KR100463379B1 - Method for adaptive beamforming system of transmission and receiver smart antenna - Google Patents

Abstract

The present invention relates to a method for adaptive beamforming of a transmit / receive smart antenna, comprising: a transmission weight vector unit for receiving an original signal to receive and a signal source for multiplying the original signal to be received by the transmission weight vector; And an antenna, a reception smart antenna for receiving a signal received from the transmitting smart antenna, and a reception weight vector unit for receiving a signal from the smart antenna and multiplying a reception weight vector. The improved channel capacity can be obtained.

Description

Adaptive beamforming method of transmitting and receiving smart antenna {METHOD FOR ADAPTIVE BEAMFORMING SYSTEM OF TRANSMISSION AND RECEIVER SMART ANTENNA}

The present invention relates to a method of using a smart antenna, and more particularly, to an adaptive beamforming method of a smart antenna for transmission and reception.

An antenna that improves wireless communication efficiency by forming an optimal beam pattern by adjusting a beam pattern by multiplying a received signal by a weight using an array antenna is called a smart antenna. In addition, the signal processing method that controls the beam pattern, which is a key part of the smart antenna, is called an adaptive beamforming method. The array antenna referred to herein refers to a smart antenna.

Meanwhile, the beamforming methods of the conventional smart antennas include the following.

First, there is a method of repeatedly searching for an eigenvector corresponding to the maximum eigenvalue of the autocorrelation matrix of the received signal in order to obtain a weight that maximizes the SNR / SIR of the received signal. A method for obtaining a weight that maximizes the SNR / SIR of the received signal as described above is described in the paper "Ayman F. Naguib," Adaptive Antennas for CDMA Wireless Networks "[Ph. D. Dissertation, Dept of Electrical Engineering, Stanford University, Aug. 1996, for example.

An implementation of this is the MCGM (LCGM) method. Such a method related to MCGM is disclosed in Korean Application No. 1998-0057416 (name of the invention: a signal processing method and apparatus for optimal weight vector calculation of an adaptive array antenna system based on a conjugate gradient method, and Applicant: Seungwon Choi). For example.

Secondly, there is SGM, which is a method of analyzing a received signal using the position of the array of receive antennas. The method of analyzing the received signal using the position of the array of the receiving antenna is disclosed in Korean Application No .: 2001-0026629 (name of the invention: weight vector detection method of the smart antenna, Applicant: Jae-don Park, Ki-wan Yoon, Je-woo Kim) Can be mentioned.

On the other hand, there is also an ILSP method that obtains the least-squares (Least Square) based on the Mexican Likelihood (Maximum Likelihood) method. The ILSP method for obtaining Least Squares based on the Maxim Likelihood method is described in the article "Blind Separation of Synchronous Co-Channel Digital Signals Using an Antenna Array-Part 1: Algorithms". Examples are disclosed in T. Shilpa, V. Mats, and P. Arogyaswami, IEEE Trans.Signal Processing, 1996, vol. 44, no. 5, pp. 1184-1197.

Conventional smart antenna systems were only used on the receiving side (base station). In other words, it is considered impossible to use an array antenna due to its small size. However, with the development of communication technology, the necessity of using an array antenna for a mobile phone as well as a base station is raised. In a system such as a wireless LAN and Bluetooth, there is no inconvenience in using an array antenna as both a base station and a base station for communication. Therefore, there is now a need for a multi-user detection adaptive beamforming algorithm that can rapidly increase channel capacity by using array antennas on both the transmitting and receiving sides.

Meanwhile, a conventional technique compared with the OFDM beamforming method is in the following paper. Paper "Adaptive beamforming algorithm for OFDM systems with antenna arrays" [Chan Kyu Kim; Kwangchun Lee; Yong Soo Cho, Consumer Electronics, IEEE Transactions, Volume: 46 Issue: 4, Nov. 2000, Page (s): 1052 1058.], for example.

Briefly looking into the paper, the received signal matrix V (n) received by the S smart antenna is expressed by the following equation (1) and (2).

Where M is the number of users and N is the number of sub-carriers. Is the user orientation matrix for the user and B (n) is AWGN. And Denotes a user signal vector. here Denotes a complex relation transpose matrix. W (n) is a weight vector, and the adaptive beamforming signal vector R (n) passing through the adaptive beam former is calculated as shown in Equation 3 below.

here Denotes the complex relational transpose matrix of W (n). R (n) and W (n) are configured as in Equation 4 below.

