KR20020037965A - System and method for downlink beamforming using uplink array response vector - Google Patents

Abstract

PURPOSE: A down link beam forming system and a method thereof are provided to transmit a maximal power to a direction in which a desired terminal is located using a receiving array response vector obtained in a uplink. CONSTITUTION: An adaptive array modulator(105) generates a reference signal necessary to estimate a receiving array response vector. A correlator(106) estimates the receiving array response vector by obtaining an average with respect to a time of a signal obtained by multiplying the reference signal by digital receiving data. A transmission beam form weight calculator(110) removes an uplink channel component from the receiving array response vector to obtain a receiving antenna response vector. The transmission beam form weight calculator(110) calculates a transmission beam form weight using the receiving antenna response vector. A down link beam form device(107) multiplies the beam form weight by a user signal to perform a down link beam form function.

Description

System and method for downlink beamforming using uplink array response vector using backward array response vector

The present invention relates to a forward beamforming system and a method using a reverse array response vector, and more specifically, to a transmission having a maximum gain in a direction in which a desired user is located and a low side lobe level in order to reduce interference signals to other users. A forward beamforming system and method thereof for forming an antenna beam.

In the adaptive array antenna base station system, there is a 'maximum signal-to-interference ratio' method in which a signal is transmitted to a desired user while minimizing signal power transmitted in a direction in which another user is located using a forward beamforming technique. This method requires a spatial correlation matrix for all users, so when the number of users increases, the amount of calculation increases rapidly. In addition, if the number of users is much larger than the number of array antenna radiating elements, the advantage of reducing the interference signal by forming a null in the transmitting antenna beam is sufficiently reduced.

Therefore, the maximum signal-to-noise ratio method, which maximizes the signal-to-noise ratio in the terminal by maximizing the gain of the transmitting antenna in the direction in which the desired user is located, does not consider the direction of other users at all, even though the performance is somewhat degraded. In addition, since the implementation is simple, it is suitable for an adaptive array antenna base station system. This 'maximum signal-to-noise ratio' method utilizes the transmit antenna response vector as the forward beamforming weight vector, and the beamforming weight is multiplied by the window or sidelobe level adjustment coefficient to reduce the sidelobe level in the formed antenna pattern. It has been widely used in other applications using antennas.

Meanwhile, there are various problems in the conventional 'maximum signal-to-noise ratio' method.

First, in the FDD scheme, since the transmit and receive frequencies are different, the antenna response vectors are different accordingly. In general, the forward beamforming algorithm often uses the antenna response vector obtained in the reverse direction, but there is a problem in that the change of the antenna response vector according to the frequency is considered.

Second, in order to reduce the signal to other users, the side beam level of the transmission beam should be lowered using the window. Since the beamforming weight vector multiplied by the window lowers the side lobe level of the transmit beam, it has been widely used in other array antenna applications. However, if the existing window is used as it is, the side lobe level of the formed beam is significantly lowered. However, in the case of the mobile communication base station system in which the main leaf width is greatly increased and the number of radiating elements of the array antenna is limited, the angle resolution is low. As a result, it is difficult to obtain a desired performance improvement.

Third, due to the gain of a single antenna, the array antenna gain in the sector is variable depending on the orientation angle. In a conventional base station system using a fixed beam antenna, an antenna is allocated for each sector. In this case, since the antenna directed to a specific sector should be designed to minimize interference with other sectors, there is a problem in that the antenna gain varies depending on an angle even within the directed sector.

Accordingly, the present invention is to solve the above problems of the prior art, an object of the present invention is to use a spread spectrum code division multiple access (CDMA) base station using an array antenna, the desired reception using the received response response vector obtained in the reverse direction It is an object of the present invention to provide a forward beamforming system and method for transmitting maximum power in a direction in which a terminal is located.

