KR101768362B1 - A method for estimation of channel state information in massive antenna-based wireless communication systems - Google Patents

Abstract

The present invention relates to a method for real-time estimation of users' channel information in a multi-user wireless communication system using a large-scale multi-antenna. The instantaneous channel information estimation method using the public pilot of the existing system is not easy to implement because a large number of pilot signals are required in a large scale multi-antenna system. In order to solve this problem, a method of estimating instantaneous channel information using spatial correlation information of a channel has been proposed, but an estimation method of spatial correlation information of a channel should be additionally considered. Accordingly, the present invention proposes a technique for significantly increasing the instantaneous channel information estimation efficiency by accurately estimating spatial correlation information of a channel using only a small pilot signal by using a far-field effect.

Description

[0001] The present invention relates to a method for estimating channel information in a multi-user wireless communication system using a large-scale antenna,

The present invention relates to a technique for estimating spatial correlation information of a channel using a far-reaching local effect in a wireless communication system using a large-scale antenna, and estimating instantaneous channel information based on the estimated spatial correlation information.

Wireless data usage due to the activation of wireless multimedia service due to the spread of smartphone and the surge of video traffic is creating a big data environment which is not easy to handle with the conventional transmission method. In order to secure future communication capacity, many studies have been conducted focusing on high density small cell, high frequency band communication, and increasing frequency efficiency. Recently, a massive multiple-input multiple output (M-MIMO) system has been considered as a technique for increasing the frequency efficiency. M-MIMO technology is attracting much attention as one of energy efficient green communications technologies because of its low energy and high frequency efficiency.

The M-MIMO system can utilize a high degree of freedoms from many antenna uses and can achieve high communication capacity by simultaneously serving a plurality of users through various spatial multiplexing techniques. However, in order to maximize such a gain, a base station must acquire instantaneous channel information of a plurality of users. The existing LTE-A system allocates orthogonal frequency division multiplexing (OFDM) orthogonal resources to all the antennas used by the base station, transmits the pilot signal, and users estimate the channel using the received pilot signal and feedback . However, in the case of the M-MIMO system, as the number of antennas increases, the number of pilot signals required for channel estimation increases significantly and orthogonal resources for transmitting data may be inefficiently consumed. On the other hand, the time division duplex (TDD) system can easily obtain the downlink channel information through the sounding signal using the uplink and downlink channel reciprocity. However, the FDD (Frequency Division Multiplexing) division duplex system can not utilize the homology of the channel, it is necessary to fundamentally solve the problem of increasing the pilot signal. In order to solve this problem, a pilot signal design capable of effectively estimating a channel using only a small pilot signal using channel correlation information of users is considered, and eigen-space channel estimation is a representative example . However, in order to utilize the channel estimation technique, it is necessary to additionally consider the channel correlation information in addition to the instantaneous channel information.

In order to estimate the channel correlation information, an orthogonal resource is allocated to each of the base station antennas, a pilot signal is transmitted, and a least square (LS) estimation technique using the orthogonal resources is considered. However, when the dimensions of an inter-antenna spatial correlation matrix (hereinafter referred to as SCM) of a channel to be estimated, such as an M-MIMO system, are very large, a large number of corresponding received signal samples So that a large amount of OFDM resource consumption due to the pilot signal is inevitable.

delete

Accordingly, the present invention proposes a SCM estimation method that can achieve a high SCM estimation performance using a small number of pilot signals in order to solve the above-described problem. The present invention improves the accuracy of SCM estimation through improved LS estimation using the far-field effect and then removes the estimated correlation information components with low accuracy, thereby obtaining only a few SCM components, which are significant for channel estimation, And to provide an estimation performance.

In order to solve the above-mentioned problems, according to an embodiment of the present invention, in a frequency division duplex-based multi-user wireless communication system

Lt; RTI ID = 0.0 > UE < / RTI >

A method for estimating instantaneous channel information of a user, the method comprising the steps of: (A) estimating instantaneous spatial correlation matrix (SCM)

Periodically transmitting an SCM estimation pilot signal composed of a plurality of orthogonal signals to the user periodically prior to transmission of a normal instantaneous channel pilot signal; (B) receiving the pilot signal for SCM estimation and generating a far- field effect, and (C) receiving the estimated SCM from the BS by the BS.

