KR20130013182A - Smart antenna system for rejection of coherent and incoherent interferences - Google Patents

Abstract

PURPOSE: A smart antenna system for removing coherent and incoherent interference is provided to effectively remove interference without arrival angle information about the coherent interference and depending on complex calculation. CONSTITUTION: A signal blocking part(120) generates a matrix from a sensor array. Desired signal component is removed from a received signal containing coherent interference in the matrix. A weighting vector calculation part(130) estimates interference subspace from the matrix outputted from the signal blocking part. The weighting vector calculation part finds a weighting vector by using the estimated interference subspace. A linear combination part(140) generates an array output by multiplying the signal data received from the sensor array by the weighting vector. [Reference numerals] (120) Signal blocking part; (130) Weighting vector calculation part; (140) Linear combination part

Description

Smart Antenna System for Rejection of Coherent and Incoherent Interferences

The present invention relates to smart antenna technology for effectively canceling coherent, incoherent interference.

Recently, with the development of various technologies and systems in the field of wireless and mobile communication, the spread of services is fast. Accordingly, the use of radio waves is frequent, and the problem of efficient use of frequency resources has emerged as an important issue.

The smart antenna system generates an array output by arranging a plurality of antenna elements in a specific shape and multiplying each antenna output by a complex weight to produce an array output. The smart antenna system generates a beam pattern according to a given signal environment. Can be changed as appropriate. The beam pattern can be formed to make a null in the direction of the interference signal while retaining the beam gain in a desired direction, thereby eliminating the interference signal, thereby efficiently using frequency resources using a smart antenna.

Application of smart antenna in mobile communication For example, base station array antennas can be beam-directed to a specific terminal desired to maintain high quality call quality with small transmission power while reducing other users' interference. The interference signal can be removed to increase the capacity. In particular, as the demand for data communication that requires greater power than voice in recent years, a strong interference signal problem has emerged.

Coherent interference is caused by multipath propagation or smart jamming, and its characteristics are very similar to the desired signal, making it difficult to remove. Spatial smoothing is widely known as a technique for removing coherent interference. Spatial smoothing is a method that uses a spatially smoothed correlation matrix, which has a disadvantage in that interference is not easily removed because a correlation between a desired signal and interference remains in the smoothing matrix. To improve this problem, a weighted spatial smoothing method has been proposed. However, in this method, since the angle of arrival information of coherent interference is required, there is a problem that cannot be applied when the angle of arrival information is not known, and it can be applied only in a very limited case.

Spatial smoothing and weighted spatial smoothing require inverse transformation of the matrix, and in the latter method, the implementation is complicated.

Republic of Korea Patent Publication No. 10-2003-0089215 (2003.11.21)

The present invention has been devised to solve the above problems, and when a coherent interference and an incoherent interference signal arrive at the sensor array, complex calculations without the angle of arrival information on the coherent interference, for example It is an object of the present invention to provide a smart antenna that effectively eliminates son interference without the inversion of the matrix.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

In the present invention, the smart antenna system uses a sensor array arranged with a plurality of sensors for receiving a signal, a signal blocking unit for blocking a specific signal among the signals received from the sensor array, and a matrix output from the signal blocking unit. And a weight vector calculator for calculating a weight vector, and a linear combiner for linearly combining the signals received by the sensor array using the weight vector to generate an array output.

The sensor array is composed of M sensors arranged at equal intervals.

The signal blocker composes an M × (p + 1) matrix having M and p + 1 sizes of rows and columns, respectively, from the signals received by the sensor array, and uses the angle of arrival information of the desired signal, This removed (M-1) x p matrix is generated. Where p is an integer determined in consideration of the number of coherent interferences and the total number of interferences.

The weight vector calculation unit calculates a basis for the interference subspace by using a (M-1) × p matrix from which a signal component generated by the signal blocking unit is removed, and calculates a vector orthogonal to the basis. After the calculation, the vector is used to obtain a weight vector.

The linear combiner multiplies element data of the weight vector corresponding to the data received from each sensor, and adds each multiplied result to generate an array output.

According to the present invention, it is possible to remove more interference signals and improve reception performance by minimizing the reduction of the array effective aperture size of the array for removing coherent and incoherent interference.

In addition, the smart antenna can be easily implemented without the inverse transformation of the matrix, and it can be applied to a wider field because information on the direction of the interference signal is not required.

1 is a block diagram showing the configuration of a smart antenna system according to an embodiment of the present invention.

In the following description of the present invention, if it is determined that the detailed description of the related public notice, function, or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. In addition, throughout the specification, when a part is said to "include" a certain component, it means that it may further include other components, without excluding the other components unless otherwise stated. .