1 is a configuration diagram of a forward beamforming system and a peripheral device required to implement the same in a base station system according to an embodiment of the present invention;

FIG. 2 is a diagram illustrating the configuration of the beamforming weight calculator shown in FIG. 1;

3 is a diagram illustrating a relationship between beam width and sidelobe level according to a window when the antenna directivity angle is set to 0 degrees in a forward beamforming system using a reverse array response vector according to an embodiment of the present invention;

4 and 5 are diagrams showing that the array antenna gain is changed according to the direction angle due to the gain characteristics of the array antenna radiating element in the forward beamforming system using the reverse array response vector according to an embodiment of the present invention.

4 is a view showing an antenna pattern without gain loss compensation;

5 is a diagram illustrating an antenna pattern with gain loss compensation.

FIG. 6 is a configuration diagram of a forward beamforming weight calculator that calculates a forward beamforming weight by predicting an angle of arrival of each user using a reception array response vector.

7 is a view showing a window for controlling the mainlobe of the antenna beam,

8 is a view showing the strength of the received signal for each angle according to the window shown in FIG.

9 is a diagram illustrating a concept of improving an angle of arrival estimation according to an embodiment of the present invention.

※ Explanation of code about main part of drawing ※

101: array antenna 102: duplexer

103: multi-channel RF down converter 104: multi-channel RF down converter

105: adaptive array demodulator 106: cross correlator

107: forward beamformer 110: beamforming weight calculator

201: normalizer 202: antenna response converter

203: antenna response conversion matrix storage

601: receive antenna response matrix storage

602: angle of arrival estimator

603: Beamforming Weight Lookup Table

According to the present invention for achieving the above object, adaptive array demodulation means for generating a reference signal required for the reception array response vector estimation; Cross-correlation means for estimating a received array response vector by multiplying the reference signal generated by the adaptive array demodulation unit and the digital received data to obtain an average value over time; Transmission beamforming weight calculation means for receiving a reception array response vector, removing a reverse channel component to obtain a reception antenna response vector, and calculating transmission beamforming weights using the reception antenna response vector; And a forward beamforming means for performing a forward beamforming function by multiplying the beamforming weights with a user signal. The forward beamforming system using the reverse array response vector is provided.

In addition, a first step of generating a reference signal required for the reception array response vector estimation; A second step of estimating a reception arrangement response vector by obtaining an average value with respect to time of a signal generated by multiplying the reference signal generated in the first step by the digital reception data; A third step of obtaining a reception antenna response vector by removing a reverse channel component from the reception array response vector and calculating a transmission beamforming weight using the reception antenna response vector; And a fourth step of forming a transmission antenna beam by multiplying the transmission beamforming weights converted in the third step by a user signal, thereby providing a forward beamforming method using a reverse array response vector.

In addition, a first step of generating, in a computer, a reference signal necessary for estimating a reception array response vector; A second step of estimating a reception arrangement response vector by obtaining an average value with respect to time of a signal multiplied by the reference signal generated in the first step and the RF data; A third step of obtaining a reception antenna response vector by removing a reverse channel component from the reception array response vector and calculating a transmission beamforming weight using the reception antenna response vector; And a fourth step of multiplying the transmission beamforming weights converted in the third step by a user signal to form a transmission antenna beam, and providing a computer-readable recording medium having recorded thereon a program capable of being executed.

Hereinafter, a forward beamforming system and a method using a reverse array response vector according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a forward beamforming system and a peripheral device for implementing the same in a spread spectrum CDMA base station system according to an embodiment of the present invention, which is an array antenna 101, a duplexer 102, and a multi-channel RF. It consists of a down converter 103, a multi-channel RF up converter 104, an adaptive array demodulator 105, a cross correlator 106, a beamforming weights operator 110, and a forward beamformer 107.

In the present embodiment, the array antenna 101 is responsible for transmitting and receiving signals at the same time, but it is also within the scope of the present invention to design an array antenna for transmitting and receiving separately. In addition, although it is assumed that the transfer function characteristics of the multi-channel RF down converter 103 and the multi-channel RF up-converter 104 are the same, the present invention may be applied even if the transfer function characteristics are different.