Also, according to a preferred embodiment, in the step (A), the base station transmits a pilot signal corresponding to a pilot signal of each antenna component

Lt; / RTI > signals using orthogonal transmission resources

(Hereinafter, referred to as " P _SCM ") of a matrix structure, and generating an SCM estimation pilot signal (hereinafter referred to as " below T _SCM) the decision process, a transmission frequency of the _SCM P in consideration of the transmission resources by the base station to allow the

The base station transmits the P _SCM at the determined T _SCM period

And transmitting the data to the mobile station in succession.

Further, according to a preferred embodiment, the base station transmits the number P of the _SCM in consideration of the amount of resources and its SCM estimation accuracy according to estimation using the SCM

The ratio of the amount of actual data transmission resources to the total amount of available resources of the base station per unit time is calculated as

, A given threshold < RTI ID = 0.0 >

About,

Minimum

.

Also, according to a preferred embodiment, the step (B)

SCM

And the user

end

_Lt; RTI ID = 0.0 _> P _{SCM < /} RTI _>

About

And,

of

Second element

, antenna

And antenna

The distance vector of

, remind

Lt; RTI ID = 0.0 > a < / RTI >

, Set

The number of elements in

In other words,

of

And

Component of the third column

of

, And transmitting the corrected SCM to the base station after the user corrects the element values with low estimation accuracy among the elements of the estimated SCM.

Further, according to a preferred embodiment, the estimated element

And the step of transmitting the corrected SCM to the base station after the user corrects the element values with low estimation accuracy in the estimated SCM,

Using

If the estimated accuracy error is larger than when the estimated error is large,

And correcting the estimated value using the estimated value,

ego

An antenna

The far-field effect, in which the spatial correlation between the two antennas and the estimation accuracy are lowered as the distance between the two antennas increases,

As shown in FIG.

Further, according to a preferred embodiment, the estimated element

Estimation accuracy error using

If the estimated error is larger than the estimated error,

, The process of correcting the estimated value is performed so that the average is 0 and the variance is

When the user receives the pilot signal in an additive noise environment of a normal distribution,

and

The average time / frequency correlation of the antenna-specific channels

, The estimated

The mean square error (MSE)

To

, And a step of calculating

To

The MSE when

When you say

, And considering the weight according to the frequency of spatial correlation estimation,

For all elements of

If

As shown in FIG.

In an FDD wireless communication system using a large-scale multi-antenna, accurate SCM estimation is performed using only a small number of SCM estimation pilot signals using an effective SCM estimation using a remote area effect, and a pilot signal is generated based on the estimated SCM We propose an instantaneous channel information estimation method suitable for large scale multi - antenna usage environment by estimating instantaneous channel information.

1 is a diagram illustrating an example of a process in which a base station estimates users' real-time channels in an M-MIMO system model considered in the present invention.
2 is a time-based diagram illustrating signal exchange between a base station and a user for the real-time channel estimation process;
Figure 3 is a simplified diagram of the process for estimating a user's CSI;
FIG. 4 is a detailed view showing a correction step of step 303 of FIG. 3. FIG.

1 is a diagram illustrating an example of a process of a base station estimating users' real-time channels in an M-MIMO system model considered in the present invention. The base station 101 uses a uniform large-scale antenna arrangement and uses the P _SCM and the _SCCH to estimate the SCM and instantaneous channel state information (CSI)

(103). The users estimate SCM and CSI based on the received pilot signal and feed back 104 to the base station.

FIG. 2 is a time-based diagram illustrating a signal exchange between a base station and a user for the real-time channel estimation process.

Since the channel correlation information acquisition is to be pre-established, the base station has a step 203 of estimating the SCM of all users prior to the channel estimation. In this step,

&_Lt; / RTI > of the orthogonal signal vectors < RTI ID = 0.0 >

≪ / RTI >

RTI ID = 0.0 > OFDM < / RTI > orthogonal resources.

Is determined in consideration of the pilot dedicated resource overhead allowed to the base station. For example, the ratio of the actual available data transmission resource amount to the total available resource amount of the base station per unit time

, A given threshold < RTI ID = 0.0 >

About

Maximum

Can be determined as the number of consecutive transmissions of the P _SCM . At this time,

The SCM estimation pilot signal for the base vector < RTI ID = 0.0 >

The pilot signal < RTI ID = 0.0 >

silver

Consisting of

Size signal matrix. As a simple example

The

The second element is 1 and the remaining elements are 0

silver

Size identity matrix. The base station then feeds back the SCM estimated by the users

