WO2008006310A1

WO2008006310A1 - A method, system and device for forming wideband random beam

Info

Publication number: WO2008006310A1
Application number: PCT/CN2007/070210
Authority: WO
Inventors: Meng Zeng; Jun Wang; Shaoqian Li; Linjun Lv
Original assignee: Huawei Technologies Co., Ltd.; University Of Electronic Science And Technology Of China
Priority date: 2006-07-03
Filing date: 2007-06-29
Publication date: 2008-01-17
Also published as: CN101102136A; CN101102136B

Abstract

A method, system and device for forming a wideband random beam. The method comprises the steps of: generating a set Q having N unitary matrices, and determining the accumulated rate sum of each unitary matrix in the set Q according to the signal to interference plus noise ratio (SINR) of the sub-carrier fed back from an user device (101), where N is a natural number; determining an unitary matrix Qi* having the largest accumulated rate sum in the set Q, and determining an accumulated rate sum of a new unitary matrix Q_rand generated randomly (102); comparing the accumulated rate sum of the unitary matrix Qi* with the accumulated rate sum of a new unitary matrix Q_rand generated randomly, selecting the larger one as a beam-forming matrix, and forming a beam using the beam-forming matrix (103).

Description

Method, system and device for forming broadband random beam

Technical field

The present invention relates to wireless communication technologies, and in particular, to a method, system and device for forming a broadband random beam. Background of the invention

Multiple Input Multiple Output (MIMO) systems can greatly increase system capacity and become a hot topic of research. It is well known that when the channel state information (CSI) of the transceiver is ideally known, a variety of techniques can be used to achieve the capacity limit. However, because CSI's feedback overhead is very large, it is not practical. These technologies are point-to-point communication links. In contrast, recent research focuses on the application of multiple antennas in multi-user network environments, especially for broadcast channels and multiple access scenarios.

The capacity domain of the MIMO broadcast channel can be reached by dirty paper precoding (DPC), but the DPC needs to know CSI, and its complexity severely limits its practical application. If random beamforming is used, the user only needs to feedback the signal to interference and noise ratio (SINR), which can greatly reduce the amount of feedback and the need for CSI is much lower. When the number of users is large, the capacity of the random beamforming can approach the capacity of a transmission scheme with an ideal CSI.

In the case where the random beamforming technology in the prior art is mostly applied to a narrowband, the technique for forming a broadband beam has yet to be studied and improved. In the prior art, there is a technique of using multiple beams and multiple beam random beamforming for each user. . This technique uses a randomly generated 酉 matrix as the beamforming matrix for each time slot. Since the rate of randomly generated 酉 arrays may be low, this technique can approach the optimal capacity when the number of users is large. However, when the number of users is small, the performance of the formed random beam is still far from ideal. Summary of the invention

In view of this, the embodiments of the present invention provide a method for forming a broadband random beam, so that the formed random beam can have good performance even in the case of a small number of users.

The embodiment of the invention also provides a system for forming a broadband random beam, so that the formed random beam can have good performance even in the case of a small number of users.

The embodiment of the invention also provides a device for forming a broadband random beam, so that the formed random beam can have good performance even in the case of a small number of users.

The embodiment of the invention further provides a user equipment, so that the formed random beam can have good performance even in the case of a small number of users.

A method of forming a broadband random beam, the method comprising:

Generating a set 2 having N 酉 arrays, and determining an accumulation rate sum of each 酉 matrix in the set 2 according to an SINR of the subcarriers fed back by the user equipment, where N is a natural number; determining that the set β has the largest accumulation Rate sum 酉 array Q, and determine the cumulative rate sum of a randomly generated new 酉 array Q~ ^rand ;

The accumulated rate is compared with the sum of the accumulated rates of the randomly generated new array Q- ^rand , and the larger one is selected as a beamforming matrix, and the beam is formed using the beamforming matrix.

