CN101355381A - Method and apparatus for scheduling and pre-encoding based on channel vector quantification - Google Patents
Method and apparatus for scheduling and pre-encoding based on channel vector quantification Download PDFInfo
- Publication number
- CN101355381A CN101355381A CNA2007101390739A CN200710139073A CN101355381A CN 101355381 A CN101355381 A CN 101355381A CN A2007101390739 A CNA2007101390739 A CN A2007101390739A CN 200710139073 A CN200710139073 A CN 200710139073A CN 101355381 A CN101355381 A CN 101355381A
- Authority
- CN
- China
- Prior art keywords
- information
- code book
- channel vector
- channel
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Radio Transmission System (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a dispatching and pre-coding method based on signal channel vector quantization and a device, wherein the method comprises: step A, a sending terminal does the multi-user dispatching according to first signal channel vector quantization information sent by all users; step B, after the sending terminal acquires the pre-coding matrix according to the second signal channel vector quantization information sent by dispatched users, and the dispatched user data are pre-coded by utilizing the acquired pre-coding matrix and sent, and the accuracy of the second signal channel vector quantization information is higher than that of the first signal channel vector quantization information; or step B', after the sending terminal combines the third signal channel vector quantization information and the first signal channel vector quantization information sent by dispatched users so as to acquire the pre-coding matrix, and the dispatched user data are pre-coded by utilizing the acquired pre-coding matrix and sent. The invention can achieve lower feedback cost under the condition of ensuring that the method of the invention has the same performance as the prior method, otherwise, the invention can achieve better system performance.
Description
Technical field
The present invention relates to the transmission technique field of wireless communication system, particularly a kind of scheduling and method for precoding and device that is applied in MIMO (Multi Input Multi Output, the multiple-input and multiple-output) system based on channel vector quantification.
Background technology
Future broadband wireless communication systems requires the more and more higher rate of information throughput and communication quality can be provided.In order to realize this goal on limited frequency spectrum resources, MIMO (Multi Input Multi Output, multiple-input and multiple-output) technology has become one of requisite means that adopted in the following radio communication.
In mimo system, transmitting terminal utilizes many antennas to carry out the transmission of signal, and receiving terminal utilizes many antennas to carry out the reception of signal.Studies show that than traditional single antenna transmissions method, the MIMO technology can improve channel capacity significantly, thereby improve the rate of information throughput.
Simultaneously, in mimo system, adopt the sending method of precoding (Pre-coding) can effectively improve the performance of mimo system.
The basic thought of precoding is according to current channel information, and data to be sent were carried out preliminary treatment before sending, and comprises linear process and Nonlinear Processing etc.
The MIMO precoding is a kind of transmission technology of closed loop, difference according to its required feedback information form, can be divided into two kinds, based on PVQ (Precoder Vector Quantization, precoding vector quantizes) precoding and based on the precoding of CVQ (Channel Vector Quantization, channel vector quantification).
In the MIMO precoding based on CVQ, transmitting terminal is set a channel quantitative code book in advance with receiving terminal before communicating by letter, wherein comprise some channel quantitative code words (Codeword).Each Resource Block (Resource Block, RB) send before, receiving terminal all can at first use this code book that the current characteristic of channel is quantized, and the result after will quantizing, promptly CVQ information feeds back to transmitting terminal.Particularly, from this code book, choose the code word of mating most exactly, and the index (Index) of this code word is fed back to transmitting terminal with current channel.
The purpose that the characteristic of channel is quantized is in order to reduce the feedback overhead of system.Next, after transmitting terminal can obtain current channel information according to feedback information, carry out multi-subscriber dispatching and calculate pre-coding matrix, and carry out corresponding signal according to this and send according to each user's channel information.
Present studies show that, compares with the method for precoding based on PVQ, can improve the system throughput performance based on the method for precoding of CVQ, reduces the implementation complexity of terminal simultaneously, is a kind of comparatively efficient precoding technology.In the existing scheduling and precoding solution based on CVQ, receiving end is carried out one time the CVQ feedback information to making a start, that is to say, make a start user scheduling with calculate pre-coding matrix in use identical CVQ feedback information, this scheme is with regard to system throughput performance or feedback overhead, do not accomplish optimum, therefore be necessary on the basis of existing technology, further optimization system throughput performance and feedback overhead.
Summary of the invention
The purpose of this invention is to provide a kind of scheduling and method for precoding and device based on channel vector quantification, perhaps the optimization system throughput performance perhaps reduces feedback overhead.
To achieve these goals, the invention provides a kind of scheduling and method for precoding, comprising based on channel vector quantification:
Steps A, transmitting terminal carries out multi-subscriber dispatching according to the first channel vector quantification information that all users send; With
Step B, after the second channel vector quantization information that transmitting terminal sends according to the user who is scheduled for is obtained pre-coding matrix, the pre-coding matrix that use obtains is scheduled for user's data to carry out sending after the precoding, and the precision of described second channel vector quantization information is higher than the precision of the first channel vector quantification information; Or
Step B ' after the 3rd channel vector quantification information that the user that the transmitting terminal combination is scheduled for sends and the first channel vector quantification information are obtained pre-coding matrix, uses the pre-coding matrix that obtains to be scheduled for user's data and carries out sending after the precoding.
Above-mentioned method, wherein, among the described step B, the control of the precision of described channel vector quantification information realizes that by code book the number of code word is big more in the code book, and channel directional information feedback overhead is big more, and described channel vector quantification precision of information is high more.
Above-mentioned method, wherein,
The first channel vector quantification information in the described steps A obtains according to the current channel and first code book, and the second channel vector quantization information among the described step B obtains according to the current channel and second code book;
Number of codewords in described second code book is greater than the number of codewords in described first code book.
