KR20100012283A - Apparatus and method for regularized coordinated beamforming in multiple antenna - Google Patents

Abstract

PURPOSE: A device for normalization collaboration beam forming in a multiple antenna system and a method thereof are provided to prevent amplification of noises by performing regularized coordinated beamforming in the multiple antenna system. CONSTITUTION: A transmitter includes an encoder(101), a modulator(103), a demultiplexer(105), a pre-coder(107), an RF(Radio Frequency) processor, a channel confirming unit(111), a first beamforming vector generator(113), a second beamforming vector generator(115), and a beamforming vector selector. The first beamforming vector selector measures a transmitting beamforming vector by channel information. The second beamforming vector generator normalizes the transmitting beamforming vector.

Description

APPARATUS AND METHOD FOR REGULARIZED COORDINATED BEAMFORMING IN MULTIPLE ANTENNA}

TECHNICAL FIELD The present invention relates to coordinated beamforming (hereinafter referred to as "CBF") in a multi-antenna system, and more particularly, to an apparatus and method for normalized cavity beamforming that minimizes interference and noise.

Precoding and post-processing methods for simultaneously transmitting data streams to multiple user terminals in a base station equipped with multiple antennas include a linear scheme and a nonlinear scheme. Coordinated beamforming (hereinafter referred to as "CBF") corresponds to a linear scheme. In addition, the linear scheme is further divided into a unitary precoding scheme and a non-unitary precoding scheme, and the CBF corresponds to a non-unitary precoding scheme. That is, the CBF performs multi-user communication through linear operation in a transmitter and a receiver, but the precoding matrix of the transmitter is not a unitary matrix.

The CBF scheme calculates a precoding matrix and a reception beamforming vector (or matrix) by using downlink channel state information of receivers at a transmitter. However, at this time, the precoding matrix and the reception beamforming vector are combined at the receiver by using the reception beamforming vector, so that no inter-user interference or multiple-access interference occurs. You must be satisfied. A precoding matrix and a reception beamforming vector satisfying the above properties can generally be obtained through an iteration characteristic.

In the CBF scheme, a base station may use the following two methods to deliver a reception beamforming vector to a terminal. The first method is to use pilot beamforming. After assigning a dedicated pilot to each terminal so as not to be duplicated, and then beamforming a pilot using a transmission beamforming vector corresponding to each terminal, the terminal estimates an effective channel and then A matched filter is formed and used as a reception beamforming vector. In this method, since pilots should be allocated to each UE so that they do not overlap (orthogonally), more dedicated pilots are required as the number of user streams increases. The second method is to quantize the transmission beamforming vector of each terminal and transmit it through a forwarding channel (downlink control channel). In this case, since the transmission beamforming vector is different for each terminal, the capacity of the forwarding channel increases in proportion to the number of terminals.

The prior art uses a zero-forcing based iterative algorithm to calculate the transmission and reception beamforming vectors. That is, from the initial reception beamforming vector, the transmission and reception beamforming vector is repeatedly updated until the predetermined condition is satisfied to determine the transmission and reception beamforming vector. Therefore, in the case of using the transmission / reception beamforming vector obtained in the prior art, in the ideal case (no channel error and noise), the interference between the users is all eliminated. However, the actual environment may not be ideal, resulting in a significant amplification of noise due to the constraint of completely eliminating user-to-user interference. In particular, the lower the signal-to-noise ratio (SNR), the greater the performance degradation.

The regularized inversion technique combines regularization with the existing zero-forcing beamforming (ZF-BF) technique, which adds a regularization factor to inverse operation, resulting in bad channel conditions. By preventing the inverse of the ill-conditioned channel matrix, the above problem can be solved.

However, in the case of a CBF having multiple antennas in a receiver and calculating a transmission / reception beamforming vector through an iterative algorithm, a method of applying normalization is not known, and a method of obtaining an optimal normalization factor is also known. Not.

An object of the present invention is to provide an apparatus and method for regularized coordinated beamforming in a multi-antenna system.

Another object of the present invention is to provide an apparatus and method for simultaneously minimizing interference and noise through normalized co-beamforming in a multi-antenna system.