A system for forming a broadband random beam, the system comprising:

User equipment, SINR for feedback subcarriers;

a beamforming device, configured to generate a set 2 of N cells; receive an SINR of a subcarrier fed back by the user equipment, and determine, according to the SINR, an accumulation rate sum of each of the arrays in the set 2, where N is a natural number; determining a 酉 array T having the largest accumulated rate sum in the set 2; randomly generating a new CL matrix CL ^rand , and determining an accumulation rate sum of the Q ^rand ; and an accumulated rate sum and the randomly generated new 酉The cumulative rate of Q~ ^rand is compared and compared. The larger one is selected as a beamforming matrix, and a beam is formed using the beamforming matrix.

A device for forming a broadband random beam, the device comprising:

a 产生 array generating unit, configured to generate a set having N 酉 arrays; randomly generating a new 酉 matrix; a storage unit, configured to store the set 2 generated by the 产生 array generating unit;

a transceiver unit, configured to receive an SINR of a subcarrier fed back by the user equipment, and send the SINR;

a calculating unit, receiving a new array Q- ^rand generated by the array generating unit and an SINR sent by the transceiver unit, reading the set 2 from the storage unit, and calculating an accumulation of each of the sets β in the set β according to the SINR a rate sum and an accumulation rate sum of the new array Q- ^rand ; a comparison unit, comparing the accumulation rates of each of the arrays in the set 2 calculated by the calculation unit, and determining the maximum accumulation rate sum in the set 2 unitary matrix, and the rate of accumulation, selects the larger of the Q- ^rand is a unitary matrix and calculated by the computing unit;

a beam forming unit, wherein the comparing unit selects the accumulating rate with the ^rand and the larger 酉 matrix to form a beam;

Where N is a natural number.

A user equipment, the user equipment comprising:

a clustering unit, configured to cluster subcarriers;

a calculating unit, configured to calculate a SINR of a central subcarrier of each cluster according to a result of clustering of the clustering unit;

a sending unit, configured to send, to the device forming the broadband random beam, the SINR calculated by the calculating unit;

The SINR is used by the broadband random beamforming device to determine that each of the generated N The summation rate sum of each of the arrays in the set 2 of the unitary array, using one of the set 2 having the largest accumulated rate sum and the randomly generated new square array Q- ^rand as the beamforming matrix .

It can be seen from the above that the method, system and device provided by the embodiments of the present invention compare the randomly generated matrix rand with the matrix Q* having the largest accumulation rate sum in the 酉 matrix set β, and select the accumulation rate of the two. And the larger one is used as a beamforming matrix, that is, a memory-based method, each time using a better performing 酉 matrix as a beamforming matrix, instead of using a randomly generated 酉 matrix as a random beam as in the prior art. The forming matrix of the present invention, which is based on the memory broadband beamforming method, can ensure that the matrix used in forming the beam each time has a high accumulating rate and can ensure that even in the case of fewer users, Achieve good performance. BRIEF DESCRIPTION OF THE DRAWINGS

1 is a flowchart of a method for forming a broadband random beam according to an embodiment of the present invention; FIG. 2 is a specific flowchart of forming a broadband random beam according to an embodiment of the present invention; FIG. 3 is a schematic diagram of forming a broadband random beam according to an embodiment of the present invention; FIG. 4 is a structural diagram of a beam forming apparatus according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a user equipment according to an embodiment of the present disclosure;

6 is a first schematic diagram of rate and performance applied to a narrowband system according to an embodiment of the present invention;

7 is a second schematic diagram of rate and performance applied to a narrowband system according to an embodiment of the present invention;

8 is a schematic diagram of comparison of aggregate size (Ν) and performance applied to a narrowband system according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of rate and performance applied to a broadband system according to an embodiment of the present invention. Mode for carrying out the invention

The present invention will be described in detail below with reference to the drawings and specific embodiments.

FIG. 1 is a flowchart of a method for forming a broadband random beam according to an embodiment of the present invention. As shown in FIG. 1 , the method mainly includes the following steps:

Step 101: Generate a set 2 with N 酉 matrices, and determine an accumulation rate of each 酉 matrix in the set 2 according to feedback of a signal to interference ratio (SINR) of the subcarriers of the user equipment. , where Ν is a natural number;

The 集合 array set 2 may be generated when the location of the user equipment is relatively fixed, and the base station knows the location information of the user equipment. At this point, each of the arrays in set 2 can be formed by a combination of randomly selected user direction vectors, where W ' is the number of transmit antennas. In this way, the user's location information can be fully utilized to avoid pointing to the direction of no user in the random beamforming, so that the generated initial array 2 has better performance.