Above-mentioned method, wherein, described second code book is the resulting code book of the first code book premultiplication unitary matrice.
Above-mentioned method, wherein, among the described step B ', the first channel vector quantification information in the described steps A obtains according to the current channel and first code book, and the 3rd channel vector quantification information among the described step B obtains according to current channel and the 3rd code book;
Described the 3rd code book is the difference code book, is used to quantize the difference of the current channel and the first channel vector quantification information.
Above-mentioned method, wherein, described difference code book is that code word is the code book of unitary matrice.
Above-mentioned method, wherein, the code word of channel directional information index correspondence is a unitary matrice in described the 3rd channel vector quantification information, among the described step B ', utilize the product of unitary matrice of the index correspondence of pairing code word of index that the mould of the pairing code word of index that the user be scheduled for feeds back for the first time, the user that is scheduled for feed back for the first time and the user feedback that is scheduled for to calculate pre-coding matrix.
Above-mentioned method, wherein, the second channel vector quantization information that the described user who is scheduled for sends only comprises the channel directional information, uses the channel quality information in the first channel vector quantification information to calculate pre-coding matrix among described step B and the B '.
Above-mentioned method, wherein, the order of the feedback information that the user that transmitting terminal utilizes information due in, identification information or information position judgment to be scheduled for sends.
In order better to realize above-mentioned purpose, the present invention also provides a kind of scheduling and pre-coding apparatus based on channel vector quantification, comprises scheduler module and precoding module, wherein:
Described scheduler module is used for carrying out multi-subscriber dispatching according to the first channel vector quantification information that all users send;
After described precoding module is used for obtaining pre-coding matrix according to the second channel vector quantization information that the user that is scheduled for sends, the pre-coding matrix that use obtains is scheduled for user's data to carry out sending after the precoding, and the precision of described second channel vector quantization information is higher than the precision of the first channel vector quantification information; Or after being used for the 3rd channel vector quantification information that sends in conjunction with the user that is scheduled for and the first channel vector quantification information and obtaining pre-coding matrix, use the pre-coding matrix that obtains to be scheduled for user's data and carries out sending after the precoding.
Above-mentioned device, wherein, when described precoding module is used for obtaining pre-coding matrix according to the second channel vector quantization information that the user that is scheduled for sends, the described first channel vector quantification information obtains according to the current channel and first code book, and described second channel vector quantization information obtains according to second code book;
Number of codewords in described second code book is greater than the number of codewords in described first code book.
Above-mentioned device, wherein, described second code book is the resulting code book of the first code book premultiplication unitary matrice.
Above-mentioned device, wherein, described precoding module is used for the 3rd channel vector quantification information that sends in conjunction with the user that is scheduled for and the first channel vector quantification information when obtaining pre-coding matrix, the described first channel vector quantification information obtains according to the current channel and first code book, described the 3rd channel vector quantification information obtains according to current channel and the 3rd code book, described the 3rd code book is the difference code book, is used to quantize the difference of the current channel and the first channel vector quantification information.
Above-mentioned device, wherein, described difference code book is that code word is the code book of unitary matrice.
Above-mentioned device, wherein, the code word of channel directional information index correspondence is a unitary matrice in described the 3rd channel vector quantification information, and described precoding module specifically is used to utilize the product of unitary matrice of the index correspondence of pairing code word of index that the mould of the pairing code word of index that the user that is scheduled for feeds back for the first time, the user that is scheduled for feed back for the first time and the user feedback that is scheduled for to calculate pre-coding matrix.
Utilize method and apparatus of the present invention,, therefore, compare, have following beneficial effect with existing method based on the CVQ information that different accuracy is used in the scheduling and the precoding of channel vector quantification:
When using the CVQ precision identical during precoding of the present invention with the method that has now, can guarantee obtain the systematic function identical with existing method, but because the CVQ precision of using during scheduling can be lower than the CVQ precision of the method for having now, therefore, in general, can reduce feedback overhead;
Under the overhead situation identical with existing method, because scheduling phase uses the CVQ information of low precision, can economize out the part feedback overhead, and this part feedback overhead is used for improving in the precoding stage precision of CVQ information, so can obtain the systematic function higher than existing method.
Description of drawings
Fig. 1 be in actual channel and the code book code word concern schematic diagram;
Fig. 2 is the schematic flow sheet of the inventive method first embodiment;
Fig. 3 is the schematic flow sheet of the inventive method second embodiment;
Fig. 4 is the simulation result schematic diagram of the inventive method second embodiment.
Embodiment
When transmitting terminal carries out based on the scheduling of channel vector quantification and precoding, mainly comprise two steps:
The multi-subscriber dispatching step is carried out multi-subscriber dispatching according to the CVQ information of receiving terminal;
The precoding step is carried out precoding according to the CVQ information of receiving terminal to the user who is dispatched to.
In two above-mentioned steps, all need to use CVQ information, according to the real network transmission as can be known, the precision of CVQ information is very big to the influence of systematic function, and the precision of CVQ information is high more, and feedback overhead is big more, but the system throughput performance is good more.
Below the CVQ precision is further elaborated.
CVQ comprises CDI (Channel Direction Information, the channel directional information) and CQI (Channel Quality Indicator, channel quality information), wherein the size of the feedback overhead of CQI can not brought tangible influence to systematic function, do not consider at this, therefore when considering the CVQ precision, only consider that CDI is to its influence that brings.
As everyone knows, CDI characterizes the position of code word in code book of selecting to be used for representing the actual channel direction, the expense of CDI is big more, code word in the expression code book can be many more, and code word is many more in the code book, the code word and the actual channel that choose expression actual channel direction are approaching more, so the CVQ precision is high more.