The data received through the array antenna 101 is input to the adaptive array demodulator 105 after passing through the duplexer 102 and the multi-channel RF down converter 103. The adaptive array demodulator 105 provides the reference signal of the cross correlator 106 required for the received array response vector estimation. In this case, the reference signal is generated by using a pseudo noise code or a symbol fed back from the modem.

The cross correlator 106 obtains an average value of time multiplied by the reference signal and the measurement data obtained from the adaptive array demodulator 105 according to Equation 1 below.

Here, x (n) is a column vector received at the n th time, and d _i is a reference signal for the i th terminal.

The beamforming weight calculator 110 receives a result value of the cross correlator 106 to obtain a forward beamforming weight, and the forward beamformer 107 multiplies the beamforming weight by a user signal to perform forward beamforming. do.

The multi-channel RF upconverter 104 receives the result of the forward beam former 107 and performs RF conversion, and then transfers the result to the duplexer 102.

2 is a block diagram showing the configuration of the beamforming weight calculator 110 shown in FIG. 1 according to an embodiment of the present invention. The beamforming weight calculator 110 includes a plurality of normalizers 201 and a plurality of normalizers. An antenna response converter 202 and an antenna response conversion matrix memory 203.

The plurality of normalizers 201 remove the reverse channel components by normalizing the reception array response vector, r _{xd, i} ^' obtained from the cross correlator 106, as shown in Equation 2 below, thereby receiving the reception antenna response vector. Find v _i .

Here, v _i is a reception antenna response vector estimate of the i-th terminal.

The antenna response conversion matrix memory 203 stores the antenna response conversion matrix T. The antenna response conversion matrix converts the reception antenna response vector into a transmission beamforming weight vector, as shown in Equation 3 below.

Here, w _i is the forward beamforming weight vector for the i-th terminal, the antenna response conversion matrix T is a matrix for converting the reception array antenna response matrix A _r into a value close to the transmission beamforming weight target matrix B, and the transmission beamforming The weighted target matrix is a matrix composed of a target beamforming weight vector to approximate the reception antenna response vector.

Meanwhile, in Equation 3 above, T can be obtained by various optimization methods. If one of them is introduced, it can also be obtained using Pseudo-Inversion as shown in Equation 4 below. .

As can be seen from Equation 4 above, the antenna response conversion matrix is a function of the transmit / receive antenna response matrix and the beamforming weight target matrix, and thus can be applied to all received array response vectors.

On the other hand, the receiving array antenna response matrix, A _r and the beamforming weighted target matrix, B is expressed by Equation 5 below.

Where N _s is the number of sample angles measuring the antenna response vector, a _{r, i} is a vector representing the response of the receiving array antenna with respect to the i th angle, and b _i is a transmission antenna response vector for the i th angle. A vector multiplied by a window to reduce the side lobe level and a value to compensate for the gain of the array antenna radiating element, a _{t, i} is the transmit antenna response vector for the i th angle, Λ _i is a diagonal matrix, These are the values that make up the window.

On the other hand, the window may be set variably according to the orientation angle of the antenna. α _i is set so that the single antenna gain is inversely proportional to the single antenna gain as a value for compensating for the gain change depending on the array antenna orientation angle.

3 is a view showing the relationship between the beam width and the side lobe level according to the window when the antenna directivity angle is set to 0 degrees in the forward beam forming system according to an embodiment of the present invention.

The rectangular window has a relatively narrow width of the main lobe, but because of the high level of the sublobe level, it transmits a relatively large interference signal to other users.If the existing Hamming window is used, the sublobe level is low enough, but the main lobe width It becomes so wide that it turns out that the angle resolution of an array antenna falls. On the contrary, the beam pattern obtained by using 20% of both ends of the whole hamming window and only 60% of the middle part is obtained. The beam pattern is an intermediate step between the rectangular window and the hamming window in terms of the side lobe level and the main lobe width. It is more suitable for the environment.