Preferably, when the user equipment feeds back the SINR of the subcarrier, the subcarrier can be clustered, and only the SINR of the central subchannel in each cluster is fed back, thereby reducing the feedback amount.

In addition, when the SINR of the subcarrier is fed back, the SINR threshold of the subcarrier may be preset, and then the SINR greater than the SINR threshold may be fed back. Alternatively, the feedback feedback is performed in the process of feeding back the SINR, that is, the value of the SINR is quantized, and the quantized result is fed back to reduce the feedback amount.

Step 102: Determine a matrix of the largest accumulated rate sum in the set 2 and determine an accumulated rate sum of a randomly generated new array ^Q - ^rand ;

Similarly, when the location of the user equipment is relatively fixed, and the base station knows the location information of the user equipment, the randomly generated new array can also be formed by randomly selecting a known user equipment direction vector, where W ' is the number of transmitting antennas. Step 103: Compare the accumulated rate with the sum of the accumulated rates of the randomly generated new array ^Q - ^rand , and select the larger one as the beamforming matrix, and use the beamforming matrix to form the beam.

Here, the rate of accumulation and the unitary matrix of the new randomly generated and compared ^ rate of accumulation, and select the larger of the ^rand as a beamforming matrix to form the beam, and in which the larger and Q * Replace it.

Further, if the new randomly generated unitary matrix Q ^ ^and accumulation rate set greater than 2 and has a minimum rate and cumulative unitary matrix ^Q, then in set 2 by Q ~ ^rand Alternatively, this unitary matrix may be used such Set 2 has better performance.

In the above process, all the 酉 arrays in the set 2 and the randomly generated new 酉 arrays preferably satisfy the equal-square distribution.

In the embodiment of the present invention, the time slots may be continuously executed for each time slot, and the new matrix randomly generated in each time slot is compared with the beam forming matrix adopted in the previous time slot, that is, the maximum accumulation rate in the set 2 and the set 2 The array is compared, and the larger the accumulation rate is selected as the beamforming matrix of the time slot. 2 is a specific flowchart of forming a broadband random beam according to an embodiment of the present invention, as shown in the following figure.

As shown in 2, the method includes two phases, an initialization phase and a data transmission phase. The initialization phase consists of the following steps:

Step 201: The base station (BS) generates a set with 酉 arrays

Where w ' is the number of BS transmit antennas, <3⁄4 obeys the equidistant distribution, ie the elements of the column vector of Q obey CW^ ¹ ), and the column vectors are orthogonal to each other. N is a design quantity, and the value is preferably greater than 2. If N is larger, the system throughput can be close to optimal when the number of users is small. Simulation N can be set to 10, in actual use to reduce feedback overhead can take 2 ~ 5

Step 202: The BS broadcasts each of the set of Qs determined by the BS, that is, 3⁄4 of the broadcast set Q.

Step 203: Each mobile station (MS) clusters the subcarriers and feeds back the maximum SINR of the central subcarriers in the cluster.

After receiving the 广播 matrix of the BS broadcast, the kth MS calculates the SINR of the central subcarrier of each cluster for the mth ( ^m = ^-N ') beams, and feeds back the maximum SINR of the central subcarrier in the cluster as ^S1NR , The MS may also feed back the label m of the corresponding beam and the central subcarrier label q;

Step 204: The BS calculates a rate sum of the qth subcarrier corresponding to the matrix Q according to the feedback SINR, and then calculates an accumulation rate sum of Q for all feedback subcarriers, determines an accumulation rate sum of all ^Qs in the set 2, and selects the largest one. , denoted, where q is the central subcarrier label of each cluster.

At this point, the initialization phase is completed and then the data transfer phase is entered. The data transfer phase includes:

Step 205: Randomly generate and broadcast a transmission matrix ^rand in each time slot t BS; Step 206: The MS feeds back the maximum SINR of the beam vector, the label of the corresponding beam, and the subcarrier label with the ^Q - ^rand column vector.