Another object of the present invention is to provide an apparatus and method for reducing overhead by using a pilot when calculating a transmit / receive beamforming vector in a multi-antenna system.

Another object of the present invention is to provide an apparatus and method for regularized coordinated beamforming in a common pilot based wireless communication system such as Long Term Evolution (LTE) system.

According to a first aspect of the present invention for achieving the above object, a transmitter for normalized co-beamforming in a multi-antenna system, the first beamforming vector generator for calculating a transmission beamforming vector using channel information; And a second beamforming vector generator for normalizing the transmission beamforming vector, and a pre-encoder for precoding and transmitting data using the normalized transmission beamforming vector.

According to a second aspect of the present invention for achieving the above objects, a receiver for normalized co-beamforming in a multi-antenna system, comprising: receiving a received beamforming vector from a first pilot signal beamformed by a reference transmit beamforming vector; And a channel estimator for calculating, an effective channel gain calculator for calculating an effective channel gain from the calculated reception beamforming vector, and a back encoder for performing channel equalization using the effective channel gain and receiving data. It features.

According to a third aspect of the present invention for achieving the above objects, there is provided a transmission method for normalized joint beamforming in a multi-antenna system, the method comprising: calculating a transmission beamforming vector using channel information; And normalizing the vector and precoding the data using the normalized transmission beamforming vector.

According to a fourth aspect of the present invention for achieving the above objects, in a receiving method for normalized co-beamforming in a multi-antenna system, a reception beamforming vector is obtained from a first pilot signal beamformed by a reference transmission beamforming vector. And calculating the effective channel gain from the calculated reception beamforming vector, performing channel equalization using the effective channel gain, and receiving data.

As described above, by performing normalized co-beamforming in a multi-antenna system, there is an advantage in that noise generated due to a constraint for removing interference between users can be prevented from being greatly amplified. That is, joint beamforming may be performed in consideration of interference and noise. Thus, in a low SNR radio channel environment, significantly better throughput can be achieved than in the prior art.

 In the normalized joint beamforming technique, an overhead may be reduced by using a pilot when the received beamforming vector is transmitted to a receiver, and the present invention may be applied to a common pilot-based LTE system.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, detailed descriptions of related well-known functions or configurations will be omitted if it is determined that the detailed description of the present invention may unnecessarily obscure the subject matter of the present invention. Terms to be described later are terms defined in consideration of functions in the present invention, and may be changed according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the specification.

Hereinafter, an apparatus and method for regularized coordinated beamforming in a multi-antenna system will be described.

In particular, an apparatus and method for performing normalized co beamforming to prevent an increase in noise due to interference constraints and to apply a pilot technique for reducing overhead will be described.

In addition, the present invention will be described a method of applying the present invention to a common pilot (common pilot) based system, such as Long Term Evolution (LTE).

In order to perform the operation according to the present invention, the following assumptions are required. First, in order to obtain downlink channel information for all receivers, the transmitter uses an uplink sounding channel in a time division duplexing (TDD) scheme or estimates the terminal through a feedback channel. Downlink channel information may be used. Finally, assume a flat fading channel where the channel changes slowly.

1 illustrates a transmitter for performing normalized joint beamforming in a multiple antenna system according to the present invention.

Referring to FIG. 1, the transmitter includes an encoder 101, a modulator 103, a demultiplexer 105, a pre encoder 107, an RF processor 109, a channel checker 111, and a first beamforming vector. And a generator 113, a second beamforming vector generator 115, and a beamforming vector selector 117.

The encoder 101 encodes and transmits the transmission data according to a modulation level (eg, Modulation and Coding Scheme (MCS) level). The modulator 103 modulates the coded data from the encoder 101 according to the modulation level and outputs the modulated data. The demultiplexer 105 demultiplexes the modulation symbols provided from the modulator 103 to be equal to the number of transmit antennas and outputs a plurality of data streams.