Illustrate as follows to above-mentioned situation.
As shown in Figure 1, two code book C1={A1 have wherein been represented, A2, A3, A4} and C2={B1, B2, B3, B4, B5, B6, B7, code word among the B8} and actual channel R.
When the feedback overhead of CDI was 2 bits, then CDI can only represent in 4 code words one, and this moment, corresponding code book was C1, can find from Fig. 1, and the CDI of receiving terminal feedback is 00, and the code word of expression CDI correspondence is A1;
When the feedback overhead of CDI was 3 bits, then CDI can represent in 8 code words one, and this moment, corresponding code book was C2, can find from Fig. 1, and the CDI of receiving terminal feedback is 001, and the code word of expression CDI correspondence is B1;
Can find clearly that from Fig. 1 when the CDI feedback overhead was 3 bits, the code word B1 of selection more approached actual channel R with respect to code word A1, that is to say that corresponding CVQ precision is higher.
Therefore, the parameter that characterizes the CVQ precision among the present invention is the number of the equivalent code book code word calculated of the feedback overhead of CDI or finally be used for.
Simultaneously, can find by above-mentioned description, two steps of multi-subscriber dispatching and precoding all relate to CVQ, and can find by further emulation, when transmitting terminal carried out based on the scheduling of channel vector quantification and precoding, in the multi-subscriber dispatching step, the variation of CVQ precision can not cause tangible influence to systematic function, and in the precoding step, the variation of CVQ precision causes very significantly influence to systematic function.
Based on above-mentioned description, scheduling and method for precoding based on channel vector quantification of the present invention use the CVQ information of different accuracy to improve systematic function and/or reduction feedback overhead to multi-subscriber dispatching step and precoding step.
And the CVQ information of employed different accuracy can be brought in processing by receiving terminal or transmission in multi-subscriber dispatching step and the precoding step, is described by different embodiment respectively below.
＜the first embodiment 〉
In the first embodiment of the present invention, the precision of employed CVQ information is directly controlled by the code book of the different number of codewords of transmitting terminal utilization in multi-subscriber dispatching step and the precoding step, is elaborated below.
As shown in Figure 2, the method for first embodiment of the invention comprises the steps:
Wherein, step 11 specifically comprises:
Step 111, channel estimating;
User k estimates current channel, obtains characteristic of channel matrix H
k, it is n
RRow n
TRow, wherein n
RAnd n
TBe respectively reception antenna and number of transmit antennas.
Step 112 receives weighting;
The purpose that receives weighting is that the mimo channel matrix is converted into vector form, thereby can more effectively carry out channel quantitative, and equivalent channel is after the reception weighting:
h
k＝v
kH
k
V wherein
kBe 1 * n
RWeighing vector, it can obtain by existing method, as the signal to noise ratio (snr) maximization approach.
Step 113, channel information quantize and feedback;
Existing channel quantitative method has two kinds, a kind of quantification that is based on the amplitude maximum, and another kind is based on the maximum quantification of Signal to Interference plus Noise Ratio (SINR).Do not adopt which kind of quantization method in the real system because the inventive method relies on, therefore only with first kind of quantization method, promptly the example that is quantified as based on the amplitude maximum describes.
Suppose that the code book of setting in advance that is used for channel quantitative is C1={f
1, f
2..., f
N, wherein, f
iBe i code word in the code book, it is 1 * n
TVector, N is the size of code book, the number of the code word that promptly comprises.So, the CVQ result of equivalent channel after the weighting after quantizing comprises two parts information: CDI (Channel Directional Information, the channel direction signal) and CQI (Channel QualityIndicator, channel quality information), wherein CDI is:
Wherein:
* for gripping the transposition symbol altogether, || || be the mould of vector.In addition, the result of CQI is || h
k||
2
Next, receiving terminal can be with the result after quantizing, and promptly CDI and CQI feed back to and make a start, and wherein only needs its index information corresponding in code book of feedback to get final product concerning CDI.
Wherein, carry out multi-subscriber dispatching according to a CQV information of feedback in the step 12,,, therefore, need to calculate pre-coding matrix in this step because multi-subscriber dispatching need be used pre-coding matrix at this.
Because it is multiple that multi-user dispatching method has, such as traversal search and Greedy method etc., it is multiple that the method for precoding also has, such as ZF (Zero Forcing, ZF) algorithm and MMSE (MinimumMean Square Error, least mean-square error) algorithm etc.
Because the inventive method does not rely on concrete which kind of scheduling and the method for precoding of using in the system, therefore we are that example describes with traversal scheduling and ZF method for precoding only here.
Suppose certain user's combination { k
1, k
2... } and scheduled, k wherein
iBe user index;
According to the CVQ information calculations ZF pre-coding matrix W at this moment of user feedback, as follows:
Wherein
The pseudoinverse of representing matrix H, wherein:
T is the matrix transpose symbol.
Calculate the SINR on each user this moment, as follows:
Wherein, ρ and σ
2Be respectively transmitted power and noise power on each stream.Thus, can obtain this moment the user handle up and, as follows:
∑R(SINR
i)
Wherein R () is for being calculated the function of handling up by SINR;
By the possible user's combination of above step traversal institute, dispatched users is handled up and maximum user's combination then.
In the prior art, scheduling of making a start and calculating pre-coding matrix are operated here and just are through with simultaneously, next make a start and will notify the receiving end scheduling result, and the user in scheduling adopts corresponding pre-coding matrix W to carry out the signal transmission.
But the scheduling result that method of the present invention only utilizes above-mentioned steps to obtain, and with its notice receiving terminal.
The user who is scheduled for who receives the transmitting terminal notice carries out the CQV feedback information second time to transmitting terminal.
