WO2014049416A2

WO2014049416A2 - Method and apparatus for coordinated multipoint downlink transmission between two cells

Info

Publication number: WO2014049416A2
Application number: PCT/IB2013/002095
Authority: WO
Inventors: Huan Sun
Original assignee: Alcatel Lucent
Priority date: 2012-09-28
Filing date: 2013-09-02
Publication date: 2014-04-03
Also published as: WO2014049416A3; TW201424290A; CN103716079A; CN103716079B

Abstract

The invention relates to a method and apparatus for coordinated multipoint downlink transmission between two cells. In a base station in each cell, a reference vector is determined; a channel vector of each user equipment in the cell is obtained to represent a valid downlink channel when a receiving vector is used having interference from another base station aligned to the reference vector; one or more user equipments are selected according to the reference vector, the channel vectors and a predetermined scheduling criterion; and a pre-coding vector for downlink transmission of each of the selected one or more user equipments is determined according to the reference vector, the channel vectors and a predetermined beam forming criterion. Thus interference from the other base station can be aligned to a reference vector and a desired signal can be placed in a spatial dimension different from the reference vector, which alleviates inter-cell interference and enables the respective base stations to separately perform multi-user scheduling and beam forming at a transmitter so as to thereby achieve a multi-user diversity gain and a beam forming gain.

Description

Method and Apparatus for Coordinated Multipoint Downlink Transmission between Two Cells

Field of the invention

The present disclosure relates to a wireless communication network and particularly to an inter-cell coordinated multipoint downlink transmission technology.

Background of the invention

The Multiple Input Multiple Output (MIMO) technology has attracted extensive attention due to its ability to significantly increase a data throughput and improve link reliability without incurring any extra bandwidth or transmission power. The MIMO technology has been adopted in numerous wireless communication standards, e.g., IEEE 802.11η, the 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE), Worldwide Interoperability for Microwave Access (WiMAX), etc. In a conventional cellular system, each user equipment (UE) receives a desired signal from its serving base station (BS) as well as interference from a base station in a neighboring cell, referred to as inter-cell interference (ICI). Therefore a downlink capacity of the conventional cellular system is limited by an interference level of ICI. As a distributed MIMO technology, the Coordinated Multi-Point (CoMP) scheme is capable of mitigating ICI through cooperation of multiple base stations.

Downlink CoMP transmission schemes are roughly categorized into two types: joint transmission (JT) and coordinated scheduling/beam-forming (CS/CB). In the joint transmission scheme, base stations in the same CoMP cluster share user information including Channel State Information (CSI) and user data. Although the joint transmission scheme can achieve a large performance gain, it also has a range of problems, e.g., a high backhaul requirement for exchanging the channel state information and user data, high computational complexity due to user scheduling and transmission of a pre-coding design as well as synchronization among the base stations in the same CoMP cluster. By contrast, the CS/CB scheme is capable of lowering the backhaul requirement because no user data is required to be exchanged among the base stations in the CoMP cluster.

Interference alignment (IA) has recently been proposed for a single-user MIMO system, that is, for a scenario where there is only one user in each cell. Interference alignment refers to spatial separation of a desired signal from an interference signal for each receiver through pre-coding by a transmitter in the case that there is a known channel matrix. For example, interference to each receiver is limited to a part of received signal space while leaving the remaining part of the received signal space to the desired signal. Interference alignment has the advantages of low complexity and good performance.

However in a practical MIMO system, there are a plurality of users in each cell, and a base station typically adopts multi-user MIMO (MU-MIMO) instead of single-user MIMO (SU-MIMO) to serve the plurality of users simultaneously. Furthermore the number of users served simultaneously by the base station is limited by the number of antennas of the base station, so the base station can only serve a part of the users over a time-frequency resource. In other words, the base station has to schedule the users to select a part of them to be served. Thus, the existing interference alignment can not be applicable directly to a practical multi-user MIMO system.

Summary of the invention

In view of the foregoing technical problem, an object of the invention is to provide a coordinated scheduling/beam forming solution based upon interference alignment so as to implement interference alignment and coordinated scheduling/beam forming in a multi-user MIMO system to thereby effectively alleviate inter-cell interference and achieve a multi-user diversity gain and a beam forming gain.