The first beamforming vector generator 113 generates transmission and reception beamforming vectors using channel state information of each stream provided from the channel identification unit 311, and generates the transmitted beamforming vector values. Both are provided to the first encoder 107, the second beamforming vector generator 115, and the beamforming vector selector 117. In a conventional iterative approach, a transmission / reception beamforming vector is generated using the channel state information. That is, the transmission / reception beamforming vector is updated repeatedly from the initial beamforming vector until the condition is satisfied.

The transmission / reception beamforming vector will be represented by Equation 1 below.

Here, M is a transmit beamforming vector, m _k is a transmit beamforming vector value of the k-th antenna, W is a receive beamforming vector, and w _k is a receive beamforming vector value of the k-th antenna.

The second beamforming vector generator 115 normalizes transmission beamforming vector values from the first beamforming vector generator 113. In this case, by normalizing the transmission beamforming vector M from the second beamforming vector generator 115 as shown in Equation 2, noise increase due to user interference restriction may be prevented.

는 프로베니우스놈(frobenius norm) 연산을 의미한다.Here, M _reg is a normalized transmission beamforming vector, I is a unit matrix or unit vector, M is a transmission beamforming vector provided from the first beamforming vector generating unit 113, and (·) ^H is a mission (Hermitian) operation, (·) ^-1 is inverse operation, P is transmit power, N _o is noise power, H _k is channel matrix of k-th terminal, m _k is k-th transmission Beamforming vector value,

Means the Frobenius norm operation.

The channel checking unit 111 checks channel states of streams for transmitting data to receivers located in a service area. For example, the channel checking unit 111 estimates channel states of the streams using sounding signals transmitted by the receivers through an uplink channel. In another embodiment, the channel checking unit 111 checks the channel state included in the feedback signals provided from the receivers.

The beamforming vector selector 117 selects a reference transmission beamforming vector (m _j ) value according to Equation 3 below and provides it to the pre-encoder 107. That is, the receiver selects a value of the transmission beamforming vector for increasing the beamforming signal strength for the radio channel and determines it as the reference transmission beamforming vector.

Here, H _k is a channel matrix of the k-th terminal, m _j is a j-th transmission beamforming vector value.

In order to form a reference beam, the precoder 107 multiplies the first transmission signal by a reference transmission beamforming vector (m _j ) value from the beamforming vector selection unit 117 to a corresponding RF processor ( 109). In addition, the pre encoder 107 generates a second pilot signal as follows and outputs the second pilot signal to the RF processor 109 so as to compensate channel distortion for the signal received from the corresponding receiver.

The pre-encoder 107 pre-encodes the normalized transmission beamforming vector values provided from the second beamforming vector generator 115 and the data stream to be transmitted through each antenna provided from the demultiplexer 105. To print. That is, the pre encoder 107 multiplies the data to be transmitted in each stream by the normalized transmission beamforming vector provided from the second beamforming vector generator 115 to form a beam for each stream. This can be expressed as Equation 4 below. Here, the first encoder 107 outputs the first antenna signal to the first RF processor 109-1, and outputs the N _t -th antenna signal to the N _t RF processor 109 -N _t .

Where m _k ^R is a normalized transmission beamforming vector value of the k th antenna, b _k is a data stream transmitted through the k th antenna, and β is a factor for normalizing the transmission power.

The RF processors 109-1 to 109-N _t convert the data stream provided from the pre encoder 107 into an analog signal. Thereafter, the RF processors 109-1 to 109-N _t convert the analog signal into a radio frequency (RF) signal that can be actually transmitted and transmit the same through an antenna.

2 illustrates a receiver for performing normalized joint beamforming in a multiple antenna system according to the present invention.

Referring to FIG. 2, the receiver includes an RF processor 201, a post coder 203, a channel estimator 205, an effective channel gain calculator 209, a channel equalizer 211, a demodulator 213, The decoder 215 is comprised.

The RF processors 201-1 through 201 -N _R convert and output a radio frequency (RF) signal received through each antenna into a baseband signal. The received signal includes a signal by a data subcarrier and a signal by a pilot subcarrier.

The channel estimator 405 estimates a channel with a transmitter and calculates a reception beamforming vector by using a pilot included in a signal provided from the RF processors 401-1 through 401-N _R. Here, the pilot is divided into a first pilot and a second pilot, wherein the first pilot is used to form a reference beam at the transmitter to estimate the received beamforming vector, and the second pilot is used to compensate for channel distortion at the receiver. To be used. The received beamforming vector value is calculated as in Equation 5 below.