And proceeds with the CSI estimation for the users (204). At this time

The

Pilot signal vectors < RTI ID = 0.0 >

Size pilot signal matrix, and in the existing system

to be. Since the channel correlation information is closely related to the location of users, the channel correlation information has a characteristic that changes relatively slowly compared with CSI. The period T SCM for updating the _SCM can be set to be very long as compared with the period for estimating the CSI, and the additional consumption of OFDM resources due to the SCM estimation can be minimized. However, the longer the T _SCM , the greater the error between the previously estimated SCM and the actual SCM of the current user, resulting in an increase in the instantaneous channel estimation error. Therefore, it is necessary to determine an appropriate T _SCM so as to maintain the instantaneous channel estimation performance satisfying the required capacity of users while minimizing the OFDM resources consumed in the SCM update considering the change of the SCM according to the mobility of the users. The base station _{adjusts the c} to the T _SCM determined in the above process,

And the SCM of the users is estimated.

FIG. 3 is a diagram briefly illustrating the above process of estimating a user's CSI. In step 301 the base station transmits the P _SCM, and the user can utilize the remote location based on the effect the _SCM P received in step 302 by performing a modified LS estimate for the SCM in order to estimate the SCM of the user. gun

When there are a number of users,

Can be represented as Equation (1) below. ≪ EMI ID = 1.0 >

here

Me

User for

Lt; / RTI >

The

th

The base station and user experiencing this

Between

The channel vector,

Each element has an average of 0 and variance

Additive white Gaussian noise < RTI ID = 0.0 >

Represents a noise vector. Assuming that the SCM of the user does not change while the P _SCM is being transmitted,

SCM

(2) < / RTI >

In this case, since the distance between the antennas of the base station is very short compared to the distance between the base station and the user, the two antenna pairs having the same distance and direction exhibit a very similar channel spatial correlation regardless of the position of the antenna. At this time,

,

The channel spatial correlation between

of

And

Which corresponds to the element in the third column. This

. The antenna pair

The distance vector of

, The set of antenna pairs with the same distance vector

The antenna pair < RTI ID = 0.0 >

&Quot; (3) &quot;

From this relationship,

(P _SCM is composed of orthogonal signals, so there is an inverse matrix), and the user

And the SCM estimated by Equation (4)

Can be obtained.

here

Antenna antenna pair

The estimated value of the channel spatial correlation of

Is a set

Lt; / RTI &gt;

The

of

, And * denotes a complex conjugate. Step 303

Is expected to be low,

By applying

. This step consists of steps 401 to 404 in Fig. 4 in detail. Step 401

And calculates a mean square error (hereinafter referred to as MSE) of Equation (5), which can be expressed as Equation (5).

here

The

of

Lt; / RTI &gt; At this time, the farther the distance between the two antennas is, the smaller the number of antenna pairs having the same distance as the corresponding distance,

And the accuracy of the estimated spatial correlation of the antenna pair is lowered. On the other hand, in order to calculate Equation (5)

There is a problem that the user must know in advance. In order to solve this problem, the distance effect is used as shown in Equation (6)

Can be approximated.

here

Channel

and

And is easily obtained from Equation (1) in view of the characteristics of a large scale multi-antenna arrangement.

Estimates of expected MSE

In other words,

Can be expressed as Equation (7) by substituting Equation (6) into Equation (5).

At this time

The conditions for correcting

Is compared with the MSE at the time of correction, and the former is larger than the latter.

(7) is used, and the MSE at the time of correction is expressed by Equation

, It can be calculated as shown in Equation (8) in Step 402.

The result is that the actual MSE when correcting the estimated spatial correlation is < RTI ID = 0.0 >

. Therefore, the estimated spatial correlation correction process can be expressed as Equation (9).

However, this only considers the accuracy of the estimated spatial correlation. That is, since the influence of the estimation accuracy varies depending on the magnitude of the spatial correlation, a weight is required considering this. At this time,

The bigger

Becomes an accurate value,

A strict weighting of the spatial correlation magnitude is required. Using this,

Rather than using it again,

The correction process of Equation (10) is performed in step 403.

The calibration process

By applying to all elements of

The

Sparse < / RTI >

Can be obtained. In this case, as the distance between the antennas is larger, the space correlation is smaller.

Can be replaced with zero. That is, in step 404,

in

ego

An antenna

If so,

. In step 304,

Obtained by the above process

To the base station. In step 305,

SCMs received from users

And transmits it to the users.

Is an arbitrary real time channel estimation pilot signal that utilizes spatial correlation information of users, assuming one user as an example,

Can be expressed simply as Equation (11).