Suppose that the code book of setting in advance that is used for channel quantitative is C2={g
1, g
2..., g
M, wherein, g
iBe i code word in the code book, wherein, the number of codewords M among the code book C2 is greater than the quantity N of code word among the employed code book C1 of scheduling.
Certainly, in the first above-mentioned embodiment, the 2nd CVQ information with the first time CVQ information compare, can only comprise CDI in the 2nd CVQ information of receiving terminal feedback, and not feeding back CQI, transmitting terminal utilizes the CQI in the CVQ information to get final product when calculating pre-coding matrix.
The CVQ result of equivalent channel after the weighting after quantizing comprises two parts information: CDI (ChannelDirectional Information, channel direction signal) and CQI (Channel Quality Indicator, channel quality information), and wherein CDI is:
Wherein:
* for gripping the transposition symbol altogether, || || be the mould of vector.In addition, the result of CQI is || h
k||
2
According to the CVQ information calculations pre-coding matrix W at this moment of user k1, k2... feedback, following descending:
At last, utilizing this pre-coding matrix W that the data of the user k1, the k2... that are dispatched to are carried out sending after the precoding gets final product.
In the first embodiment of the present invention, because the difference of number of codewords in the code book that transmitting terminal uses, the precision of calculating the employed CVQ information of pre-coding matrix is higher than the precision of dispatching the employed CVQ information of step and is lower than.
Under first kind of situation, suppose that number of users is 16, the number of users that is scheduled for is 2, at this, the feedback overhead of supposing CDI in the CVQ information of conventional method is 4 bits, and utilizing method of the present invention, CDI is 2 bits in the CVQ information of supposing to feed back for the first time, and CDI is 4 bits in the CVQ information of feeding back for the second time.
Under the situation of above-mentioned hypothesis, according to the description of front as can be known, since be used in the method for the present invention calculating the CDI of pre-coding matrix and conventional method CVQ information CDI be 4 bits, the CVQ precision of promptly calculating pre-coding matrix is identical, therefore the systematic function that reaches of two kinds of methods is basic identical, yet, yet utilize the inventive method, total feedback overhead of CDI is 16 * 2+2 * 4 (40) bits, much smaller than total CDI feedback overhead of 16 * 4 (64) bits of conventional method.
Under second kind of situation, suppose that number of users is 16, the number of users that is scheduled for is 2, at this, the feedback overhead of supposing CDI in the CVQ information of conventional method is 4 bits, and utilizing method of the present invention, CDI is 2 bits in the CVQ information of supposing to feed back for the first time, and CDI is 16 bits in the CVQ information of feeding back for the second time.
Under the situation of above-mentioned hypothesis, the total CDI feedback overhead that utilizes method of the present invention is 16 * 2+2 * 16 (64) bits, and is identical with total CDI feedback overhead 16 * 4 (64) bits that utilize conventional method.
But because in the method for the present invention, the bit number that is used for calculating the CVQ information CDI of pre-coding matrix is 16,4 bits of its precision in the conventional method, lifting has very significantly influence and because the variation of CVQ precision is to systematic function, therefore, utilize method of the present invention guaranteeing significantly to improve systematic function under the situation that feedback overhead does not increase.
Certainly, under second kind of situation, if suppose CDI in the CVQ information for the second time feedback overhead be 8 bits, the total feedback overhead of CDI that then utilizes method of the present invention is 16 * 2+2 * 8 (48) bits, less than CDI feedback overhead 64 bits of conventional method, simultaneously, because being used for calculating the bit number of the CVQ information CDI of pre-coding matrix in the method for the present invention is 8, its precision is greater than the CDI of 4 bits in the conventional method, and therefore, systematic function also can be improved.
＜the second embodiment 〉
In the second embodiment of the present invention, the precision of employed CVQ information is directly controlled by the code book of the different number of codewords of transmitting terminal utilization in multi-subscriber dispatching step and the precoding step.
In the first embodiment of the present invention, separate between second code book and first code book, the number of codewords that only needs to guarantee second code book gets final product greater than first code book.
And in the second embodiment of the present invention, further construct second code book according to first code book, the number of codewords of second code book that assurance simultaneously obtains is greater than the number of codewords of first code book.
At this, the method that this second code book C2 utilizes the first code book C1 to obtain is as described below.
Each code word in first code book is carried out difference processing respectively obtain a plurality of code words, and, illustrate as follows the code word of these a plurality of code words that obtain as second code book.
Suppose that first code book comprises 4 code word A, B, C and D, is used for independent quantized channel;
Respectively code word A, B, C and D being carried out difference processing obtains:
A1、A2、A3、A4；
B1、B2、B3、B4；
C1、C2、C3、C4；
D1、D2、D3、D4。
Therefore second code book that obtains comprises these 16 code words of A1, A2, A3, A4, B1, B2, B3, B4, C1, C2, C3, C4, D1, D2, D3 and D4, and therefore, the CVQ precision improves.
Certainly, this second code book also can be to comprise:
These 7 code words of A1, A2, A3, A4, B, C and D; Or
These 7 code words of A, B1, B2, B3, B4, C and D; Or
These 7 various situations such as code word of A, B, C1, C2, C3, C4 and D.
Certainly, this second code book also can be to comprise:
These 6 code words of A1, A2, A3, B, C and D; Or
These 6 various situations such as code word of A, B1, B2, B3, C and D.
Certainly, second code book can also be other combination, does not enumerate one by one at this.
At this, this difference processing can obtain by code word premultiplication unitary matrice.
As g
nA (n=1,2,3 ..., 8) then the code word difference is become 8 code words.
Because the second embodiment of the present invention and the difference of first embodiment only are the different of code word design in second code book, other handling processes are identical, do not repeat them here.