Step 203 of the specific calculation and initialization phase in this step is the same;

Step 207: Calculate the rate of Q~ ^rand and ^SR( Q~ ^rand) , according to the SINR of the MS feedback at this moment, update β'*

That is, if Q ^ RiCL mnc, Q is selected; a beamforming matrix; otherwise using 0 ^"rand, ^rand and replace Meanwhile, if the SR ^(Q_, preferably by 2 alternative ^rand ^Q.; Step 208: Form a beam using the selected beamforming matrix and transmit the data. In this step, in order to ensure the fairness of users in the network, a proportional fair scheduling (PFS) may be adopted. Here, the specific techniques of partial fair scheduling are clear to those skilled in the art, and therefore, they are not mentioned.

The following is a description of a specific embodiment of the application of the 802.22 proposal scenario. First, the broadband channel system model is given.

The branch now has a wideband channel 具有 with L taps, and by employing orthogonal frequency division multiplexing (OFDM), the L taps are decomposed into N subcarriers. The input and output signals are expressed as:

( 1 )

Let H = [/3⁄4, /3⁄4,...,; 3⁄4Γ , then the response of Η at frequency q is:

Expressed in matrix form,

w ].

Suppose that the set of N 酉 arrays ² = {Q 'Q ₂ , Q, ^ c^' is the beamforming matrix. The symbol transmitted on the qth subcarrier is denoted by ^, where is the column vector of Q, and ^W is the transmitted symbol on the mth beam.

The embodiments of the present invention are described below based on the above model in combination with specific parameters. The setting parameters are: FFT points are 2048, data and pilot subcarriers are 1680, divided into 30 subchannels, and each subchannel is divided into 4 BINs. Each BIN includes 12 data subcarriers and 2 pilot subcarriers, and 14 contiguous subcarriers of one BIN can be set as a cluster.

For the initialization phase: Step 1: The BS generates a set of N squares.

Q = {Q ₁ , Q ₂ , -, Q _A , }, Q _I e ^ _; where, the number of antennas is transmitted for the BS, and Q is broadcast to all users through the broadcast channel, where 3⁄4 is obediently distributed. Here, N is a design quantity, and the larger N is, the system throughput can be close to optimal when the number of users is small.

Step 2: Each user estimates its own channel ^ by one pilot in each BIN. According to the correlation, the channel coefficient can be used as the channel coefficient of all other subcarriers of this BIN.

The kth MS calculates the SINR of the mth beam of the central subcarrier of each BIN, wherein the central subcarrier is labeled q;

H ^k *V*VH ^k

SINR"

1 . -^-ιNΛ',,,

-+ H ^k *V*VH ^k

P (3) where m = l, ''', NH _m ^k = H _m ; is the signal-to-noise ratio at the receiving end, _{q = 1} , ..., 1680; is the mth column vector of Q, representing the mth Beam.

Step 3: Cluster the subcarriers, where the size of the cluster is 14 subcarriers, including 12 data subcarriers and 2 pilot subcarriers (corresponding to a BIN), and the maximum SINR of the central subcarrier in the MS feedback cluster is ^ ^SINRqm and the label m of the corresponding beam and the subcarrier label.

Step 4: BS based on the feedback SINR, by

Calculate the qth subcarrier rate and (where q is the central subcarrier label of each BIN), and the formula of the formula (4) adopts the formula existing in the prior art, and details are not described herein again. From the cumulative rate sum calculated by Σ _?= Α (5), determine the rate sum of all elements in set 2, and select the matrix with the largest rate sum, denoted as .

At this point, in the data transfer phase, the following steps are included:

Step 1: Randomly generate an equilateral distribution of the emission matrix ^Q - ^rand at each time slot t BS and broadcast the emission matrix ^rand

Step 2: The MS feeds back Q- ^rand as the maximum SINR of the transmit matrix, the label of the corresponding beam, and the subcarrier label.

