KR100311513B1 - Adaptive Beamforming Method - Google Patents

Abstract

In a direct spread CDMA system, a convergence speed is increased to adaptively form a beam of an array antenna by using information of a pilot channel for analyzing a channel state and a traffic channel information for transmitting actual information. An adaptive beamforming method, comprising: determining respective weights for beamforming from a pilot signal and a traffic signal respectively inputted through a pilot channel and a traffic channel, and using each coefficient value of a filter obtained by using the determined weights By updating the determined weight according to the recognized state of each channel, the present invention relates to an adaptive beamforming method capable of not only fast beamforming but also beamforming even when one of the pilot channel and the traffic channel is disabled. .

Description

Adaptive Beamforming Method

TECHNICAL FIELD The present invention relates to a direct spread CDMA system, and in particular, faster convergence in adaptively forming a beam of an array antenna by using information of a pilot channel for analyzing channel conditions and traffic channel information for transmitting actual information. The present invention relates to a method for forming an adaptive beam to have a speed.

In general, in the conventional adaptive beamforming method of the array antenna, a method using only the pilot channel information or a blind method using only the traffic channel information without the pilot channel information is typically used.

As a representative method using only the pilot channel information, there is a Least Mean Square (LMS) method, and a typical blind method is a constant modulus algorithm (hereinafter, referred to as CMA) method. There is this.

Here, referring to the principle of forming the beam of the array antenna according to the LMS method and the CMA method is as follows.

1 is a block diagram illustrating a general configuration for forming an adaptive beam of an array antenna. 1 applies to both pilot and traffic channels.

Referring to FIG. 1, first, signals received through each array antenna are input to the filters 10, 11, and 12 with carriers removed. The filters 10, 11, and 12 used here are (PSF: Phase Sharping Filters).

The output of each filter is subjected to demodulation and down sampling and despread. The illustrated configuration has a structure in which the weight of each beam forming unit 30 is updated every moment after despreading.

Here, when there are K transmission signals and L array antennas as shown in FIG. 1, the traffic channel reception signal and the pilot channel reception signal may be defined as follows.

----------------------------------------- Equation (1)

----------------------------------------- Equation (2)

Here, r_k` (t) is a traffic transmission signal, p_k` (t) is a pilot transmission signal, and w (t) is background noise.

The k th traffic transmission signal, r_k` (t), is defined as follows.

------------------------------- Equation (3)

----------------------------------- Equation (4)

Where A_k is the traffic transmission signal size, s_k is the traffic symbol, and T is the symbol interval.

d_k` (t; n) is the waveform of the n th spreading code of the k th traffic transmission signal, c_k is the l th spreading code for spreading the n th traffic symbol, N is the number of spreading codes for spreading one symbol, phi (t) is the chip waveform. T_c is the chip spacing of the spreading code.

A k-th pilot transmission signal, p_k` (t), is defined as follows.

Equation (5)

Equation (6)

From here, Is the pilot transmission signal size, Is a pilot symbol, Is the waveform of the nth spreading code of the kth pilot transmission signal, Is the l-th spreading code for spreading the n-th pilot symbol.

The traffic transmission signal r_k` (t) and the pilot transmission signal p_k` (t) pass through the filter, sample, and despread, respectively, as shown in FIG. The despread signal accumulates during the symbol period and averages. The code used for despreading is the c_k code on the traffic channel, and on the pilot channel. Is the code.

First traffic reception signal r_1` (n) among K despread signals received and pilot reception signal If is defined as:

-------------- Formula (7)

Equation (8)

Here, i_r, 1` [n] and i_p, 1` [n] are the sum of interference and noise in the traffic channel and the pilot channel, respectively.

2 is a block diagram illustrating a conventional LMS algorithm.

Referring to FIG. 2, the principle of the LMS algorithm in the structure of updating the weight of the beam forming unit 30 at every instant will be described with reference to FIG. 2 and equation (8).