＜the three embodiment 〉
In above-mentioned first embodiment and second embodiment, the user directly utilizes the different code book of number of codewords to carry out feedback processing, and therefore, transmitting terminal can directly obtain corresponding pre-coding matrix according to user's feedback.
And the third embodiment of the present invention is different from first embodiment and second embodiment, in the 3rd embodiment, after the scheduling step, the receiving terminal that is scheduled for sends the 3rd CVQ information, the 3rd CVQ feedback information is behind transmitting terminal, and transmitting terminal needs to carry out precoding processing in conjunction with being scheduled for the 3rd CVQ information and the CVQ information that the user feeds back.
As shown in Figure 3, the method for third embodiment of the invention comprises the steps:
Wherein, step 21 specifically comprises:
Step 211, channel estimating;
User k estimates current channel, obtains characteristic of channel matrix H
k, it is n
RRow n
TRow, wherein n
RAnd n
TBe respectively reception antenna and number of transmit antennas.
Step 212 receives weighting;
The purpose that receives weighting is that the mimo channel matrix is converted into vector form, thereby can more effectively carry out channel quantitative, and equivalent channel is after the reception weighting:
h
k＝v
kH
k
V wherein
kBe 1 * n
RWeighing vector, it can obtain by existing method, as Signal to Interference plus Noise Ratio (SNR) maximization approach.
Step 213, channel information quantize and feedback;
Existing channel quantitative method has two kinds, a kind of quantification that is based on the amplitude maximum, and another kind is based on the maximum quantification of Signal to Interference plus Noise Ratio (SINR).Not adopting which kind of quantization method in the real system because the inventive method relies on, is that example describes with first kind of quantization method only therefore.
Suppose that the code book of setting in advance that is used for channel quantitative is C={f
1, f
2..., f
N, wherein, f
iBe i code word in the code book, it is 1 * n
TVector, N is the size of code book, the number of the code word that promptly comprises.So, the CVQ result of equivalent channel after the weighting after quantizing comprises two parts information: CDI (Channel Directional Information, the channel direction signal) and CQI (Channel QualityIndicator, channel quality information), wherein CDI is:
Wherein:
* for gripping the transposition symbol altogether, || || be the mould of vector.In addition, the result of CQI is || h
k||
2
Next, receiving terminal can be with the result after quantizing, and promptly CDI and CQI feed back to and make a start, and wherein only needs its index information corresponding in code book of feedback to get final product concerning CDI.
Wherein, carry out multi-subscriber dispatching according to a CQV information of feedback in the step 22, at this, therefore this multi-subscriber dispatching, needs to calculate pre-coding matrix owing to need use pre-coding matrix in this step.
Because it is multiple that multi-user dispatching method has, such as traversal search and Greedy method etc., it is multiple that the method for precoding also has, such as ZF (Zero Forcing, ZF) algorithm and MMSE (MinimumMean Square Error, least mean-square error) algorithm etc.
Because the inventive method does not rely on concrete which kind of scheduling and the method for precoding of using in the system, therefore we are that example describes with traversal scheduling and ZF method for precoding only here.
Suppose certain user's combination { k
1, k
2... } and scheduled, k wherein
iBe user index;
According to the CVQ information calculations ZF pre-coding matrix W at this moment of user feedback, as follows:
Wherein
The pseudoinverse of representing matrix H, wherein:
T is the matrix transpose symbol.
Calculate the SINR on each user this moment, as follows:
Wherein, ρ and σ
2Be respectively transmitted power and noise power on each stream.Thus, can obtain this moment the user handle up and, as follows:
∑R(SINR
i)
Wherein R () is for being calculated the function of handling up by SINR;
By the possible user's combination of above step traversal institute, dispatched users is handled up and maximum user's combination then.
In the prior art, scheduling of making a start and calculating pre-coding matrix are operated here and just are through with simultaneously, next make a start and will notify the receiving end scheduling result, and the user in scheduling adopts corresponding pre-coding matrix W to carry out the signal transmission.
But the scheduling result that method of the present invention only utilizes above-mentioned steps to obtain, and with its notice receiving terminal.
The user who is scheduled for who receives the transmitting terminal notice carries out the CQV feedback information second time to transmitting terminal.
Certainly, in the second above-mentioned embodiment, can only comprise CDI in the 2nd CVQ information of receiving terminal feedback, and not feed back CQI, transmitting terminal utilizes the CQI in the CVQ information to get final product when calculating pre-coding matrix.
The difference of the third embodiment of the present invention and first embodiment is that among first embodiment, the 2nd CVQ only needs to guarantee the precision of its precision greater than a CVQ, but in the 3rd embodiment, pre-coding matrix calculates need take all factors into consideration a CVQ.
Suppose that user's equivalent channel is h
k, the CDI after the first step quantizes is
Employing is based on the method for amplitude quantizing, and the CDI that so here carries out after second step quantized is:
Wherein:
At this, the quantification code book is C
2={ g
1, g
2..., g
N.Here, quantize difference, at this moment each the code word g in the code book with the first step
iBe unitary matrice, and primary CDI feedback has been considered in the calculating of cos θ.
Next, the user in the scheduling will the CDI after quantizing for the second time as a result the index of the code word of correspondence feed back to and make a start.
After transmitting terminal received the channel information of user's second step feedback, meeting obtained according to the channel information of the reconstruct reality as a result of two step feedbacks:
Then, recomputate pre-coding matrix according to scheduling result and corresponding subscriber channel information.Particularly, suppose user's combination { k
1, k
2... scheduled, its equivalent channel is
Ci Shi pre-coding matrix is so
Next, transmitting terminal just the user in scheduling adopt corresponding pre-coding matrix W to carry out signal to send.