Step 3: Calculate the rate of Lrand and SWCLranc, according to the SINR update of the MS feedback at this moment. Wherein, if ^(Q;)>SR(Q_mnd), Q is selected; data is transmitted as a beamforming matrix, otherwise Q" ^{rand is} used as a beamforming matrix to transmit data. Meanwhile, if

SR(Q_rand)>SR(Q _imm ) _; preferably in the set β with Q_mnd instead of Q, in the initialization phase: the feedback amount is

N xNumberofUser x Num ber of Cluster SINR and

N xNumberofUser xNumberof Cluster beam subscripts corresponding to the maximum SINR, where N is the size of the 酉 array set G, Wwmtera/t/ser is the number of user equipments, and Number of Cluster is the number of clusters of subcarriers. During the data transfer phase: the amount of feedback is

INumberofUser x Numberof Cluster SINR and

INumberofUser xNumberof Cluster beam subscript corresponding to the maximum SINR.

For the application under the 802.22 proposal, each 6 MHz band includes 30 subchannels in the 2K FFT mode, each subchannel includes 4 BINs, and each BIN includes 14 subcarriers. So in the initialization phase the system has a total of ¹² feedbacks ^(W umbefofUser _SINR and

120N NumberofUser beam subscript, feedback ^{240 x} W ^ro/User during data transmission SINR and 240 x N"mtero/[/ser beam subscripts.

For a single user, if a user is assigned 64 subcarriers, a set of 4 subcarriers, the number of transmit antennas is 4, and the SINR is quantized by 3 bits, because 4 transmit antennas correspond to 4 beams, and each user not only feeds back SINR, It also needs feedback to make it reach the beam number of the optimal SINR. 4 beams need 2bit quantization, then each user feedbacks 16*3+16*2=10byte; if the transmit antenna is 2, lbit quantization is needed, then each user feedback 16*3+16*l=8byte.

In the above process, the SINR threshold may be preset, and the MS only feeds back the SINR that is greater than a predetermined threshold, so that the feedback amount in the network is further reduced to

_T , Pr(max SINR . ≥ η) _τ , , Pr(max SINR . ≥ η) _T , _TT , , K i ≤ TM ^l ' ^m , , where K is the number of corpses, I ≤ TM ≤ M ^l - ^m "The probability that the SINR is greater than the threshold.

The above is a method for forming a broadband random beam provided by an embodiment of the present invention. The system for forming a broadband random beam provided by the embodiment of the present invention is described in detail below. 3 is a structural diagram of a system for forming a broadband random beam according to an embodiment of the present invention, the system comprising: a beamforming device 310 and a user equipment 320;

a beamforming device 310, configured to generate a set 2 of N cells; receive an SINR of a subcarrier fed back by the user equipment 320, and determine an accumulation rate sum of each of the arrays in the set 2 according to the SINR, where N is a natural number; determining a 酉 matrix having the largest cumulative rate sum in the set 2 ( ; randomly generating a new Q matrix Q_ rand , and determining an accumulation rate sum of the Q rand; a cumulative rate sum and the randomly generated new The accumulation rate of the Q array Q- ^rand is compared and compared, the larger one is selected as the beamforming matrix, and the beamforming matrix is used to form the beam.

The user equipment 320 is configured to feed back the SINR of the subcarrier to the beamforming device 310. More preferably, the user equipment 320 is further configured to perform clustering on subcarriers, and calculate an SINR of a central subcarrier of each cluster. When the user equipment 320 performs feedback, the child load may be first The waves are clustered, and only the SINR of the central subchannel in each cluster is fed back, thereby reducing the amount of feedback. The user equipment 320 may further preset a threshold P 子 of the subcarrier, compare the calculated SINR with the threshold, and feed back an SINR greater than the threshold. This also reduces the amount of feedback.

The user equipment 320 may further perform a quantization calculation on the calculated SINR, and feed back the quantized SINR. This also reduces one of the methods of feedback. The beamforming device 310 is further configured to replace ^Q - ^rand when the accumulating rate of CL ^rand is greater than the sum of the accumulating rates.