Finally, the received signal extracted after the FFT (Fast Fourier Transform: FFT) processing is obtained as shown in Equation 5 below.

Here, F (n) means a Fourier transform matrix. Is expressed as in Equation 6 below.

And Is It means a reception signal of a subcarrier. Here, the k th subcarrier.

The adaptive beamforming method of the received signal in the adaptive beamformer is determined in such a way as to minimize the square solution of the pilot signal and the received signal. The method as described above is calculated as shown in Equation 7 below.

here Denotes the square of the error signal. so Is the average value. Is the frequency interval between pilot signals, q is the subcarrier position of the first pilot signal. so Is the magnitude of the error signal. Is the number of pilot signals included in an orthogonal frequency division multiplexing (OFDM) block. Also, Wow Is It means a pilot signal and a reception signal of a subcarrier. Equation as described above is expressed in a vector form as shown in Equation 8.

Here, each element of Equation 8 is calculated as in Equation 9 below.

here Is the error signal vector, and the Means a pilot signal vector, Denotes a predicted received signal.

The MSE (Mean Square Error: hereinafter referred to as MSE) in the time domain is calculated by Equation 10 below.

here, Is an Inverse Fast Fourier Transform (IFFT) transformed signal. Time domain vector And frequency domain vector Is obtained by the FFT matrix of Equation 11 as follows.

Is Means the complex relational transpose matrix.

Where the Fourier transform matrix Is obtained as in Equation 12 below.

The weight vector update equation of the paper is based on the Least Mean Square (LMS) of Equation 13.

here, of The gradient for is calculated as in Equation 14 below.

As a result, the weight vector update equation of the OFDM system is calculated as in Equation 15 below.

The OFDM beamforming method has a problem in that the weight vector is processed in units of blocks by a pilot signal.

Accordingly, an object of the present invention is to provide a multi-user adaptive beamforming method using a smart antenna on both the transmitting and receiving sides. It is another object of the present invention to provide an OFDM adaptive beamforming method of a semi-blind MSE-based algorithm based on MSE.

According to an aspect of the present invention to achieve the above object, in the adaptive beamforming method of the transmission and reception smart antenna, the process of initializing the weight vector, calculating the signal source with the initialized weight vector, and calculating Projecting a reference signal by projecting the signal source to a predetermined signal position; calculating an error signal using the calculated reference signal; calculating a gradient using the calculated error signal; And updating the weight vector of the transmit / receive smart antenna with the calculated gradient.

According to another aspect of the present invention, in the adaptive beamforming method of the transmission and reception smart antenna, the step of extracting the original signal from the received signal is to update the weight vector by block unit by a pilot signal, and the update weight vector Training by the pilot signal, and applying the blind vector to the weight vector by training the update weight vector by the pilot signal.

1 is a schematic overall configuration diagram of an adaptive beamforming system of a transmit / receive smart antenna according to an embodiment of the present invention;

2 is an operation flowchart of obtaining a weight vector in the adaptive beamforming method of a receiving smart antenna according to an embodiment of the present invention;

3 is an operation flowchart of obtaining a weight vector in the adaptive beamforming method of a transmitting smart antenna according to an embodiment of the present invention;

4 is an operation flowchart of updating a transmission weight vector of a transmitting smart antenna according to an embodiment of the present invention;

5 is a flowchart of a transmission / reception weight vector update operation of a transmission / reception smart antenna according to an embodiment of the present invention;

6 is a schematic overall configuration diagram of an orthogonal frequency division multiplexing (OFDM) beamforming system according to an embodiment of the present invention;

7 is an operation flowchart of an OFDM beamforming method of an adaptive beamforming method of a transmission / reception smart antenna according to an embodiment of the present invention;

8 is a graph showing the performance of an adaptive beamforming method of a transmission / reception smart antenna as the number of transmitting antennas is changed in an AWGN channel according to an embodiment of the present invention.

9 is a graph showing the performance of the adaptive beamforming method of the transmission and reception smart antenna as the number of transmit antennas in the Rayleigh fading channel according to an embodiment of the present invention,

10 is a graph showing the performance of the adaptive beamforming method of the transmission and reception smart antenna as the number of receiving antennas in the AWGN channel according to an embodiment of the present invention,

11 is a graph showing the performance of the adaptive beamforming method of the transmission and reception smart antenna as the number of receiving antennas in the Rayleigh fading channel according to an embodiment of the present invention,

12 is a graph showing the performance of the OFDM beamforming method in the adaptive beamforming method of the transmission and reception antenna in the AWGN channel according to an embodiment of the present invention.