In the solution of the invention, in a base station in each cell, a reference vector is determined; a channel vector of each user equipment in the cell is obtained to represent a valid downlink channel when a receiving vector is used having interference from another base station aligned to the reference vector; one or more user equipments are selected according to the reference vector, the channel vectors and a predetermined scheduling criterion; and a pre-coding vector for downlink transmission of each of the selected one or more user equipments is determined according to the reference vector, the channel vectors and a predetermined beam forming criterion.

Thus interference from the other base station can be aligned to the first reference vector and a desired signal can be placed in a spatial dimension different from the first reference vector, that is, interference can be spatially separately from the desired signal, to thereby alleviate inter-cell interference and enable the respective base stations to separately perform multi-user scheduling and beam forming at a transmitter, that is, multi-cell multi-user scheduling can be simplified successfully into single-cell multi-user scheduling, to thereby achieve a multi-user diversity gain and a beam forming gain without involving high complexity.

In an example, the respective cells share the same reference vector,and in another example, reference vectors determined by the respective cells can be different.

According to an aspect of the invention, there is provided a method, in a first base station in a first cell, of coordinated multipoint downlink transmission between the first cell and a second cell, wherein both the first base station and a second base station in the second cell are equipped with a plurality of antennas, the method includes the steps of: A. determining a first reference vector; B. obtaining a first channel vector corresponding to each of a plurality of user equipments in the first cell, the first channel vector representing a valid downlink channel when the corresponding user equipment uses a first receiving vector, and the first receiving vector being determined according to an interference alignment criterion that interference from the second base station is aligned to the first reference vector; C. selecting one or more of the plurality of user equipments in the first cell according to the first reference vector, the first channel vectors of the plurality of user equipments in the first cell and a predetermined scheduling criterion; and D. determining a first pre-coding vector for downlink transmission of each of the selected one or more user equipments according to the first reference vector, the first channel vectors of the selected one or more user equipments and a predetermined beam forming criterion.

According to a particular embodiment of the invention, the first reference vector is a right singular vector of downlink channel matrixes of the plurality of user equipments in the first cell and one of a plurality of user equipments in the second cell. In an example, a user equipment can be selected randomly, and a right singular vector corresponding to the largest singular value of the downlink channel matrix of the user equipment can be taken as the first reference vector. In another example, a singular value corresponding to the right singular vector that serves as the first reference vector can be the largest among singular values of the downlink channel matrixes of the plurality of user equipments in the first cell and all of the plurality of user equipments in the second cell.

According to a particular embodiment of the invention, the method further includes in the step A the steps of: Al . determining and sending to the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the first cell and the corresponding right singular vector; A2. receiving from the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the second cell and the corresponding right singular vector; and A3, determining the first reference vector according to the determined singular value and corresponding right singular vector and the received singular value and corresponding right singular vector.

According to a particular embodiment of the invention, the interference alignment criterion is represented in the formula of:

¾¾wi (( )^H Η¹;· ¹ ) = span ((b^ref f ) , wherein b^ref represents the first reference vector, represents the first receiving vector of a user equipment i among the plurality of user equipments in the first cell j, and represents the downlink channel matrix of the second base station in the second cell 3-j to the user equipment i in the first cell j. As can be apparent, the respective user equipments design the receiving vectors such that interference from the other cells is aligned to the direction of the first reference vector.

According to the invention, the first channel vectors representing the valid channels of the respective user equipments can be obtained in the step B in numerous ways.

According to a particular embodiment of the invention, the method further includes in the step B the steps of: B l l . obtaining the downlink channel matrix of the user equipment and an interference downlink channel matrix of the second base station to the user equipment; and B 12. determining the first downlink vector of the user equipment according to the first reference vector, the obtained downlink channel matrix and the obtained interference downlink channel matrix. Fore example, the user equipment can obtain the downlink channel matrix and the interference downlink channel matrix through channel reciprocity or a channel feedback.

According to another particular embodiment of the invention, the method further includes in the step B the steps of: B21. sending the first reference vector to the plurality of user equipments in the first cell; and B22. receiving the first channel vector of each of the plurality of user equipments in the first cell from the user equipment. Correspondingly the respective user equipments determines and send to the base station the first channel vectors according to their own downlink channel matrixes and interference downlink channel matrixes upon reception of the first reference vector from the base station. The amounts of information in the first reference vector and the first channel vectors are less than the downlink channel matrixes and interference downlink channel matrixes of the user equipments, which facilitate a channel feedback.