이 되고, 기준 빔을 사용하지 않는 수신기일 때(k≠j), 수신 빔포밍 벡터는 f(ㆍ)에 의한 직교 벡터로 산출된다. 상기 f(ㆍ)는 2x1 벡터에 대해서 하나의 직교 벡터를 출력하는 함수이다. 예를 들어, [p, q]^T 인 벡터에 대해 쉽게 [q*, -p*]^T 와 같은 직교 벡터를 구할 수 있다. y_{k , pilot -A} 는 제 1 파일럿을 수신한 신호이다.Here, when the receiver uses the reference beam (j = k), the reception beamforming vector is

When the receiver does not use the reference beam (k? J), the reception beamforming vector is calculated as an orthogonal vector by f (·). F (·) is a function for outputting one orthogonal vector to a 2 × 1 vector. For example, an orthogonal vector such as [q *, -p *] ^ T can be easily obtained for a vector with [p, q] ^ T. y _{k , pilot -A} is the signal that received the first pilot.

The effective channel gain calculator 209 calculates an effective channel gain for the second pilot signal by using the calculated reception beamforming vector value from the channel estimator 205. The effective channel gain is expressed by Equation 6 below.

는 k 번째 수신 빔포밍 벡터값이고, (ㆍ)^H는 허미션(Hermitian) 연산이고, y_{k , plot -B}은 제 2 파일럿에 의한 수신신호, β는 송신전력을 일반 화(normalization)하기 위한 인자, m_k ^R은 k 번째 안테나의 정규화된 송신 빔포밍 벡터값,

는 k 번째 단말의 수신빔포밍 벡터에서의 위상에러이다.here,

Is the k th received beamforming vector value, (·) ^H is a Hermitian operation, y _{k , plot -B} is the received signal by the second pilot, and β is for normalizing the transmission power. Where m _k ^R is the normalized transmit beamforming vector of the k th antenna,

Is a phase error in the reception beamforming vector of the k-th terminal.

The channel equalizer 211 equalizes the effective channel gains from the effective channel gain calculator 209 through the inverse.

The back encoder 203 detects the received data by using the valid channel information provided from the channel equalizer 211. This is shown in Equation 7 below.

는 k 번째 수신 빔포밍 벡터값이고, (ㆍ)^H는 허미션(Hermitian) 연산이고, y_k은 데이터 부반송파에 의한 수신신호, n_k'는 잡음이다.Where here

Is the k-th received beamforming vector value, (·) ^H is a Hermitian operation, y _k is a received signal by the data subcarrier, and n _k 'is noise.

The demodulator 213 demodulates the signal received from the post coder 203 according to the modulation level (Modulation and Coding Scheme (MCS) level).

The decoder 215 detects original data by decoding a signal provided from the demodulator 213 according to a corresponding modulation level.

3 illustrates a transmitter operation for performing normalized co-beamforming in a multi-antenna system according to the present invention.

Referring to FIG. 1, in step 301, the transmitter checks channel states of streams for transmitting data to receivers located in a service area. For example, the transmitter may estimate a channel state of the streams using a sounding signal transmitted by the receivers through an uplink channel, or check a channel state included in a feedback signal provided from the receivers.

Thereafter, in step 303, the transmitter generates the transmission / reception beamforming vectors using the channel state information of each stream. The generated transmission beamforming vector values are generated by using the channel state information in a conventional iterative approach. That is, the transmission and reception beamforming vector is repeatedly updated from the initial beamforming vector, and is performed until the value satisfies the threshold.

Thereafter, the transmitter normalizes the transmission beamforming vector values generated in step 305. By normalizing the transmission beamforming vector, an increase in noise due to user interference limitation may be prevented (see Equation 2).

Thereafter, in step 307, the transmitter multiplies the reference transmission beamforming vector by the first pilot and performs a precoding to form reference beamforming. In this case, the reference transmission beamforming vector is selected according to Equation 3 above.