That is, it can be seen that as the increase and decrease of the window function increases, the side lobe level decreases, but the degree of widening the main lobe width increases. Therefore, by appropriately adjusting the degree of increase and decrease of the window function, the side leaf level and the main leaf width can be properly adjusted.

4 and 5 illustrate an embodiment of the forward beamforming system shown in FIG. 2, wherein the array antenna gain is changed according to a direction angle due to the gain characteristics of a single antenna radiating element constituting the array antenna, respectively, and a transmission beam. A diagram showing compensation of such a change in gain by appropriately selecting the formation weighting target matrix, which will be described in detail as follows.

FIG. 4 is a diagram illustrating an antenna pattern without gain loss compensation. When the gain change of the radiating element is not compensated, the gain change according to the angle of the array antenna is almost similar to the gain change of the radiating element. In the case of the compensated antenna pattern, the gain of the array antenna is almost constant regardless of the angle. Therefore, according to the proposal of the present invention, by using the method of adjusting the forward beamforming weight vector target value, it can be seen that the gain according to the angle can be compensated without estimating the angle of signal arrival.

6 is a block diagram of a forward beamforming calculator for calculating forward beamforming weights by predicting the angle of arrival of each user using a reception array response vector. Is composed of an angle of arrival estimator 602 and a beamforming weight lookup table 603.

This method forms a transmit antenna beam with a low Sidelobe level with maximum gain in the direction of the desired user, regardless of the direction of the other user.

The beamforming weight lookup table 603 stores beamforming weights previously calculated for each angle in consideration of the transmission antenna response characteristics. Therefore, once the signal arrival angle estimation is completed, the forward beamforming weights obtain a beamforming weight corresponding to the arrival angle from the beamforming weight lookup table 603. The arrival angle estimator 602 removes the reverse channel component by normalizing the reception array response vector, r _{xd, i} ^' obtained from the cross correlator 106 as shown in Equation 2, thereby receiving the reception antenna response vector, v. After _i is obtained, as shown in Equation 6 below, the received antenna response matrix, A _r , is measured and correlated with each other to obtain the strength of the signal according to the angle, P _{VC, i} (θ), and compare with the threshold. To estimate the angle of arrival.

In the present invention, instead of calculating the intensity of the received signal at narrow angle intervals in the entire angle of arrival range as a method for reducing the amount of calculation when estimating the angle of arrival using such a vector correlation, the angle of arrival is obtained by obtaining the received signal strength at large angle intervals. An approximation is first obtained, and then, a method of estimating a more accurate signal arrival angle by obtaining received signal strengths at narrow angle intervals with respect to the angle around the angle of arrival is presented. This technique is described in more detail with reference to FIGS. 7, 8, and 9 as follows.

FIG. 7 is a diagram illustrating a window for controlling a mainlobe of an antenna beam, and shows a window for controlling an antenna beam width when the number of radiating elements constituting the array antenna is eight. Window 1 is a window in which the second to seventh elements have Hamming window coefficients and all others have zeros. Windows 2 is 65% of the center of the Hamming window, and Windows 3 is a rectangular window.

FIG. 8 is a diagram illustrating the strength of the received signal for each angle according to the window shown in FIG. 7, wherein a vector obtained by multiplying the window drawn in FIG. 7 by a reception antenna response vector, v _i and angles within -60 to +60 degrees. It shows the received signal strength for each angle obtained by correlating the reception array antenna response matrix. In the figure, received signal strength 1, received signal strength 2 and received signal strength 3 correspond to window 1, window 2 and window 3, respectively. In the figure, you can see that the beamwidth changes with the window. Therefore, when estimating the angle of arrival, the received array antenna response vector, b _i is multiplied by a specific window to widen the beam width, and then obtains an approximation of the angle of arrival by obtaining a correlation value at wide angle intervals, and narrows the beam width to another window again. Computation at narrow angle intervals around the approximation can be used to reduce the amount of computation by estimating a more precise angle of arrival.