According to a particular embodiment of the invention, the step C includes: selecting the one or more user equipments from the plurality of user equipments in the first cell based on an iterative zero-forcing algorithm.

According to a particular embodiment of the invention, the scheduling criterion is maximum rate scheduling, proportional fair scheduling or round-robin scheduling.

According to a particular embodiment of the invention, the beam forming criterion is a zero-forcing or minimum mean square error criterion.

According to another aspect of the invention, there is provided an apparatus, in a first base station in a first cell, for coordinated multipoint downlink transmission between the first cell and a second cell, wherein both the first base station and a second base station in the second cell are equipped with a plurality of antennas, the apparatus includes: a first determining unit configured to determine a first reference vector; an obtaining unit configured to obtain a first channel vector corresponding to each of a plurality of user equipments in the first cell, the first channel vector representing a valid downlink channel when the corresponding user equipment uses a first receiving vector, and the first receiving vector being determined according to an interference alignment criterion that interference from the second base station is aligned to the first reference vector; a selecting unit configured to select one or more of the plurality of user equipments in the first cell according to the first reference vector, the first channel vectors of the plurality of user equipments in the first cell and a predetermined scheduling criterion; and a second determining unit configured to determine a first pre-coding vector for downlink transmission of each of the selected one or more user equipments according to the first reference vector, the first channel vectors of the selected one or more user equipments and a predetermined beam forming criterion.

Brief description of drawings

Other features, objects and advantages of the invention will become more apparent upon review of the following detailed description of non-limiting embodiments thereof with reference to the drawings in which:

Fig.l illustrates a schematic diagram of coordinated downlink transmission by two cells;

Fig.2 illustrates a flow chart of a method of coordinated multipoint downlink transmission according to a particular embodiment of the invention; and

Fig.3 illustrates a block diagram of an apparatus for coordinated multipoint downlink transmission according to a particular embodiment of the invention. Detailed description of embodiments

Fig.l illustrates a schematic diagram of coordinated downlink transmission by two cells. Generally, two cells performing coordinated downlink transmission are adjacent cells. As illustrated in Fig.l , a cell 110 and a cell 120 are adjacent, and downlink transmission in each of the two cells brings interference to the other. The cell 110 includes a base station 130 and a plurality of user equipments, e.g., user equipments 151 and 152, wherein the base station 130 is an enhanced Node B (eNB). Similarly the cell 120 includes a base station 140 and a plurality of user equipments, e.g., user equipments 153 and 154. Moreover both the base station 130 and the base station 140 are equipped with a plurality of antennas, and the respective user equipments are also equipped with a plurality of antennas.

Assumed here both the base station 130 and the base station 140 are equipped with M antennas, and the respective user equipments are equipped with N antennas. It shall be noted that although it is assumed here that both of the base stations are equipped with the same number of antennas and the respective user equipments are also equipped with the same number of antennas, those skilled in the art shall appreciate that the method and apparatus according to the invention can also be applicable to a scenario where the two bases stations are equipped with different numbers of antennas and the respective user equipments are also equipped with different numbers of antennas. Moreover each cell is further assumed to include K user equipments. Those skilled in the art can also appreciate that the numbers of user equipments included in respective cells in a practical system tend to be different and will vary over time, and the method and apparatus according to the invention can also be applicable to such a scenario.

Hereinafter a downlink channel matrix of a base station j to a user equipment i in the current cell is denoted as H^{[ ]} , and a downlink channel matrix of another base station 3-j to the user equipment i is denoted as H^^] . As illustrated, the respective user equipments in the cell 110 (i.e., a cell 1) are served by the base station 130, and downlink channel matrixes of the base station 130 to the K user equipments in the cell 110 where the base station 130 is located are Η5^{1 1]} ,· · · , Η^ ^1] respectively. Also a downlink signal of the base station 140 will bring interference to the respective user equipments in the cell 110, and interference downlink channel matrixes of the base station 140 to the K user equipments in the cell 110 are HE,^U] ,- - -, HE ^1] . Similarly the respective user equipments in the cell 120 (i.e., a cell 2) are served by the base station 140, and downlink channel matrixes of the base station 140 to the K user equipments in the cell 120 where the base station 140 is located are H^[ ₂ ^] ,- - -, HE '^2] respectively. Also a downlink signal of the base station 130 will bring interference to the respective user equipments in the cell 120, and interference downlink channel matrixes of the base station 130 to the K user equipments in the cell 120 are H^²¹ ,- - -, !!^'²¹ .