Scheduling and pre-coding apparatus based on channel vector quantification of the present invention comprise scheduler module and precoding module, wherein:
Described scheduler module is used for carrying out multi-subscriber dispatching according to all users' the first channel vector quantification information;
Described precoding module is used the pre-coding matrix that obtains to be scheduled for user's data and is carried out sending after the precoding after being used for obtaining pre-coding matrix according to the user's that is scheduled for second channel vector quantization information;
The precision of second channel vector quantization information is higher than the precision of the first channel vector quantification information.
The described first channel vector quantification information obtains according to first code book, and described second channel vector quantization information obtains according to second code book;
Number of codewords in described second code book is greater than the number of codewords in described first code book.
At least two code words in this second code book are obtained by a code word difference processing in first code book.
This difference processing can be the operation of premultiplication unitary matrice.
In the method and apparatus of the present invention, the user who is scheduled for need carry out twice feedback, therefore, exist transmitting terminal how to distinguish that this feedback belongs to the feedback information that is used to dispatch or is used for the problem of the feedback information of precoding, the present invention uses following method to distinguish:
1, the user adds the order identification parameter in feedback information, and transmitting terminal is differentiated this feedback information according to this order identification parameter and is used for scheduling or is used for precoding; Or
2, when twice feedback information has significant difference a time of advent, can utilize time of reception to distinguish;
3, differentiate according to the position of feedback information, be positioned over the primary importance of Frame as the feedback information that will be used to dispatch, and the feedback information that will be used to dispatch is positioned over the second place of Frame, and transmitting terminal can be judged according to the residing position of feedback information.
In the emulation to the second embodiment of the present invention, simulated conditions is as follows:
n
T=2, n
R=1, channel is a flat fading channel, and number of users is 16, and dispatching algorithm adopts the MaxC/I scheduling, and the speed of a motor vehicle is 3km/h, and feedback delay is 3ms, and wherein B is the bit number of CDI feedback.
In the emulation, for fairness relatively, conventional method has identical feedback overhead with sending method.Such as, during conventional method B=3, always be fed back to 16 * 3=48 bit, and utilize the method for invention, 2 bits of each feedback of 16 users of the first step, two scheduled users feed back 8 bits again in second step, and 16 * 2+2 * 8=48 is arranged this moment.In addition, traditional B=4 also have identical total feedback overhead with the method for invention.
Fig. 4 is the simulation result schematic diagram, as can be seen from Figure 4, utilizes method of the present invention can obtain better throughput performance under the situation that does not increase feedback overhead.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (15)
1. scheduling and method for precoding based on a channel vector quantification is characterized in that, comprising:
Steps A, transmitting terminal carries out multi-subscriber dispatching according to the first channel vector quantification information that all users send; With
Step B, after the second channel vector quantization information that transmitting terminal sends according to the user who is scheduled for is obtained pre-coding matrix, the pre-coding matrix that use obtains is scheduled for user's data to carry out sending after the precoding, and the precision of described second channel vector quantization information is higher than the precision of the first channel vector quantification information; Or
Step B ' after the 3rd channel vector quantification information that the user that the transmitting terminal combination is scheduled for sends and the first channel vector quantification information are obtained pre-coding matrix, uses the pre-coding matrix that obtains to be scheduled for user's data and carries out sending after the precoding.
2. method according to claim 1 is characterized in that, among the described step B, the precision control of described channel vector quantification information realizes by code book, the number of code word is big more in the code book, and channel directional information feedback overhead is big more, and described channel vector quantification precision of information is high more.
3. method according to claim 2 is characterized in that:
The first channel vector quantification information in the described steps A obtains according to the current channel and first code book, and the second channel vector quantization information among the described step B obtains according to the current channel and second code book;
Number of codewords in described second code book is greater than the number of codewords in described first code book.
4. method according to claim 3 is characterized in that, described second code book is the resulting code book of the first code book premultiplication unitary matrice.
5. method according to claim 1, it is characterized in that, among the described step B ', the first channel vector quantification information in the described steps A obtains according to the current channel and first code book, and the 3rd channel vector quantification information among the described step B obtains according to current channel and the 3rd code book;
Described the 3rd code book is the difference code book, is used to quantize the difference of the current channel and the first channel vector quantification information.
6. method according to claim 5 is characterized in that, described difference code book is that code word is the code book of unitary matrice.
7. method according to claim 6, it is characterized in that, the code word of channel directional information index correspondence is a unitary matrice in described the 3rd channel vector quantification information, among the described step B ', utilize the product of unitary matrice of the index correspondence of pairing code word of index that the mould of the pairing code word of index that the user be scheduled for feeds back for the first time, the user that is scheduled for feed back for the first time and the user feedback that is scheduled for to calculate pre-coding matrix.
8. according to any described method of claim in the claim 1 to 7, it is characterized in that, the second channel vector quantization information that the described user who is scheduled for sends only comprises the channel directional information, uses the channel quality information in the first channel vector quantification information to calculate pre-coding matrix among described step B and the B '.
9. according to any described method of claim in the claim 1 to 7, it is characterized in that the order of the feedback information that the user that transmitting terminal utilizes information due in, identification information or information position judgment to be scheduled for sends.
10. scheduling and pre-coding apparatus based on a channel vector quantification comprise scheduler module and precoding module, it is characterized in that:
Described scheduler module is used for carrying out multi-subscriber dispatching according to the first channel vector quantification information that all users send;
After described precoding module is used for obtaining pre-coding matrix according to the second channel vector quantization information that the user that is scheduled for sends, the pre-coding matrix that use obtains is scheduled for user's data to carry out sending after the precoding, and the precision of described second channel vector quantization information is higher than the precision of the first channel vector quantification information; Or after being used for the 3rd channel vector quantification information that sends in conjunction with the user that is scheduled for and the first channel vector quantification information and obtaining pre-coding matrix, use the pre-coding matrix that obtains to be scheduled for user's data and carries out sending after the precoding.