The beamforming device 310 is further configured to compare an accumulation rate of the randomly generated new matrix Q- ^rand with a matrix having a minimum accumulation rate sum in the set β, if

Q_rand's accumulating rate and greater than the accumulating rate sum, then Q-rand is replaced (3⁄4, wherein, as shown in FIG. 4, the beam forming apparatus 310 may include: a matrix generating unit 311, a storage unit 312, and a transceiving unit 313 a calculating unit 314, a comparing unit 315, and a beam forming unit 316;

The 酉 matrix generating unit 311 is configured to generate a set 2 having N 酉 arrays; randomly generate a new 酉 matrix Q_ rand . The storage unit 312 is configured to store the set generated by the 产生 array generating unit 311 2

The transceiver unit 313 receives the SINR of the subcarrier fed back by the user equipment, and sends the SINR to the computing unit 314;

The calculating unit 314 receives the new array Q- ^rand generated by the matrix generating unit 311 and the SINR sent by the transceiver unit 313, and reads the set 2 from the storage unit 312, and according to the

SINR, calculating an accumulation rate sum of each of the arrays in the set 2 and an accumulation rate sum of the new arrays ^rand ;

The comparing unit 315, the accumulative acceleration of each of the arrays in the set 2 calculated by the calculating unit 314 Comparing the rate and the comparison, determining a matrix having the largest accumulated rate sum in the set 2, and comparing the sum of the accumulated rates of the Q- ^rand calculated by the calculating unit 314, and selecting a larger one of the arrays;

The beam forming unit 316 forms a beam by using the summation rate selected by the comparison unit 315 and the accumulated rate in the Q~ ^rand and a larger matrix.

Where N is a natural number.

Further, the comparing unit 315 is further configured to: when the accumulation rate of ^rand is greater than the above, provide the ^rand to the storage unit 312, and send the first replacement notification to the storage unit 312;

The storage unit 312 is further configured to receive the Q- ^rand provided by the comparison unit 315, and after receiving the first replacement notification, store the Q- ^rand instead.

More preferably, the comparison unit 315, further for the accumulation rate of ^rand rate having the minimum accumulated unitary matrix and β are compared with the set, if the accumulated Q_ ^rand rate greater than the rate of accumulation ^Q And sending Q- ^rand to the storage unit 312, and sending a second replacement notification to the storage unit 312;

The storage unit 312 is further configured to receive the Q- ^rand provided by the comparison unit 315, and after receiving the second replacement notification, store Q- ^rand instead of ^Q.

5 is a schematic structural diagram of a user equipment 320 according to an embodiment of the present invention. As shown in FIG. 3.C, the user equipment includes: a clustering unit 321, a computing unit 322, and a sending unit 323;

The clustering unit 321, is configured to cluster the subcarriers;

The calculating unit 322 is configured to calculate, according to the result of the clustering unit 321 clustering, the SINR of the central subcarrier of each cluster;

The sending unit 323 is configured to send the SINR calculated by the calculating unit 322 to the beam shape Into device 310.

The user equipment 320 may further include: a gate P艮 setting unit 324 and a comparing unit 325; a threshold setting unit 324, configured to set a threshold value of the subcarrier, and provide the comparison unit

325;

The comparing unit 325 is configured to receive the gate P艮 value of the subcarrier provided by the gate P艮 setting unit 324 and the SINR provided by the calculating unit 322, and compare the two, and send the SINR greater than the threshold to the sending unit.

The simulation of the embodiment of the present invention will be described below.

The embodiment of the present invention is directed to the problem of how the broadband system approaches the capacity limit in the case of a small number of users. Obviously, the narrowband system can be considered as a special case of the broadband system (where the number of subcarriers is 1). For the convenience of simulation, the present invention is first described in the case of application to a narrow band.

FIG. 6 is a first schematic diagram of the rate and performance applied to a narrowband system according to an embodiment of the present invention, wherein the number of users is 20. The simulation conditions in Figure 6 are as follows: The channel is an independent Rayleigh fading channel, the maximum multi-lingual frequency shift is 10 Hz, the signal-to-noise ratio is 0 dB, each time slot is lms, the simulation is 1000 time slots, and the number of users is 20.