FIG. 9 is a diagram illustrating a concept of improving an angle of arrival estimation according to an embodiment of the present invention. In this embodiment, the angle of arrival of an actual signal is set to 34 degrees.

First, the reception antenna response vector is multiplied by window 1, the correlation value is obtained at intervals of 20 degrees within -60 to 60 degrees, and the maximum value is found to obtain an angle of arrival of 40 degrees. Next, multiply the receive antenna response vector by window 2 to obtain a correlation value at intervals of 5 degrees within 25 to 55 degrees, and find the maximum value to obtain a more accurate angle of arrival, 32.5 degrees. Finally, the correlation value is obtained at intervals of 1.5 degrees within the range of 30 to 35 degrees, and the maximum value is found to obtain the final arrival angle, 34 degrees. This method reduces the amount of computation compared to the method of sequentially obtaining correlation values at 1.5 degree intervals within the full range, -60 to +60 degrees. Controlling the antenna beamwidth by multiplying the different windows in this way is that if the antenna beamwidth is too narrow when the windows are not multiplied, and the received signal strength is calculated at large angular intervals, the received signal strength cannot be calculated for the angle around the signal arrival angle. Is to prevent this from happening. Therefore, controlling the beam width by multiplying the window is particularly effective when the number of radiating elements of the array antenna is large, and it is not necessary to change the window when the number of array antenna radiating elements is not large.

While the invention has been described above based on the preferred embodiments thereof, these embodiments are intended to illustrate rather than limit the invention. It will be apparent to those skilled in the art that various changes, modifications, or adjustments to the above embodiments can be made without departing from the spirit of the invention. Therefore, the protection scope of the present invention will be limited only by the appended claims, and should be construed as including all such changes, modifications or adjustments.

As described above, the present invention provides a method for obtaining a forward beamforming weight value in a base station system using an array antenna, which maximizes a signal-to-noise ratio in a terminal by maximizing a transmission antenna gain in a direction in which a desired user is located and an antenna beam. Lowering the level of the side lobe of has the effect of reducing interference to other users.

Claims (14)