Thereafter, in step 309, the transmitter transmits a second pilot signal to enable the receiver to compensate for channel distortion.

Thereafter, the transmitter transmits a data stream to be transmitted through each antenna using the normalized transmission beamforming vector in step 311.

The transmitter then terminates a normalized joint beamforming procedure.

The operation of the transmitter for applying normalized joint beamforming in a common pilot based system such as 3GPP LTE is shown in FIG. 4.

4 shows a flow diagram of a transmitter operation for performing normalized joint beamforming in a multiple antenna system according to the present invention.

Referring to FIG. 4, in step 401, the transmitter derives a transmission / reception beamforming vector for all codewords (reference transmission beamforming vector) of a codebook, and sums the sums of the double data rates. Find the codebook that maximizes the rate.

Thereafter, the transmitter obtains a normalized transmission beamforming vector as in FIG. 3 using the transmission beamforming vector found in step 403.

Thereafter, the transmitter transmits a transmission beamforming vector normalized to compensate for channel distortion at the receiver through port 5 of LTE, which is used for a dedicated pilot in step 405, and calculates an optimal reference beam index calculated above ( The reference beam index is transmitted to the receiver through a control channel.

Thereafter, in step 407, the transmitter transmits beamformed data through a normalized transmission beamforming vector through ports 0 to 3.

In this case, the receiver may calculate its own received beamforming vector using the reference beamforming index received through the control channel, and obtain an effective channel gain using the signal received through port 5. After receiving the data using the reception beamforming vector, equalization is performed through the effective channel gain to obtain a sufficient statistic of the data.

The transmitter then terminates a normalized joint beamforming procedure.

5 illustrates a flow diagram of a receiver operation for performing normalized joint beamforming in a multiple antenna system according to the present invention.

Referring to FIG. 5, the receiver estimates an effective channel using a dedicated pilot tone in step 501.

In operation 503, the receiver determines whether the user is a reference beam user. Each receiver may know whether it is a reference beam user (ie, a terminal using a reference transmission beamforming vector) through the control information MAP.

이 되고, 기준 빔을 사용하지 않는 수신기일 때(k≠j), 수신 빔포밍 벡터는 f(ㆍ)에 의한 직교 벡터로 산출된다. 상기 f(ㆍ)는 2x1 벡터에 대해서 하나의 직교 벡터를 출력하는 함수이다. 예를 들어, [p, q]^T 인 벡터에 대해 쉽게 [q*, -p*]^T 와 같은 직교 벡터를 구할 수 있다. y_{k , pilot -A} 는 제 1 파일럿을 수신한 신호이다.Thereafter, the receiver calculates a reception beamforming vector from the reference transmission beamforming vector for the signal of the first pilot (see Equation 5). That is, when the receiver uses the reference beam (j = k), the reception beamforming vector is

When the receiver does not use the reference beam (k? J), the reception beamforming vector is calculated as an orthogonal vector by f (·). F (·) is a function for outputting one orthogonal vector to a 2 × 1 vector. For example, an orthogonal vector such as [q *, -p *] ^ T can be easily obtained for a vector with [p, q] ^ T. y _{k , pilot -A} is the signal that received the first pilot.

Thereafter, the receiver calculates an effective channel gain for the second pilot signal from the received beamforming vector calculated in step 507 (see Equation 6).

In step 509, the receiver receives data using the reception beamforming vector and performs channel equalization (see Equation 7).

The receiver then terminates the normalized cobeamforming procedure.

As described above, in obtaining a transmission and reception beamforming vector for performing coordinated beamforming, by obtaining a transmission beamforming vector for minimizing a system-wide mean squared error (MSE), interference is obtained. The problem of noise increase due to the limitation can be solved. In addition, a pilot use technique may be applied to reduce overhead when performing joint beamforming.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described. However, various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the scope of the following claims, but also by the equivalents of the claims.

1 is a transmitter for performing normalized joint beamforming in a multi-antenna system according to the present invention.