FIG. 7 is a second schematic diagram of rate and performance applied to a narrowband system according to an embodiment of the present invention, wherein the number of users is 20 and the number of transmitting antennas is 8. The simulation conditions in Figure 7 are: The channel is an independent Rayleigh fading channel, the maximum Doppler shift is 10 Hz, the signal-to-noise ratio is 0 dB, each time slot is lms, the simulation is 1000 time slots, and the number of users is 20. The number of antennas is fixed at 8. It can be seen from Fig. 6 and Fig. 7 that the present invention has a higher rate sum in the case where the number of users is small.

FIG. 8 is a schematic diagram of comparison of aggregate size (N) and performance applied to a narrowband system according to an embodiment of the present invention, where tx is the number of antennas, and u is the number of users. The situation of 4 antenna 20 users, 8 antenna 20 users, and 8 antenna 40 users is shown in FIG. In the figure, 4 antenna 20 users, 8 antenna 20 users, and 8 antenna 40 users are represented as (20u, 4tx), (20u, 8tx), respectively. (40u, 8tx). As can be seen from Fig. 8, in the case of a narrow band, the effect of the size N of the set 2 on the rate sum is not significant.

FIG. 9 is a schematic diagram of rate and performance applied to a broadband system according to an embodiment of the present invention. The simulation conditions of Figure 7 are: According to the proposed 2k mode, 1680 data and pilot subcarriers are allocated, the number of channel taps is L = 6; the size of the cluster is 14 subcarriers; the number of transmit antennas is 2; The size of 2 is 5.

It can be seen from Fig. 9 that the simulation of the parameters of the 2k mode in the proposal can achieve a larger rate sum when the number of users is small. The accumulating rate is calculated by ^? , where the accumulating rate and the rate sum of all used subcarriers when using the transmit matrix Q are represented.

It can be seen from the above that the method, system and device provided by the embodiments of the present invention compare the randomly generated matrix rand with the matrix Q* having the largest accumulation rate sum in the 酉 matrix set β, and select the accumulation rate of the two. The larger one is used as a beamforming matrix, that is, a memory-based approach, each time using a better performing matrix as a beamforming matrix. The memory broadband beamforming method adopted by the embodiment of the invention can ensure that the matrix used in forming the beam has a higher accumulating rate and each time, thereby ensuring good performance even in the case of fewer users. .

In addition, the embodiment of the present invention adopts a method of clustering subcarriers and only feeding back the maximum SINR of the central subcarriers in the cluster, which can reduce the feedback amount of the system. It is also possible to use a set threshold value to feed back only the SINR greater than the set threshold, which in turn reduces the feedback of the system.

The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalents, improvements, etc., which are made within the spirit and principles of the present invention, should be included in the present invention. Within the scope of protection.

Claims

Claim

A method for forming a broadband random beam, the method comprising: generating a set 2 of N 酉 matrices, and determining the set 2 in accordance with a signal to interference and noise ratio SINR of a subcarrier fed back by a user equipment The cumulative rate sum of each 酉 matrix, where Ν is a natural number;

Determining a matrix Q having a maximum accumulation rate sum in the set 2 and determining an accumulation rate sum of a randomly generated new lattice ^rand ;

Comparing the accumulation rate of Q, * with the summation rate sum of the randomly generated new Q array Q- ^rand , selecting the larger one as the beamforming matrix, and forming the beam using the beamforming matrix.

2. The method according to claim 1, wherein after the summed rate and the cumulative rate sum of the randomly generated new array Q_rand are further included: if the accumulation of the ^rand The rate sum is greater than the sum of the accumulated rates, and the Q_rand is substituted for the set β.

3. The method according to claim 1, wherein after the comparing the accumulated rate and the accumulated rate sum of the randomly generated new array Q_rand, further comprising: if the randomly generated The accumulation rate of the new array Q- ^{rand and} the array larger than the smallest accumulation rate in the set 2 are replaced by CL ^rand in the set 2.

The method according to claim 1, wherein each of the arrays in the set 2 is combined by randomly selecting W 'known user direction vectors when the user direction is known and the position is fixed. , where W ' is the number of transmit antennas.