For convenience, the technical content of the present invention will be described using a correlation matrix. In reality, the correlation matrix is not known, and a sample correlation matrix obtained from the received data is used. The present invention encompasses both correlation matrices and their estimation.

1 is a block diagram showing the configuration of a smart antenna system according to an embodiment of the present invention. Referring to FIG. 1, the smart antenna system of the present invention includes a sensor array 110, a signal blocking unit 120, a weight vector calculation unit 130, and a linear coupling unit 140.

개의 센서가 직선 상에 같은 간격으로 놓여져 있는 ULA(uniform linear array) 어레이이다. 센서 어레이에 수신된 신호 벡터를

와 같이 나타낼 수 있다. 여기서

는 시간을 나타내는 인덱스(index),

는 행렬의 전치(transpose),

는 수신 신호벡터

의

번째 요소이다. 수신신호는 원하는 신호, 간섭신호, 잡음의 합으로 주어진다. Sensor array 110

It is a uniform linear array (ULA) array with four sensors spaced at equal intervals on a straight line. The signal vector received by the sensor array

As shown in Fig. here

Is an index representing time,

Is the transpose of the matrix,

Receive signal vector

of

Second element. The received signal is given as the sum of the desired signal, interference signal and noise.

For convenience, it is assumed that the desired signal comes from the 0 ° direction. The m-dimensional steering vector for the desired signal has m elements and is given by a _{d , m} = [1,1, ..., 1] ^T as it comes from the direction. There is one desired signal, η _c coherent interference, η _u incoherent interference, where the total number of interference signals is η _c + η _u (≡η). Although the present invention can be applied to any incoherent interfering signal correlated with other interfering signals, for simplicity, it is assumed that the incoherent interfering signals are not correlated with each other.

The signal blocking unit 120 serves to block a desired signal component among the signals received from the sensor array 110.

The correlation matrix R _x for the received signal is defined as M x M matrix as R _x = E [ x (t) x ^H (t)]. Where E means expectation and H means Hermitian (complex conjugate transpose) operation. It is also possible to use a matrix from which noise is removed from R _x . The matrix without noise components is given by R _x -σ ² I , where σ ² is the variance of noise and I is the identity matrix. The matrix R _x -σ ² I is the correlation matrix for the desired signal and interference.

The signal blocking unit 120 does not need the entire correlation matrix R _x but uses a portion thereof. R _x represents the first m-column of the matrix consisting of R _{x, m.} The signal blocking unit 120 constructs an R _{x , p + 1} matrix from the received signal data. The size of the columns of the constructed matrix is p + 1, the values of p are described later. As mentioned earlier, if necessary, noise components can be removed from R _{x , p + 1} . The invention can be applied without removing the noise component from the R _{x, p + 1,} for simplicity of _description, R _{x, p + 1} is assumed to not having a noise component. R _{x , p + 1} contains a desired signal component, and is obtained as in the formula (1) x p matrix R ^f _I (1).

R ^f _I = B ^T _M R _{x , p + 1} B _{p +1} (1)

Where B _m is m × (m-1) is a matrix with, k-th column, k th element of B _m is 1, k + 1-th element, and -1, k-th column, the other elements are all zeros.

The signal blocking unit 120 transmits the R ^f _I obtained above to the weight vector calculation unit 130.

When the signal blocking unit 120 transmits R ^f _I , the weight vector calculation unit 130 obtains the D × K matrices R ^f _{DK , l} (l = 1,2, ..., L) from R ^f _I. Configure. Wherein R ^f _{DK, l} is the R ^f _{DK, l} in the matrix portion (submatrix) of _{^{R f I (1,1), (}} D, K) component of R ^f _I each of (l, l), (l + D-1, l + K-1). These L matrices are added to obtain R ^f _DK as shown in Equation (2).

(2)

(2)

D, K, and L values are described later.

Each column of the matrix R ^f _I lies in the M-dimensional interference subspace, and each column of the matrix R ^b _I whose position of the resulting matrix element after complex conjugated operation of R ^f _I is also the M-dimensional interference subspace. It lies in space. R ^b _I is given by equation (3).

(3)

(3)

Where J _m is m × m square matrix, as defined by equation (4).

(4)

(4)

That is, J _m is a matrix having only 1 cross diagonal element and all other elements 0. The order of elements is changed.

In analogy to the procedure obtained from the R ^f _DK R ^f _I, it is obtained from R ^b R ^b _{DK I} as in expression (5).

(5)

(5)

Wherein the D × K matrix R ^b _{DK, l} consisting of a portion of the matrix _{^{R b I (1,1), (}} D, K) element of R ^b each _I (l, l), (l + D-1 , l + K-1)

Once R ^f _DK and R ^b _DK are found, they are arranged side by side as in Eq. (6) to form R ^{f / b} _DK .