For convenience, the user index is ignored,

The

&Lt; / RTI >

The

From the first column vector of

Lt; th &gt; column vector. In step 306,

And obtains CSI using the MMSE estimation method and feeds back the CSI to the base station.

Assuming that the channels of users are kept constant while being transmitted,

Can be expressed by Equation (12). &Quot; (12) &quot;

here

User

Lt; / RTI &gt;

The

Base stations and users

Between

The channel vector,

Is the average of 0

Additive white Gaussian noise &lt; RTI ID = 0.0 &gt;

Represents a noise vector. Therefore, the channel estimated through the MMSE estimation technique

Is expressed by Equation (13).

here

The

Size equality matrix. Subsequent users

Using the commercially available channel information feedback technique

To the base station, thereby completing the channel estimation process. Although the embodiments of the present invention have been described above, the embodiments disclosed in this document are provided for the explanation and understanding of the disclosed technical contents, and do not limit the scope of the technology described in this document. Accordingly, the scope of this document should be interpreted to include all modifications based on the technical idea of this document or various other embodiments.

101: base station
102: User
SCM: Interchannel Spatial Correlation Matrix
CSI: Real-time channel information
P _SCM : SCM estimation public pilot
P _CSI : CSI Estimate Public Pilot

Claims (6)

In multi-user wireless communication systems based on frequency division duplex

Lt; RTI ID = 0.0 &gt; UE &lt; / RTI &gt;

A method for estimating instantaneous channel information,
(A) the base station estimates an interchannel spatial correlation matrix (SCM) information for each of the base station antennas of the users

Periodically transmitting an SCM estimation pilot signal composed of a plurality of orthogonal signals to the user prior to transmission of a normal instantaneous channel pilot signal,
(B) receiving the pilot signal for the SCM estimation by the user and estimating the SCM using the similarity between the spatial correlation information by the far-field effect,
(C) receiving the estimated SCM from the BS by the BS.
The method according to claim 1,
The step (A)
And the base station transmits a pilot signal corresponding to a pilot signal of each antenna component

Lt; / RTI &gt; signals using orthogonal transmission resources

A process of generating the SCM estimation pilot signal (hereinafter, referred to as P _SCM )
Determining a SCM estimation cycle (hereinafter referred to as T _SCM ) at a value equal to or less than a maximum time interval at which the base station can maintain coherence in a time band of the _SCM ,
The number of transmissions of the P _SCM in consideration of the allowed transmission resources

, &Lt; / RTI &gt;
The base station transmits the P _SCM at the determined T _SCM period

And transmitting the channel information to the base station.
3. The method of claim 2,
The number of transmissions of the _SCM in consideration of the amount of resources used by the BS in SCM estimation and the SCM estimation accuracy thereof,

In the process of determining,
The ratio of the actual data transmission resource amount to the total available resource amount of the base station per unit time

, A given threshold &lt; RTI ID = 0.0 &gt;

About,

Minimum

And estimating the channel information.
The method according to claim 1,
The step (B)
user

SCM

And the user

end

_Lt; RTI ID = 0.0 _&gt; P _{SCM &lt; /} RTI _&gt;

About

And,

of

Second element

, antenna

And antenna

The distance vector of

, remind

Lt; RTI ID = 0.0 &gt; a &lt; / RTI &gt;

, Set

The number of elements in

In other words,

of

And

Component of the third column

of

, &Lt; / RTI &gt;
And correcting element values with low estimation accuracy among the elements of the estimated SCM by the user.
5. The method of claim 4,
The step of the user correcting the element values with low estimation accuracy among the elements of the estimated SCM,
The estimated element

Estimation accuracy error using

, It is determined that the estimation accuracy error is large,

By

A process of correcting the value,
After the estimation value correction process is completed,

ego

An antenna

The far-field effect, in which the spatial correlation between the two antennas and the estimation accuracy are lowered as the distance between the two antennas increases,

And correcting the channel information.
6. The method of claim 5,
The estimated

Using

If the estimated accuracy error is larger than when the estimated error is large,

Wherein the step of correcting the estimated value comprises:
If the average is 0 and the variance is

When the user receives the pilot signal in an additive noise environment of a normal distribution,

and

The average time / frequency correlation of the antenna-specific channels

, The estimated

The mean square error (MSE)

To

, &Lt; / RTI &gt;
The estimated

To

The MSE when

When you say

, &Lt; / RTI &gt;

doggy

The average correlation between the transmitted channels

Considering the weight according to the number of spatial correlation estimates,

For all elements of

If

And correcting the channel information.

Images