The method according to claim 1, wherein when the direction of the user is known and the position is fixed, the randomly generated new array ^Q - ^rand is randomly selected from known users. The direction vectors are combined and formed, where is the number of transmitting antennas.

The method according to claim 1, characterized in that all the 酉 arrays in the set 2 satisfy an equi-square distribution, and the randomly generated new 酉 array also satisfies the equal-square distribution.

The method according to claim 1, wherein the SINR of the feedback subcarrier specifically includes:

Clustering subcarriers;

The user equipment feeds back the SINR of the central subcarrier within each cluster.

The method according to claim 1, wherein the SINR of the feedback subcarrier comprises: presetting a SINR threshold of the subcarrier, and the user equipment feeds back an SINR greater than the SINR threshold.

A system for forming a broadband random beam, the system comprising: a user equipment, a signal to interference and noise ratio SINR for a feedback subcarrier;

a beamforming device, configured to generate a set 2 of N cells; receive an SINR of a subcarrier fed back by the user equipment, and determine, according to the SINR, an accumulation rate sum of each of the arrays in the set 2, where N is a natural number; determining a 酉 array T having the largest accumulated rate sum in the set 2; randomly generating a new CL matrix CL ^rand , and determining an accumulation rate sum of the Q ^rand ; and an accumulated rate sum and the randomly generated new 酉The accumulation rate of the array ^rand is compared with, the larger one is selected as the beamforming matrix, and the beamforming matrix is used to form the beam.

10. A device for forming a broadband random beam, the device comprising: a matrix generating unit, configured to generate a set having N arrays; randomly generating a new array; and a storage unit for storing a matrix array Unit generated by 2;

a transceiver unit, configured to receive a signal to interference and noise ratio SINR of a subcarrier fed back by the user equipment, and send the SINR; a calculating unit, receiving a new array Q- ^rand generated by the array generating unit and an SINR sent by the transceiver unit, reading the set 2 from the storage unit, and calculating an accumulation of each of the sets β in the set β according to the SINR a rate sum and an accumulation rate sum of the new array Q- ^rand ; a comparison unit, comparing the accumulation rates of each of the arrays in the set 2 calculated by the calculation unit, and determining the maximum accumulation rate sum in the set 2 Arraying and comparing the summation rate of the Q- ^rand calculated by the calculation unit to select a larger one of the arrays;

a beam forming unit that forms a beam by using the summation rate selected by the comparison unit and the accumulated rate in the Q- ^rand and a larger matrix;

Where N is the natural number.

The device according to claim 10, wherein the comparing unit is further configured to: when the accumulating rate of CL ^rand is greater than the sum, provide the ^rand to the storage unit, and send the ^rand to the storage unit First alternative notice;

The storage unit receives the Q- ^rand provided by the comparison unit, and after receiving the first replacement notification, stores Q- ^rand instead.

The device according to claim 10, wherein the comparing unit is further configured to compare the accumulation rate of the Q- ^rand with a 酉具有具有具有具有 , , , And the accumulating rate of the Q_mnd and the sum of the accumulating rates greater than, sending Q_rand to the storage unit, and sending a second alternative notification to the storage unit;

The storage unit receives the Q- ^rand provided by the comparison unit, and after receiving the second replacement notification, stores Q- ^rand instead of ^Q.

13. A user equipment, the user equipment comprising:

a clustering unit, configured to cluster subcarriers;

a calculation unit, configured to calculate a central subcarrier of each cluster according to a result of clustering the clustering unit SINR of the wave;

The SINR is used by the wideband random beamforming device to determine an accumulated rate sum of each of the sets 2 of N arrays generated by the wideband random beamforming device, using the largest accumulated rate sum of the set 2 and the random generation The new array of Q- ^rands has a larger rate of accumulation as a beamforming matrix.

The user equipment according to claim 13, wherein the user equipment further comprises:

a threshold setting unit, configured to set a threshold value of the subcarrier;

And a comparing unit, configured to compare the SINR calculated by the calculating unit with a threshold value set by the threshold setting unit, and send an SINR greater than the threshold to the sending unit.