Fig.2 illustrates a flow chart of a method of coordinated multipoint downlink transmission according to a particular embodiment of the invention; and Fig.3 illustrates a block diagram of an apparatus for coordinated multipoint downlink transmission according to a particular embodiment of the invention. Each base station is arranged therein with the apparatus 300 for coordinated multipoint downlink transmission. As illustrated in Fig.3, the apparatus 300 includes a first determining unit 310, an obtaining unit 320, a selecting unit 330 and a second determining unit 340.

Referring to Fig.2 and Fig.3, the first determining unit 310 determines a first reference vector ^ref in the step S210.

According to a particular embodiment of the invention, the first reference vector is a right singular vector of the downlink channel matrixes of the K user equipments in the cell 110 and one of the K user equipments in the cell 120. In an example, a user equipment is selected randomly, and a right singular vector corresponding to the largest singular value (i.e., the first singular value) of the downlink channel matrix of the user equipment is taken as the first reference vector. Specifically if the user equipment i in the cell j is selected, and singular value decomposition (SVD) of the downlink channel matrix of the user e uipment is:

then the first reference vector for interference alignment is determined as:

In another example, the first reference vector is determined such that a singular value corresponding to the right singular vector that serves as the first reference vector is the largest among singular values of the downlink channel matrixes of all user equipments. Specifically the first reference vector b^re is determined in the formulas of:

to o , k₀ } = ^max If )² }, and

where h_m ^{l 3} represents the m-th singular value of the user equipment i e cell j .

According to a particular embodiment of the invention, the foregoing step S210 includes the steps of: determining and sending to the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the first cell and the corresponding right singular vector; receiving from the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the second cell and the corresponding right singular vector; and determining the first reference vector as the right singular vector corresponding to the larger one of the determined singular value and the received singular value according to the determined singular value and corresponding right singular vector and the received singular value and corresponding right singular vector.

In other words, firstly each base station finds and sends to another base station the largest one of the singular values of the downlink channel matrixes of all of user equipments in the current cell and the corresponding singular vector; and secondly each base station finds out the largest one of the singular values of the downlink channel matrixes of all of the user equipments in the cell and another cell by comparing the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the current cell with the singular value received from the other base station (which is the largest one of the singular values of the downlink channel matrixes of all of the user equipments in the other cell).

In this way, it is only necessary for each base station to know the downlink channel matrixes of the user equipments in the current cell without any knowledge of the downlink channel matrixes of the user equipments in the other cell.

Further referring to Fig.2 and Fig.3, in the step S220, the obtaining unit 320 obtains a first channel vector corresponding to each of a plurality of user equipments in the first cell, the first channel vector representing a valid downlink channel when the corresponding user equipment uses a first receiving vector, and the first receiving vector being determined according to an interference alignment criterion that interference from the second base station is aligned to the first reference vector.

It shall be noted that actually the respective user equipments may not necessarily receive a desired signal from the base station using the first receiving vectors. The user equipments can design their receiving vectors independently of the first receiving vectors, e.g., through zero-forcing, maximum ratio combining, Minimum Mean Square Error (MMSE), etc.

¾¾wi (( )^H Η¹;· ¹ ) = span ((b^ref f ) , wherein represents the first receiving vector of the user equipment i in the cell j .

After the first receiving vector is determined, the first channel vector representing the valid downlink channel can be determined as:

wherein h _j- ^'^ represents the first channel vector of the user equipment i in the cell j .

According to a particular embodiment of the invention, the first receiving vectors and thus the first channel vectors of the user equipments in the current cell can be determined by the base station. According to a particular embodiment of the invention, the first receiving vectors and thus the first channel vectors of the user equipments can be determined by the respective user equipments.