1 is a schematic overall configuration diagram of a multi-user adaptive beamforming system of a transmit / receive smart antenna according to an embodiment of the present invention. Referring to FIG. 1, the number of antennas of the transmit / receive smart antenna is M and N, respectively. On the transmitting side, the signal source z (t) is transmitted by the transmission weight vector unit 110. After multiplying by the transmit smart antenna 120 to transmit. The transmitted signal is transmitted to the receiving smart antenna 130 through a channel modeled by the matrix H. The incident signal of the receiving smart antenna 130 is obtained as in Equation 16 below.

z (t) is the signal source to be transmitted, Means noise component. I is the number of interfering signals, Is the channel matrix of the i-th interference signal. And Is the transmit weight vector of the transmit smart antenna 120 of the i-th interference signal. Is AWGN (Additive White Gaussian Noise).

Received signal received by the receiving smart antenna 130 , Channel matrix H, transmit weight vector Is obtained as in Equation 17 below.

here Is the channel response between the m-th transmit smart antenna 130 and the n-th receive smart antenna 120. And Denotes a conjugate, a transpose, and a complex transpose.

Eventually, the signal incident on the receiving smart antenna 130 The received weight vector in the received weight vector unit 140 Complex relationship transpose matrix Multiply by to extract the original signal. The extracted original signal y (t) is calculated as in Equation 18 below.

2 is a flowchart illustrating a process of obtaining a reception weight vector of a reception smart antenna according to an embodiment of the present invention. Based on a conventional Least Mean Square (LMS) method according to an aspect of the present invention. Referring to FIG. 2, first, a signal is received by a receiving smart antenna in step 210. The received received signal x (t) is calculated as in Equation 19 below.

In Equation 19, H is a channel, Is a transmission weight vector, z (t) is a signal source, and n (t) is AWGN.

After step 210, the signal received in step 220 is multiplied by the weight vector of the receiving smart antenna to extract the signal. The extracted original signal y (t) is calculated as in Equation 20 below.

Is the complex relational transpose of the received weight vector, H is the channel, Is a transmission weight vector, and z (t) means a signal source.

In order to simplify the description of the weight vector of the reception smart antenna in the updating process, the noise component is described. And the error signal of the k-th sample Is obtained as in Equation 21 below. Here k denotes the k th of the arrival sequence of the received signal.

Denotes a reference signal, and y (k) denotes a signal predicted by multiplying a signal received by the array antenna by a weight vector.

In step 220 after step 220, the reference signal is projected to a preset signal left to replace the reference signal with the predicted signal according to an aspect of the present invention. The preset signal locus means the nearest signal locus of the predicted signal. Reference signal Is expressed as in Equation 22 below.

In step 230 and step 240, the gradient value is calculated by differentiating the square of the error signal by the weight vector of the receiving smart antenna. The equation of the gradient numerical calculation is obtained as shown in Equation 23 below.

In Equation 23 Means the gradient value.

Update the weight vector of the receiving smart antenna in step 250 after step 240 Is calculated using Equation 24 below.

here Is the received weight vector before the update, Is the gradient value at the receiving end, Is a step size constant. So the updated received weight vector Is the step size constant in the gradient value of the receiving side in the receiving weight vector before updating. Minus the product of.

3 is a flowchart illustrating a process of obtaining a transmission weight vector of a transmitting smart antenna according to an embodiment of the present invention. Referring to FIG. 3, in operation 310, gradient numerical calculation is performed by differentiating a square of an error signal by a weight vector of a transmitting smart antenna. Gradient Is obtained as in Equation 25 below.

Is the transmission weight vector before the update, Denotes an error signal.

Where error signal Is calculated as in Equation 26 below.

Is the reference signal, Denotes the extracted original signal before the update.

In step 320, a differential operation is performed using the transmission smart antenna weight vector in step 320. The equation after performing the derivative operation is obtained as in Equation 27 below.

here Denotes a complex relation, a transpose matrix, and a complex relation transpose matrix, respectively.

In operation 320, the operation of substituting Equation 27 into Equation 25 is performed. The substituted equation is calculated as in Equation 28 below.