The pilot symbol estimate is a weight vector of the beam forming unit 30 of FIG. And the despread pilot symbol signal, It can be defined as

------------------------------------------------expression( 9)

The Mean Squared Error (MSE) algorithm is used to derive the optimal weight solution in the pilot transport channel. This is possible because the pilot symbols are known.

The optimal solution is defined as follows using the MSE algorithm.

-------------------------------------------- Equation (10)

------------------------ Equation (11)

------------------- Equation (12)

----------------------------- Equation (13)

Here, C_P` (w_1`) is an evaluation function in the LMS algorithm, epsilon _LMS` [n] is an LMS error function, * is a conjugate complex number, and mu _p is an adaptive gain in the LMS algorithm. )to be.

The optimal coefficient can be obtained repeatedly by w_1` [n] defined in Equation (12).

According to the evaluation function C_P` (w_1`) in the LMS algorithm based on this principle, the weight vector is updated at every instant.

However, this method is because the signal strength of the pilot channel is smaller than that of the traffic channel ( Symbol estimation may be poor.

3 is a block diagram illustrating a CMA according to the prior art.

Referring to Fig. 3 and equation (7), the principle of the CMA algorithm will be described in a structure in which the weight of the beam forming unit 30 is updated every moment.

The traffic symbol estimate is a weight vector of the beam forming unit 30 of FIG. And the despread traffic symbol signal, It can be defined as

------------------------------------------------expression( 14)

Since there is no known reference signal at the receiver for the traffic channel signal, the final weight vector of the beamformer 30 is obtained by using the CMA algorithm.

------------------------ Equation (15)

------------------- Equation (16)

------------ Equation (17)

Here, C_r` (w_1`) is an evaluation function in the CMA algorithm, epsilon_CMA` [n] is a CMA error function, and mu_r is an adaptive gain in the CMA algorithm.

However, this method has the drawback that it does not guarantee convergence of the weighted final value. On the other hand, the LMS algorithm guarantees convergence of the final value.

The LMS method and the CMA method described above and the other recursive least square (RLS) method are methods of repeatedly searching for an optimal solution using an adaptive method, 2-D Rake Combiner) is a method of combining received signals by separating them into a time axis and a space axis.

These methods form a beam of the adaptive antenna using only one of the pilot channel information and the traffic channel information, and thus, there is a disadvantage in that the maximum available channel information is not used.

Therefore, when the use of the corresponding channel used for forming the beam becomes impossible, there is a problem in that the beam of the adaptive antenna cannot be formed at all.

SUMMARY OF THE INVENTION An object of the present invention has been devised in view of the above-mentioned problems of the prior art, and adaptively uses the maximum available channel information, that is, pilot channel information and traffic channel information, to adaptively adapt a beam of an adaptive antenna. It is to provide a beam forming method to be formed.

According to an aspect of the present invention for achieving the above object, the beamforming method of the adaptive antenna is to determine the respective weights for beamforming from the pilot signal and the traffic signal input through the pilot channel and the traffic channel, respectively; Then, the determined weight is updated according to the state of each channel recognized by each coefficient value of the filter obtained using the determined weight.

Preferably, when determining the weight, the weights for the two signals are determined differently according to the distortion degree of the pilot signal and the traffic signal, and the sum of the weights for the determined pilot signals and the weights for the traffic signals are determined. Is normalized to 1.

In addition, a least mean square method is used for beam formation using the pilot signal, and a constant modulus algorithm is used for beam formation using the traffic signal, and an average square using the pilot signal is used. The error by the method and the constant coefficient algorithm using the traffic signal are obtained, and the weight is updated according to the state of each channel recognized according to these errors.

1 is a block diagram showing a general configuration for beam forming of an array antenna;

2 is a block diagram illustrating a conventional LMS algorithm.

Figure 3 is a block diagram for explaining a CMA according to the prior art.

4 is a block diagram illustrating an adaptive beam forming method according to the present invention.

Explanation of symbols for main parts of the drawings

10,11,12: Phase Shaping Filter

20,21,22: Downsampled & Despreader

30: beam forming unit

Hereinafter, a configuration and an operation according to an exemplary embodiment of the present invention will be described with reference to the accompanying drawings.