Adaptive array demodulation means for generating a reference signal required for receiving array response vector estimation; Cross-correlation means for estimating a reception array response vector by obtaining an average value over time of a signal generated by multiplying the reference signal generated by the adaptive array demodulation unit with the digital received data; Transmission beamforming weight calculation means for calculating a reception antenna response vector by removing a reverse channel component from the reception array response vector, and using the same to calculate a transmission beamforming weight; And And forward beamforming means for multiplying the beamforming weights by a user signal to perform a forward beamforming function. The method of claim 1, And the adaptive array demodulating means generates a reference signal using a symbol fed back from a pseudo noise code or a modem in a code division multiple access (CDMA) system. The method of claim 1, The beamforming weight calculation means, An antenna response conversion matrix storage device for storing a predetermined antenna response conversion matrix; A normalizer for obtaining a reception antenna response vector by normalizing a reception array response vector that is the cross-correlation means output to remove a reverse channel component; And And an antenna response converter, which receives an antenna response conversion matrix from the antenna response conversion matrix storage device, multiplies the received antenna response vector output from the normalizer, and converts the received antenna response vector into a transmission beamforming weight vector. Forward beamforming system using vectors. Generating a reference signal required for receiving array response vector estimation; A second step of estimating a reception arrangement response vector by obtaining an average value with respect to time of a signal generated by multiplying the reference signal generated in the first step by the digital reception data; A third step of obtaining a reception antenna response vector by removing a reverse channel component from the reception array response vector of the second step and using the same to calculate a transmission beamforming weight; And And a fourth step of multiplying the transmission beamforming weights converted in the third step by a user signal to form a transmission antenna beam. The method of claim 4, wherein The first step is a method for generating a forward beam using a reverse array response vector, characterized in that for generating a reference signal using a symbol fed back from a pseudo noise code or a modem in a code division multiple access (CDMA) system. The method of claim 4, wherein The third step, A first sub step of storing a preset antenna response conversion matrix; A second substep of removing a backward fading component from the reception array response vector obtained in the second step to obtain a reception antenna response vector; And And a third substep of receiving the antenna response conversion matrix stored in the first substep and multiplying the received antenna response vector output in the second substep to obtain a transmission beamforming weight vector. Forward beamforming method using array response vector. The method of claim 6, The first sub-step, After generating the transmission beamforming weighted target matrix by collecting all the transmission beamforming weighted target vectors by angle, selecting the matrix for converting the reception antenna response matrix into an approximation of the transmission beamforming weighted target matrix as the antenna response conversion matrix. A forward beamforming method using a reverse array response vector, characterized in that. The method of claim 7, wherein The first sub-step, And a transmission function formed by multiplying a transmission antenna response vector by a window function to reduce the side lobe level of the beam as the transmission beamforming weighted target vector. The method of claim 8, The first sub-step, And adjusting the window function to adjust the side lobe level and the main lobe width. The method of claim 7, wherein The first sub-step, In order to remove the gain change according to the direction angle of the array antenna, the transmission antenna response vector corresponding to each terminal incidence angle is multiplied by a single antenna gain compensation value and designated as the transmission beamforming weighted target matrix. Forward beamforming method using response vector. The method of claim 10, The first sub-step, And specifying the single antenna gain compensation value in inverse proportion to a single antenna gain for a particular angle. The method of claim 4, wherein The third step, A first sub step of calculating and storing a beamforming weight having excellent directivity for each angle in advance; A second substep of previously measuring and storing a reception antenna response vector for all angles in the sector; A third substep of removing a backward fading component from the reception array response vector obtained in the second step to obtain a reception antenna response vector; A correlation value between the reception antenna response vector obtained in the third sub-step and the reception antenna response vector stored in the second sub-step is first obtained at wide angle intervals to obtain an angle of arrival approximation, and then around the angle of arrival approximation. A fourth sub-step of estimating a signal arrival angle by obtaining narrow angle intervals in an angular range of; And a fifth substep of using, as the forward beamforming weight, the beamforming weight corresponding to the angle of arrival acquired in the fourth substep among the beamforming weights previously calculated in the first substep. Forward beamforming method using array response vector. The method of claim 4, wherein The third step, A first sub step of calculating and storing beamforming weights having excellent directivity for each angle in advance; A second substep of previously measuring and storing a reception antenna response vector for all angles in the sector; A third substep of removing a backward fading component from the reception array response vector obtained in the second step to obtain a reception antenna response vector; The reception antenna response vector obtained in the third sub-step is multiplied by a window function that first widens the beam width, and then a correlation value is obtained at a wide angle interval with the reception antenna response vector stored in the second sub-step to obtain an approximate arrival angle. And a fourth sub-step of multiplying a window for narrowing the beam width than the previous case and estimating a signal arrival angle by obtaining a correlation value at narrow angle intervals in the angular range around the angle of arrival approximation; And a fifth substep of using, as the forward beamforming weight, the beamforming weight corresponding to the angle of arrival acquired in the fourth substep among the beamforming weights previously calculated in the first substep. Forward beamforming method using array response vector. On your computer, Generating a reference signal required for receiving array response vector estimation; A second step of estimating a reception arrangement response vector by obtaining an average value with respect to time of a signal multiplied by the reference signal generated in the first step and the RF data; A third step of obtaining a reception antenna response vector by removing a reverse channel component from the reception array response vector of the second step and using the same to calculate a transmission beamforming weight; And And a fourth step of multiplying the transmission beamforming weights converted in the third step by a user signal to form a transmission antenna beam.