The weight vector is updated after step 330 in step 340. Update weight vector as above Is obtained as in Equation 29 below.

here, Is the transmission weight vector before the update, Is the step size at the gradient It means multiplied by. Therefore, the updated transmission weight vector is the step size in the gradient value from the transmission weight vector value before the update. Minus the product of

A weight vector derivation process in step 340 of FIG. 3 will be described in more detail with reference to FIG. 4. 4 is a flowchart illustrating an update process of a weight vector equation in a process of updating a weight vector of a transmitting smart antenna according to an embodiment of the present invention. Referring to FIG. 4, the weight vector update equation can be simply changed through a series of processes as follows.

First, in operation 410, an operation of multiplying a transmission weight vector by a received signal is performed. In step 410, the received signal is calculated as shown in Equation 30 below.

In operation 410, the operation of multiplying the transmission weight vector by Equation 30 is performed. The result is calculated as in Equation 31 below.

After operation 410, in operation 420, a simple vector inverse transform operation is performed in Equation 31. The result is calculated as in Equation 32 below.

After the step 420, the operation of the transpose matrix is performed on Equation 32, which is the result of the step 430. The result of step 430 is obtained as in Equation 33 below.

After step 430, the operation of multiplying the received weight vector by Equation 33, which is the result of step 430, is performed. The result of step 440 is calculated as in Equation 34 below.

As a result of step 440, the update equation of the transmission weight vector is obtained as shown in Equation 35 below.

5 is a flowchart illustrating a transmission / reception weight vector update process among adaptive beamforming systems of a transmission / reception smart antenna according to an embodiment of the present invention. Referring to FIG. 5, first, a process of initializing a weight vector is performed in step 510. The weight vector initialization in step 510 refers to the value of the transmit / receive weight vector when k = 0. Initialization is expressed as in Equation 36 below.

After step 510, the signal source is calculated in step 520. Signal source calculation is calculated as in Equation 37 below.

After operation 520, the signal source calculated in operation 530 is projected to a preset signal position to calculate a reference signal. Here, the preset signal locus means the signal locus closest to the calculated signal source. Performing the operation in step 530 is expressed as shown in Equation 38 below.

In operation 540, an operation of obtaining an error signal using Equation 38 is performed. The error signal is calculated as shown in Equation 39 below.

After step 540, in step 550, the square of the error signal calculated using Equation 39 is differentiated into a transmission / reception weight vector to calculate a transmission / reception gradient. In operation 550, the calculation is performed as in Equation 40 below.

After operation 550, in operation 560, an operation of updating the transmission / reception weight vector is performed using Equation 40. The transmission / reception weight vector update is obtained as in Equation 41 below.

The updated transmission / reception weight vector is the transmission / reception weight vector before update. Send and receive gradients each Mu is the value of minus multiplied by a constant that determines the step size.

In step 560, it is determined whether the transmit / receive weight vector converges. If it is determined that the convergence is not converged, the operation is repeated from step 520.

6 is a schematic overall configuration diagram of an OFDM adaptive beamforming system of a receiving smart antenna according to an embodiment of the present invention.

Before briefly describing FIG. 6, OFDM is one of a multicarrier modulation schemes, and shows excellent performance in a multi-path and mobile reception environment. The frequency utilization efficiency is improved by using a plurality of carriers having mutual orthogonality, and is a method suitable for high-speed data transmission using a multi-carrier in a wired or wireless channel. In the case of a high-speed data transmission with a short symbol period in a wireless communication channel having a multipath fading, when the single carrier method is used, the inter-symbol interference becomes more severe, whereas the complexity of the receiver is greatly increased. Since the symbol period in each subcarrier can be extended by the number of subcarriers while maintaining the data transmission rate, a simple equalizer with one tap can cope with severe frequency selective fading channels by multipath. In the OFDM scheme, since a plurality of carriers having mutual orthogonality are used, frequency utilization efficiency is increased, and a process of modulating and demodulating the plurality of carriers may be implemented by FFT / IFFT.

In addition, FFT / IFFT converts FT (Fourier Transform) or IFFT (Inverse Fourier Transform (IFT) at high speed. FT is applied to aperiodic function. All signals are integer multiples of reference signal. Start with the notion that the sum of the sin and cos functions can be expressed. FT / IFT is expressed as in Equation 42 below.

Instead of converting Equation 42 for all frequency points, only the specific points are captured and only a part of the equations are called FFT / IFFT. Here, it is possible to select and convert points with features, and then compensate for the rest, because the characteristics are almost alive.