The present invention is a technique for forming a beam of an adaptive antenna using multiple antennas in a direct spread CDMA system, which is ultimately used to increase the overall system capacity.

The beamforming method of the adaptive antenna is to operate before despreading or after despreading, so as to adaptively form a beam suitable for the spatial position of various mobile stations.

Here, whether to use the beam forming method before despreading or after despreading may be determined in consideration of the complexity and performance of the system hardware.

The beamforming method of the adaptive antenna according to the present invention is a method of repeatedly obtaining an optimal solution using a best decent method for every symbol or every code. This feature is easy to implement in real hardware and can immediately follow the characteristics of a channel that changes over time.

The method of forming the beam of the adaptive antenna by using the steepest decent method is classified into two methods according to whether the pilot channel information is used or not.

As mentioned above, there are a representative LMS method and an RLS method using a pilot channel, and there is a CMA method of forming an adaptive antenna beam using only traffic channel information without pilot channel information.

In the beamforming method of the adaptive antenna according to the present invention, it is desirable to increase the efficiency of the entire system by forming a beam faster in a CDMA system using both the pilot channel signal and the traffic channel signal used in these methods. It is the key.

Therefore, in the present invention, the beams are formed by the LMS method using the pilot channel signal, and the two methods are combined and used to form the beams according to the CMA method for the traffic channel.

4 is a block diagram illustrating a method for forming an adaptive beam according to the present invention.

Referring to FIG. 4, a pilot channel and a traffic channel are both present, and the LMS method and the CMA method described above are applied to each channel.

If both channel information is available, faster beams can be formed adaptively.

To do this, we use the new evaluation function C (w_1`).

C (w_1`) `=` lambda C_P` (w_1`) `` + '' (1`-` lambda) C_r` (w_1`) ----------------- ---------- Equation (18)

Equation (18) is an evaluation function defined in the present invention, which is obtained by adding weight λ to the evaluation functions of LMS and CMA. Here, λ determines which of the LMS and the CMA should be given more weight. When the distortion of the signal caused by the channel is less, the CMA may have more weight, and the sum of the two weights is normalized to 1. .

Accordingly, the equation for updating the coefficient combining LMS and CMA is as follows.

------------------------- Equation (19)

{W_1} `` [n] defined as described above may update the weight of the beam forming unit 30 filter at every instant.

First, the error of the LMS using the pilot signal is obtained from the result of multiplying the pilot signal and the traffic signal by the weight of the beam forming unit 30, and the error of the CMA using the traffic signal.

Thereafter, the weight λ is determined according to the state of the next channel, and a new coefficient value of the filter for the beam forming unit 30 is obtained from the determined weight.

Current CDMA systems obtain signal-to-interference (SNR) information for power control or handover. Since the approximate channel state between the terminal and the base station can be known from the signal-to-interference ratio information, this signal-to-interference ratio information can be used to determine the weight λ.

As described above, the present invention forms a beam of the array antenna faster by using both the information of the pilot channel and the information of the traffic channel, so that a call connection can be made in a short time, and also the pilot channel or traffic channel Even if any one of the use is impossible, there is an effect that the beam can be formed.

Claims (4)

The respective weights for beamforming are determined from the pilot signals and the traffic signals respectively inputted through the pilot channel and the traffic channel, and the weights of the respective channels recognized by the coefficient values of the filter obtained by using the determined weights. And updating the determined weight according to the method. The adaptive beamforming method of claim 1, wherein the weights of the two signals are determined differently according to the distortion degree of the pilot signal and the traffic signal when determining the weight. 3. The method of claim 2, wherein the sum of the weights for the determined pilot signals and the weights for the traffic signals is normalized to one. The method of claim 1, wherein a least mean square method is used for beamforming using the pilot signal, and a constant modulus algorithm is used for beamforming using the traffic signal. The error by the mean square method using the and the constant coefficient algorithm using the traffic signal is obtained, and the weight is updated according to the state of each channel recognized according to these errors. .