(6)

(6)

(≡L_s)과 같이, K는

(≡K_s)와 같이 선택한다. 이때, D는 D=M-L_s, p는 p=K_s+L_s-1과 같이 주어진다. 여기서

는

를 초과하지 않는 최대 정수를 의미한다. R ^{f / b} _DK is a D × 2K matrix. If 2L≥η _c and 2L≥η, and D≥η, the column vectors of R ^{f / b} _DK create a D-dimensional interference subspace. L is such that the column space of R ^{f / b} _DK creates a D-dimensional interference subspace.

(≡L _s ), K is

Choose (≡K _s ). Where D is given by D = ML _s and p is given by p = K _s + L _s −1. here

The

It means the maximum integer that does not exceed.

D-dimensional weight vector w _D is obtained using matrix R ^{f / b} _DK . Calculate w _D to be orthogonal to the thermal space of R ^{f / b} _DK as shown in equation (7).

(7)

(7)

처럼 주어진다. Where P ^⊥ _M = I - P _M , where P _M represents the projection matrix into the column space of the matrix M. In formula (7) if c1 is so that the first array in response to the desired signal direction to a scalar factor c ₁ is

Is given as

The array output can be obtained by applying the D-dimensional weight vector w _D to a sub array consisting of D sensors. However, in this case it is not effective because neither sensor is used to generate the array output. It would be desirable to use an M-dimensional weight vector so that the output can be generated using all M sensors. The M-dimensional weight vector w _M can be obtained from w _D as in Equation (8).

(8)

(8)

Where I _l is an M × D matrix that transforms w _D into an M-dimensional vector, given by Eq. (9).

(9)

(9)

On the right side of equation (9), 0 _lk (k = 1,2) is a zero matrix, where 0 _l1 is (l-1) × D and 0 _l2 is (MD-1 + 1) × D. , I is a D × D unit matrix. In Equation (9), the scalar factor C ₀ is determined from the condition of W ^H _M a _{d , M} = 1 if the response to the desired signal direction of arrival is 1.

As mentioned above, although this invention was demonstrated using the preferable Example, an Example is an illustration and is not limited. Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the invention and the scope of the rights set forth in the appended claims.

110 sensor array 120 signal interrupter
130 Weight vector calculator 140 Linear combination

Claims (7)

A sensor array in which M sensors for receiving signals are arranged at equal intervals,
A signal blocking unit generating a matrix R ^f _I from which a desired signal component is removed from a received signal including coherent interference from the sensor array;
A weighted vector calculator for estimating an interference subspace from the matrix R ^f _I output from the signal blocker, and obtaining a weight vector using the estimated interference subspace;
Smart antenna system comprising a linear coupling unit for generating an array output by multiplying the weight data by the signal data received from the sensor array.
The method of claim 1,
The signal blocking unit obtains R _{x , p + 1} which is an M × (p + 1) sub-matrix of the sample correlation matrix,
The integer p is determined by considering the number of coherent interference signals and the total number of interference signals.
A smart antenna system, characterized by generating R ^f _I , which removes a desired signal component from R _{x , p + 1} .
The method of claim 2,
The weight vector calculation unit extracts a D × K sub-matrix R ^f _{DK, l} (l = 1,2, ..., L) of the given matrix R ^f _I from the signal blocking unit,
The constants L, K, and D are determined by considering the number of coherent interference signals and the total number of interference signals.
Smart antenna system characterized by estimating the D-dimensional interference subspace using these R ^f _{DK , l} .
The method of claim 2,
The weight vector calculation unit extracts a D × K sub-matrix R ^f _{DK, l} (l = 1,2, ..., L) of the given matrix R ^f _I from the signal blocking unit,
A complex conjugate operation on the matrix R ^f _I and the complex conjugate of D to change the position of the matrix elements calculated, to generate a R ^b R ^b _{I I} × K sub-matrix _{^{R b DK, l (l =}} 1,2, .. Extract the., L)
The constants L, K, and D are determined by considering the number of coherent interference signals and the total number of interference signals.
Smart antenna system characterized by estimating the D-dimensional interference subspace using both R ^f _{DK , l} and R ^b _{DK , l} .
The method according to claim 3 or 4,
A smart antenna system, characterized by obtaining a D-dimensional weight vector w _D orthogonal to the estimated interference subspace.
The method of claim 5,
Smart antenna system, characterized in that to obtain the M-dimensional weight vector w _M using the D-dimensional weight vector w _D.
The method of claim 2,
And removing the noise component from the sample correlation submatrix R _{x , p + 1} and generating R ^f _I.

Images