11. device according to claim 10, it is characterized in that, when described precoding module is used for obtaining pre-coding matrix according to the second channel vector quantization information that the user that is scheduled for sends, the described first channel vector quantification information obtains according to the current channel and first code book, and described second channel vector quantization information obtains according to second code book;
Number of codewords in described second code book is greater than the number of codewords in described first code book.
12. device according to claim 11 is characterized in that, described second code book is the resulting code book of the first code book premultiplication unitary matrice.
13. device according to claim 10, it is characterized in that, described precoding module is used for the 3rd channel vector quantification information that sends in conjunction with the user that is scheduled for and the first channel vector quantification information when obtaining pre-coding matrix, the described first channel vector quantification information obtains according to the current channel and first code book, described the 3rd channel vector quantification information obtains according to current channel and the 3rd code book, described the 3rd code book is the difference code book, is used to quantize the difference of the current channel and the first channel vector quantification information.
14. device according to claim 13 is characterized in that, described difference code book is that code word is the code book of unitary matrice.
15. device according to claim 14, it is characterized in that, the code word of channel directional information index correspondence is a unitary matrice in described the 3rd channel vector quantification information, and described precoding module specifically is used to utilize the product of unitary matrice of the index correspondence of pairing code word of index that the mould of the pairing code word of index that the user that is scheduled for feeds back for the first time, the user that is scheduled for feed back for the first time and the user feedback that is scheduled for to calculate pre-coding matrix.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101390739A CN101355381A (en) | 2007-07-24 | 2007-07-24 | Method and apparatus for scheduling and pre-encoding based on channel vector quantification |
| JP2008189838A JP5172517B2 (en) | 2007-07-24 | 2008-07-23 | Scheduling and precoding method and apparatus based on channel vector quantization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101390739A CN101355381A (en) | 2007-07-24 | 2007-07-24 | Method and apparatus for scheduling and pre-encoding based on channel vector quantification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101355381A true CN101355381A (en) | 2009-01-28 |
Family
ID=40308000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007101390739A Pending CN101355381A (en) | 2007-07-24 | 2007-07-24 | Method and apparatus for scheduling and pre-encoding based on channel vector quantification |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP5172517B2 (en) |
| CN (1) | CN101355381A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101989870A (en) * | 2009-08-05 | 2011-03-23 | 株式会社Ntt都科摩 | Method for acquiring channel quality indication information and base station thereof |
| CN102237950A (en) * | 2010-04-29 | 2011-11-09 | 株式会社Ntt都科摩 | User equipment, base station and channel information feedback method |
| CN102460996A (en) * | 2009-04-29 | 2012-05-16 | 英特尔公司 | Differential feedback scheme for closed-loop MIMO beamforming |
| CN102916788A (en) * | 2012-07-30 | 2013-02-06 | 东南大学 | Self-adaptation bit feedback method for multi-cell cooperative communication |
| CN104202130A (en) * | 2009-07-06 | 2014-12-10 | 英特尔公司 | Beamforming using base and differential codebooks |
| CN106330271A (en) * | 2015-06-19 | 2017-01-11 | 中国移动通信集团公司 | Multi-antenna multi-user non-orthogonal transmission method, base station, blind detection method, terminal |
| CN107005579A (en) * | 2014-09-26 | 2017-08-01 | 华为技术有限公司 | For the feedback CQI for determining UE grades method, base station and UE |
| CN111788785A (en) * | 2018-01-29 | 2020-10-16 | 华为技术有限公司 | Precoding matrix index reporting method, communication device and medium |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2219298A1 (en) * | 2009-02-12 | 2010-08-18 | NTT DoCoMo, Inc. | Method and apparatus for determining a quantized channel vector |
| WO2010148554A1 (en) * | 2009-06-23 | 2010-12-29 | 上海贝尔股份有限公司 | Method and device for signal transmission in time-division duplex mimo system |
| WO2011021261A1 (en) * | 2009-08-17 | 2011-02-24 | 株式会社日立製作所 | Wireless base station, wireless terminal, wireless communication system, and wireless communication method |
| US9137802B2 (en) | 2009-08-18 | 2015-09-15 | Qualcomm Incorporated | DL MU-MIMO operation in LTE-A |
| EP2343837A1 (en) | 2010-01-08 | 2011-07-13 | Alcatel Lucent | Channel state information feedback using a hierarchical codebook for vector quantisation |
| GB2479377B (en) * | 2010-04-07 | 2013-08-14 | Toshiba Res Europ Ltd | Dual indicator scheme for channel state information feedback |
| WO2011131210A1 (en) * | 2010-04-20 | 2011-10-27 | Telecom Italia S.P.A. | Method and system for wireless communications, corresponding network and computer |
| CN102291216A (en) * | 2010-06-17 | 2011-12-21 | 株式会社Ntt都科摩 | User equipment and channel state information feedback method |
| CN102790637A (en) * | 2011-05-16 | 2012-11-21 | 株式会社Ntt都科摩 | Reconstruction method, precoding method and device for communication channel |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101124932B1 (en) * | 2005-05-30 | 2012-03-28 | 삼성전자주식회사 | Apparatus and method for transmitting/receiving a data in mobile communication system with array antennas |