2 is a receiver for performing normalized joint beamforming in a multi-antenna system according to the present invention;

3 is a flowchart illustrating a transmitter operation for performing normalized joint beamforming in a multi-antenna system according to the present invention;

4 is a flowchart illustrating a transmitter operation for performing normalized joint beamforming in a multi-antenna system according to the present invention;

5 is a flow diagram of a receiver operation for performing normalized joint beamforming in a multiple antenna system in accordance with the present invention.

Claims (16)

A transmitter for normalized co-beamforming in a multi-antenna system, A first beamforming vector generator for calculating a transmission beamforming vector using channel information; A second beamforming vector generator for normalizing the transmission beamforming vector; And a pre- encoder for precoding and transmitting data using the normalized transmission beamforming vector. The method of claim 1, The first beamforming vector generator And transmitting and receiving the beamforming vector repeatedly from the initial beamforming vector until the condition is satisfied. The method of claim 1, The second beamforming vector generator A transmitter characterized by normalizing a transmission beamforming vector that minimizes system-wide mean squared error (MSE). The method of claim 3, wherein And the transmission beamforming vector is normalized through Equation (8).

Here, M _reg is a normalized transmission beamforming vector, I is a unit matrix or unit vector, M is a transmission beamforming vector provided from the first beamforming vector generating unit 113, and (·) ^H is a mission (Hermitian) operation, (·) ^-1 is inverse operation, P is transmit power, N _o is noise power, H _k is channel matrix of k-th terminal, m _k is k-th transmission Beamforming vector value,

Means the Frobenius norm operation. The method of claim 1, And a beamforming vector selection unit for selecting a reference transmission beamforming vector to form a reference beam among the calculated transmission beamforming vectors. The method of claim 1, The first encoder In order to form a reference beam, a reference transmission beamforming vector (m _j ) value from the beamforming vector selector is multiplied by a first pilot signal, And multiplying the normalized transmit beamforming vector by a second pilot signal to compensate for channel distortion for the signal received at the receiver. In the receiver for normalized co-beamforming in a multi-antenna system, A channel estimator for calculating a received beamforming vector from the first pilot signal beamformed by the reference transmit beamforming vector; An effective channel gain calculator for calculating an effective channel gain from the calculated received beamforming vector; And a back encoder performing channel equalization using the effective channel gain and receiving data. The method of claim 7, wherein The channel estimator, And a receiver beamforming vector is calculated through Equation (9).

In a transmission method for normalized co-beamforming in a multi-antenna system, Calculating a transmission beamforming vector using channel information; Normalizing the transmission beamforming vector; And precoding data using the normalized transmission beamforming vector and transmitting the data. The method of claim 9, The transmission beamforming vector is And transmitting and receiving beamforming vectors are repeatedly updated from the initial beamforming vector until the predetermined conditions are satisfied. The method of claim 9, The process of normalizing the transmission beamforming vector, system-wide Mean Squared Error (MSE). The method of claim 11, The transmission beamforming vector is normalized through Equation 10 below.

Here, M _reg is a normalized transmission beamforming vector, I is a unit matrix or unit vector, M is a transmission beamforming vector provided from the first beamforming vector generating unit 113, and (·) ^H is a mission (Hermitian) operation, (·) ^-1 is inverse operation, P is transmit power, N _o is noise power, H _k is channel matrix of k-th terminal, m _k is k-th transmission Beamforming vector value,

Means the Frobenius norm operation. The method of claim 9, Selecting a reference transmission beamforming vector to form a reference beam from the calculated transmission beamforming vector. The method of claim 9, During the process of precoding and transmitting data using the normalized transmission beamforming vector, In order to form a reference beam, the first pilot signal is multiplied by a value of the beamforming vector selection unit (reference transmission beamforming vector m _j ) from And multiplying the normalized transmit beamforming vector by a second pilot signal to compensate for channel distortion for the signal received at the receiver. In the receiving method for normalized co-beamforming in a multi-antenna system, Calculating a reception beamforming vector from the first pilot signal beamformed by the reference transmission beamforming vector; Calculating an effective channel gain from the calculated received beamforming vector; Performing channel equalization using the effective channel gain and receiving data. The method of claim 15, The channel estimator, A method of calculating a reception beamforming vector through Equation 11 below.

Images