In the case of being determined by the base station, the step S220 includes the steps of: for each of the plurality of user equipments in the first cell (e.g., the cell j), the first base station (e.g., the base station j) obtaining the downlink channel matrix H^[j ¹ of the user equipment and the interference downlink channel matrix H^¹ of the second base station (e.g., the base station 3-j) to the user equipment, and determining

[i ]

the first receiving vector w according to the first reference vector b^re and the obtained interference downlink channel matrix H^¹ , and then determining the first channel vector h _j- ^'^ of the user equipment according to the obtained downlink channel matrix H^{[! 1} and the

[i ]

determined first receiving vector w In the case of being determined by the user equipment, the step S220 includes the steps of: the first base station (e.g., the base station j) sending the first reference vector b^re to the plurality of user equipments in the first cell (e.g., the cell j); and receiving the first channel vector of each of the plurality of user equipments in the first cell from the user equipment. Correspondingly upon reception of the first reference vector b^re from the base station j, the user equipment i in the cell j determines the first

[i ]

receiving vector w according to the received first reference vector b^re and the interference downlink channel matrix H^¹ of the user equipment, and then determines the first channel vector h^{[ ' j]} of the user equipment according to the downlink channel matrix H^{[! 1} of the user

[i ]

equipment and the determined first receiving vector , and sends the first channel vector to the base station j.

Still referring to Fig.2 and Fig.3, in the step S230, the selecting unit 330 selects one or more of the plurality of user equipments in the first cell according to the first reference vector, the first channel vectors of the plurality of user equipments in the first cell and a predetermined scheduling criterion.

According to a particular embodiment of the invention, the one or more user equipments are selected from the plurality of user equipments in the first cell based on an iterative zero-forcing algorithm in the foregoing step S230. Moreover the scheduling criterion is maximum rate scheduling, proportional fair scheduling or round-robin scheduling.

Hereinafter, the maximum rate scheduling will be taken as an example of the scheduling criterion to describe how the base station selects one or more of the plurality of user equipments in the current cell based on an iterative zero-forcing algorithm, where the selected one or more user equipments constitute a subset of user equipments for downlink transmission. The iterative zero-forcing algorithm includes three phases:

1) Initialization, where let ^_c = {/, |/ = 1, · · · , Κ U{ ₀] and Ω₅ = 0 (that is, the subset of user equipments is initialized to NULL), where l₀ represents an index of the dummy user equipment;

2) In the first round iteration (i.e., i = 1), the dummy user equipment is selected as the first candidate user equipment, and the subset of user equipments, matrixes R^"1 and Q₁ and the maximum rate Q are updated as follows:

Q_C=Q_C\ I₀},Q_S=Q_SU{I₀},

C_sum= g₂(\ + PT), where T represents the norm square of the first reference vector, i.e.,

3) For succeeding rounds of iteration, i.e., i = 2 to M:

a) For each leH_s, perform the following procedure:

1) Let Ω₅ = Ω₅ U{/} , and update the matrixes R^"1 and Q₁ as follows:

wherein r.=(h^) Q__{1 ? ¾} = (h^) (i-Q^) , and r.. = q,

2) Calculate the corresponding maximum rate, and if

R,' = §ι>§2'^"''§/ \, then

b) Find the potential candidate user equipment according to the maximum sum scheduling criteria, where: v. = arg maxj . (/) ,/ E Ω₀ ]

c) If C_t≤C_sum , then the algorithm terminates; otherwise, the subset of user equipments, the matrixes R^"1 and Q₁ and the maximum rate Q are updated as follows:

= Ω, U {v,. } , Ω_ε = Ω_ε \ {v,. } , C_mm = C_t .

Still referring to Fig.2 and Fig.3, in the step S240, the second determining unit 340 determines a first pre-coding vector for downlink transmission of each of the selected one or more user equipments according to the first reference vector, the first channel vectors of the selected one or more user equipments and a predetermined beam forming criterion.

For example, the beam forming criterion can be a zero-forcing criterion, that is, a pre-coding vector for beam forming can be determined in the formula of:

where t i ^' ] represents the pre-coding vector for downlink transmission of the user equipment i in the cell j, a^[l,}] re resents a normalization factor to normalize the pre-coding vector so that

1 , where the symbol "† " represents pseudo inversion.

Alternatively the beam forming criterion can be a minimum mean square error criterion

Those skilled in the art shall appreciate that all the foregoing embodiments are merely illustrative but not to limiting. Different technical features appearing in different embodiments can be combined to advantage. Those skilled in the art shall appreciate and make other variant embodiments than the disclosed embodiments upon study of the drawings, the description and the claims. In the claims, the term "comprising" will not preclude another device(s) or step(s); the indefinite article "a" or "an" will not preclude plural; and the terms "first", "second", etc., are intended to designate a name but not to suggest any specific order. Any reference numerals in the claims shall not be construed as limiting the claimed scope of the invention. The mere fact that some technical features appear in different dependent claims shall not suggest that these technical features can not be combined to advantage.