| US9136974B2 (en) * | 2005-08-30 | 2015-09-15 | Qualcomm Incorporated | Precoding and SDMA support |
| US8760994B2 (en) * | 2005-10-28 | 2014-06-24 | Qualcomm Incorporated | Unitary precoding based on randomized FFT matrices |
| KR100996023B1 (en) * | 2005-10-31 | 2010-11-22 | 삼성전자주식회사 | Apparatus and method for data transmission and reception in a multi-antenna communication system |
| CN101166052B (en) * | 2006-10-19 | 2012-05-23 | 株式会社Ntt都科摩 | Precoding method of multi-input multi-output system and equipment using same |
| CN101652936B (en) * | 2007-03-30 | 2013-05-08 | 松下电器产业株式会社 | Radio communication system, radio communication device, and radio communication method |
| CN101325441A (en) * | 2007-06-11 | 2008-12-17 | 株式会社Ntt都科摩 | Method and apparatus for scheduling precoding system based on code book |
-
2007
- 2007-07-24 CN CNA2007101390739A patent/CN101355381A/en active Pending
-
2008
- 2008-07-23 JP JP2008189838A patent/JP5172517B2/en active Active
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102460996A (en) * | 2009-04-29 | 2012-05-16 | 英特尔公司 | Differential feedback scheme for closed-loop MIMO beamforming |
| CN102460996B (en) * | 2009-04-29 | 2014-08-20 | 英特尔公司 | Differential feedback scheme for closed-loop MIMO beamforming |
| US10027394B2 (en) | 2009-07-06 | 2018-07-17 | Intel Corporation | Beamforming using base and differential codebooks |
| CN104202130A (en) * | 2009-07-06 | 2014-12-10 | 英特尔公司 | Beamforming using base and differential codebooks |
| CN101989870A (en) * | 2009-08-05 | 2011-03-23 | 株式会社Ntt都科摩 | Method for acquiring channel quality indication information and base station thereof |
| CN102237950B (en) * | 2010-04-29 | 2016-06-08 | 株式会社Ntt都科摩 | A kind of subscriber equipment, base station and channel information feedback method |
| CN102237950A (en) * | 2010-04-29 | 2011-11-09 | 株式会社Ntt都科摩 | User equipment, base station and channel information feedback method |
| CN102916788B (en) * | 2012-07-30 | 2015-09-16 | 东南大学 | A kind of for the adaptive bit feedback method in multi-cell cooperative communication |
| CN102916788A (en) * | 2012-07-30 | 2013-02-06 | 东南大学 | Self-adaptation bit feedback method for multi-cell cooperative communication |
| CN107005579A (en) * | 2014-09-26 | 2017-08-01 | 华为技术有限公司 | For the feedback CQI for determining UE grades method, base station and UE |
| CN106330271A (en) * | 2015-06-19 | 2017-01-11 | 中国移动通信集团公司 | Multi-antenna multi-user non-orthogonal transmission method, base station, blind detection method, terminal |
| CN106330271B (en) * | 2015-06-19 | 2019-11-26 | 中国移动通信集团公司 | The nonopiate sending method of multi-antenna multi-user, base station, blind checking method, terminal |
| CN111788785A (en) * | 2018-01-29 | 2020-10-16 | 华为技术有限公司 | Precoding matrix index reporting method, communication device and medium |
| CN111788785B (en) * | 2018-01-29 | 2022-02-01 | 华为技术有限公司 | Precoding matrix index reporting method, communication device and medium |
| US11296759B2 (en) | 2018-01-29 | 2022-04-05 | Huawei Technologies Co., Ltd. | Precoding matrix index reporting method, communications apparatus, and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| JP5172517B2 (en) | 2013-03-27 |
| JP2009033744A (en) | 2009-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101355381A (en) | Method and apparatus for scheduling and pre-encoding based on channel vector quantification | |
| CN101166052B (en) | Precoding method of multi-input multi-output system and equipment using same | |
| CN101273548B (en) | Method for reducing feedback information spending of precoding MIMO-OFDM system | |
| CN101378277B (en) | Multi-user pre-coding and dispatching method and realize the base station of the method | |
| CN102035615B (en) | MIMO-based downlink data transmission method, device and system | |
| CN101984571B (en) | Pre-coding method for multi-user MIMO system | |
| CN101453259A (en) | Pre-encoded transmission method for MIMO system | |
| CN101583184B (en) | Radio communication apparatus and radio communication method | |
| CN101183890B (en) | Code book based CQI feedback method for multi-user precoding system | |
| CN102640440A (en) | Mobile station device, base station device, mimo system, and data transmission method | |
| CN101325441A (en) | Method and apparatus for scheduling precoding system based on code book | |
| CN101232478A (en) | Equipment and method for selecting matched data stream and corresponding precoding vectors | |
| CN103166688A (en) | A precoding implementation method, device and MIMO system | |
| JP2010529808A5 (en) | ||
| CN1909402B (en) | Self-adapting transmission method and apparatus used in spacing related MIMO system | |
| CN102545984A (en) | Linear and nonlinear comprehensive precoding method and device for multi-user multiple-input multiple-output (MIMO) system | |
| CN101094022B (en) | Transmitter, communication system, and communication method | |
| CN101860386B (en) | Multi-user random beam forming method and system | |
| CN102546123A (en) | Uplink precoding method and base station | |
| CN101741523A (en) | Finite Feedback Method for Feedback Codeword Spatial Distance Threshold in Multi-user and Multi-Antenna Systems | |
| CN101557280A (en) | Method and device for selecting pre-coding matrix/vector in multi-input and multi-output system | |
| CN102415006A (en) | Method for communicating in a network | |
| CN102136891A (en) | Improved channel state information (CSI) limited feedback method for multi-user multiple input multiple output (MU-MIMO) system | |
| CN103117780B (en) | Method for eliminating multi-user interference in multiple input and multiple output (MIMO) system | |
| CN101989867B (en) | Cooperative communication method and system, base station and mobile terminal device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20090128 |