Referring to FIG. 6 with the contents of OFDM and FFT / IFFT, first, a reception-side OFDM adaptive beamforming system includes a plurality of receiving smart antennas 610 and a plurality of analog-to-digital converters (ADCs). 620, a plurality of reception weight vectors 630, a weight controller 640, an FFT unit 650, and a decoder unit 660.

The plurality of receiving smart antennas 610 that receive the signals transmitted from the plurality of transmitting smart antennas transmit the signals received by the transmitting smart antennas from the plurality of directions to the ADC 620. The ADC 620 converts the received signal into a digital signal. Thereafter, the converted signal is transmitted to the weight controller 640 and the reception weight vector unit 630. The weight controller 640 compares the received pilot signal as a reference signal to determine whether the received signal has a certain level of weight, and transmits the received signal back to the received weight vector unit 630. The received pilot signal Is obtained as in Equation 43 below.

The information on the weight of the received signal is transmitted to the reception weight vector unit 630 with reference to Equation 42. Thereafter, the plurality of received signals and the information of the plurality of reception weight vector units 630 are added together and transmitted to the FFT unit 650. The FFT unit 650 performs Fourier transform on the transmitted signal. Then, the decoder 660 receiving the converted signal extracts the original signal by decoding the signal.

Referring to FIG. 6 in more detail using the equation, assuming that the number of users incident on the signal received from the receiving smart antenna 610 is M and the number of array antennas is K, Is a matrix of direction vectors, and G (n) is AWGN noise. Reception signal vector Is expressed by Equation 44 and Equation 45 below.

By the way Multiuser Signal Vector, User's Signal Vector Is the multi-user weight vector of the weight vector unit 630. Is multiplied by That is, the received signal multiplied by the received weight vector User multi-user signal vector Is obtained as in Equation 46 below.

Is Is a complex relational transpose matrix.

Multiple user signal vector of Equation 46 And multiuser weight vector Looking at the component of the in more detail as shown in Equation 47.

Multiuser signal vector Converts to the frequency domain via the FFT unit 650. Signal converted to frequency domain Is calculated as in Equation 48 below.

That is, the signal converted into the frequency domain Is the multiuser signal vector Calculate with the product of and F (n).

Thereafter, the decoder 660 recovers the signal source. Recovered signal source Express as

7 is a flowchart illustrating a weight vector update process in an OFDM beamforming system according to an embodiment of the present invention. 7 is a flowchart illustrating a semi-blind method based on Mean Square Error (MSE) according to an aspect of the present invention. Referring to FIG. 7, the weight vector is updated in units of blocks by the pilot signal before the transmission signal is sent in step 710. In step 720, the weight vector is trained by the pilot signal. In operation 730, the weight vector is updated in a blind manner. In the above process, the reference signal is a recovered signal source. Instead of

Training in accordance with the pilot signal in step 720 will be described in detail through the equation. Multiuser Pilot Signal And multi-user pilot signals received Is obtained as in Equation 49 below.

here Denotes a Fourier transform vector.

When performing step 720, Equation 50 is first defined.

A pilot signal received according to the definition of Equation 50 Is the multi-user weight vector And the product of the received signal U_pm (n) defined in Equation 50 above. That is, the received pilot signal Is expressed by the following equation (51).

The adaptive beamforming method as shown in FIG. 7 is performed based on MMSE (Minimun Mean Square Error: MMSE). The cost function J (n) of the OFDM beamforming system based on MMSE is calculated as in Equation 52 below.

Also, the value of the gradient Is calculated as in Equation 53 below.

here, silver Block of Correlation matrix of the signal. silver Cross-Correlation vector of the pilot signal and the pilot signal of the receiver. Thus, weight vector update Is calculated by the following equation (54).

In other words, multi-user weight vector update Is the multi-user weight vector before the update. The gradient calculated by Equation 53 above Times the step size constant mu Minus the value of.

Here, the correlation matrix and the cross-correlation vector are calculated as in Equation 55 below.

Substituting Equation 55 into Equation 54 yields Equation 56 below.

here, Is It is an error signal between the pilot signal and the pilot signal of the receiver.

In step 730 after step 720, the weight vector is updated by applying a blind method. Now, after training by the pilot signal, the method converts to a blind method. From now on, pilot signals are no longer used. The weight vector update method has the following differences compared to the conventional method using a pilot signal.

First, receive signal Is Is replaced by However, the forming method is the same as Equation 49 in step 720.