Claims

1. A method, in a first base station in a first cell, of coordinated multipoint downlink transmission between the first cell and a second cell, wherein both the first base station and a second base station in the second cell are equipped with a plurality of antennas, the method comprises the steps of:

A. determining a first reference vector;

B. obtaining a first channel vector corresponding to each of a plurality of user equipments in the first cell, the first channel vector representing a valid downlink channel when the corresponding user equipment uses a first receiving vector, and the first receiving vector being determined according to an interference alignment criterion that interference from the second base station is aligned to the first reference vector;

C. selecting one or more of the plurality of user equipments in the first cell according to the first reference vector, the first channel vectors of the plurality of user equipments in the first cell and a predetermined scheduling criterion; and

D. determining a first pre-coding vector for downlink transmission of each of the selected one or more user equipments according to the first reference vector, the first channel vectors of the selected one or more user equipments and a predetermined beam forming criterion.

2. The method according to claim 1, wherein the first reference vector is a right singular vector of downlink channel matrixes of the plurality of user equipments in the first cell and one of a plurality of user equipments in the second cell.

3. The method according to claim 2, wherein a singular value corresponding to the right singular vector that serves as the first reference vector is the largest among singular values of the downlink channel matrixes of the plurality of user equipments in the first cell and all of the plurality of user equipments in the second cell.

4. The method according to claim 3, wherein the step A further comprises the steps of:

Al . determining and sending to the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the first cell and the corresponding right singular vector;

A2. receiving from the second base station the largest one of the singular values of the downlink channel matrixes of all of the plurality of user equipments in the second cell and the corresponding right singular vector; and

A3, determining the first reference vector according to the determined singular value and corresponding right singular vector and the received singular value and corresponding right singular vector.

5. The method according to claim 1 , wherein the interference alignment criterion is represented in the formula of:

¾¾wi (( ^M )^H H ) = span ((b^ref f ) , wherein b^ref represents the first reference vector, w^^'^ represents the first receiving vector of a user equipment i among the plurality of user equipments in the first cell j, and represents the downlink channel matrix of the second base station in the second cell 3-j to the user equipment i in the first cell j .

6. The method according to claim 1 , wherein the step B comprises: for each of the plurality of user equipments in the first cell,

B l l . obtaining the downlink channel matrix of the user equipment and an interference downlink channel matrix of the second base station to the user equipment; and

B 12. determining the first downlink vector of the user equipment according to the first reference vector, the obtained downlink channel matrix and the obtained interference downlink channel matrix.

7. The method according to claim 1, wherein the step B comprises: B21. sending the first reference vector to the plurality of user equipments in the first cell; and

B22. receiving the first channel vector of each of the plurality of user equipments in the first cell from the user equipment.

8. The method according to claim 1, wherein the step C comprises: selecting the one or more user equipments from the plurality of user equipments in the first cell based on an iterative zero-forcing algorithm.

9. The method according to claim 1 , wherein the scheduling criterion is maximum rate scheduling, proportional fair scheduling or round-robin scheduling.

10. The method according to claim 1 , wherein the beam forming criterion is a zero-forcing or minimum mean square error criterion.

11. An apparatus, in a first base station in a first cell, for coordinated multipoint downlink transmission between the first cell and a second cell, wherein both the first base station and a second base station in the second cell are equipped with a plurality of antennas, the apparatus comprises:

- a first determining unit configured to determine a first reference vector;

- an obtaining unit configured to obtain a first channel vector corresponding to each of a plurality of user equipments in the first cell, the first channel vector representing a valid downlink channel when the corresponding user equipment uses a first receiving vector, and the first receiving vector being determined according to an interference alignment criterion that interference from the second base station is aligned to the first reference vector;

- a selecting unit configured to select one or more of the plurality of user equipments in the first cell according to the first reference vector, the first channel vectors of the plurality of user equipments in the first cell and a predetermined scheduling criterion; and - a second determining unit configured to determine a first pre-coding vector for downlink transmission of each of the selected one or more user equipments according to the first reference vector, the first channel vectors of the selected one or more user equipments and a predetermined beam forming criterion.