterized in that: described determining step specifically comprises:

With described signal to noise ratio δ ²Signal to noise ratio less in the MCS table is approximate; And

The inquiry information source speed corresponding in the MCS table with described approximate signal to noise ratio.

7. method according to claim 3 is characterized in that: described determining step specifically comprises:

According to certain step-length signal to noise ratio and information source speed in the MCS table are carried out interpolation arithmetic; And

Inquire about corresponding information source speed in MCS after the interpolation table again.

8. method according to claim 3 is characterized in that: described determining step specifically comprises:

Smnr data in the MCS table all is converted into linear numerical, obtains the mapping table of new signal to noise ratio and information source speed;

Described new mapping table is carried out interpolation; And

Desire is inquired about the signal to noise ratio δ of MCS table ²Be converted to linear numerical, inquire about the mapping table after the interpolation again, obtain corresponding maximum information source speed.

9. method according to claim 3 is characterized in that: described selection step specifically comprises: root

Maximum according to the information source speed sum of two or more data flow is selected weight vectors.

10. method according to claim 3 is characterized in that: described selection step specifically comprises:

Maximum according to the mean value of two or more data flow information source speed is selected weight vectors.

11. method according to claim 3 is characterized in that: described selection step specifically comprises: select weight vectors according to the maximum in the minimum value in two or more data flow information source speed.

Images