Second, a reference signal of a time domain is obtained as shown in Equation 57 below.

Third, the received signal in the time domain before decoding is obtained as in Equation 58 or Equation 59 below.

or,

Weight vector in step 720 Is obtained as in Equation 60 below.

And error signal Is obtained as in Equation 61 below.

FIG. 8 is a graph showing the performance of an adaptive beamforming system of a transmit / receive smart antenna that appears as the number of transmit smart antennas is changed in an Additive White Gaussian Noise (AWGN) channel according to an embodiment of the present invention. By comparing the Bit Error Rate (BER) to the Signal-to-Noise Ratio (SNR) as the number of transmitting smart antennas is changed from 1 to 6, the number of transmitting smart antennas is almost the same when the number of transmitting smart antennas is 3 or 4. The same is true when the number of transmitting antennas is 5 or 6. It can be seen that the larger the number of transmitting smart antennas, the better the transmission rate of the transmitting smart antenna is in the AWGN channel. As a result, it means that it is strong in AWGN channels.

FIG. 9 is a graph showing the performance of an adaptive beamforming system of a transmit / receive smart antenna that appears as the number of transmit antennas is changed in a Rayleigh fading channel according to an embodiment of the present invention. As a result, as the number of transmitting smart antennas increases, the results of the BER according to the same SNR indicate that the more transmitting smart antennas have a higher transmission rate of the transmitting smart antennas in Rayleigh fading. As a result, it means that it is resistant to Rayleigh fading.

10 is a graph showing the performance of the adaptive beamforming system of the transmission and reception smart antenna appearing as the number of receiving antennas in the AWGN channel according to an embodiment of the present invention. As a result of BER according to the same SNR, it can be seen that the higher the number of receiving smart antennas, the better the reception rate of the receiving antenna is in AWGN. As a result, it can be seen that it is resistant to AWGN.

FIG. 11 is a graph showing the performance of an adaptive beamforming system of a transmit / receive smart antenna that appears as the number of receive antennas is changed in a Rayleigh fading channel according to an embodiment of the present invention. As a result of the BER according to the same SNR, it can be seen that the higher the number of receiving antennas, the better the reception rate of the receiving smart antenna in Rayleigh fading. As a result, it can be seen that it is resistant to Rayleigh fading.

12 is a graph showing the performance of the OFDM beamforming method in the adaptive beamforming system of the transmission and reception smart antenna in the AWGN channel according to an embodiment of the present invention. It is a diagram showing the performance of the BER with respect to the SNR by changing the number of elements (element). As the number of elements increases, the value of SNR decreases, and thus the value of BER decreases.

The present invention provides an adaptive beamforming method using a smart antenna for both transmission and reception. Both the transmitter and the receiver can use the smart antenna to achieve improved channel capacity. In addition, by providing an OFDM adaptive beamforming method in a transmission and reception smart antenna adaptive beamforming system, it is possible to increase the efficiency of the frequency for the channel.

Claims (9)

In the adaptive beamforming method of the transmission and reception smart antenna applied to both the base station and the mobile station, Initializing the weight vector; Calculating a signal source using the initialized weight vector; Calculating a reference signal by projecting the calculated signal source to a predetermined signal position; Calculating an error signal using the calculated reference signal; Calculating a gradient using the calculated error signal; And updating the weight vector of the transmit / receive smart antenna by using the calculated gradient. The method of claim 1, wherein the error signal is obtained using Equation 62 below. The adaptive beam type transmission and reception method of claim 1, wherein the reference signal is calculated by projecting the reference signal to a predetermined signal left as in Equation 63. The method of claim 1, wherein the transmission and reception gradient is calculated using Equation 62 as shown in Equation 64 below. The method of claim 1, wherein the transmit / receive weight vector is calculated using Equation 64 as shown in Equation 66 below. The orthogonal frequency division multiplexing beamforming method in the adaptive beamforming method of a transmit / receive smart antenna applied to both a base station and a mobile station Updating the weight vector on a block basis by a pilot signal; Training the update weight vector by the pilot signal; And applying the blind method to the weight vector by training the update weight vector by the pilot signal. The method of claim 6, wherein the pilot signal is initially calculated using a pilot signal, and the pilot signal is used until a blind method is applied. The adaptive beamforming method of claim 6, wherein the cost function is calculated using Equation 67. The method of claim 6, wherein the gradient calculation is calculated using Equation 68 below.