KR20060101648A - Channel equalizer and its method in digital broadcasting receiver - Google Patents

Abstract

The present invention relates to a time domain equalization apparatus and method for compensating for distortion due to a channel in a residual side band (VSB) digital broadcasting receiver. In particular, the present invention estimates a channel from a received signal, varies the number of taps of the time domain equalizer according to the estimated channel value, and then performs tap coefficient update by applying the LMS algorithm to the variable number of taps to determine the received signal in the time domain. Compensate for the distortion. Accordingly, since the number of filter taps varies according to channel conditions, the present invention has an effect of reducing the amount of calculation for coefficient updating compared to a time domain equalizer having a fixed number of taps regardless of channel conditions.

Time domain equalizer, variable number of taps, channel estimation

Description

Channel equalizer and its method in digital broadcasting receiver

1 is a block diagram showing an embodiment of a channel equalization apparatus of a digital broadcast receiver according to the present invention;

2 is a block diagram showing another embodiment of a channel equalizer of a digital broadcast receiver according to the present invention;

Explanation of symbols for main parts of the drawings

110: shear filter 120: noise removing unit

121: noise predictor 122: subtractor

130: reference value generator 131: determination unit

140: subtractor 210: channel estimator

220: weight control unit

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital broadcasting receiver of a residual side band (VSB) method, and more particularly, to a time domain equalizer and a method thereof for compensating for distortion caused by a channel.

In a transmission / reception system, a symbol transmitted through a bandwidth-limited multipath channel causes interference between symbols, which distorts the transmitted signal and causes bit errors in the receiver. In addition, such intersymbol interference has been recognized as a major obstacle in transmitting high-speed data in a multipath channel.

In order to minimize the intersymbol interference, a signal processing process called channel equalization is performed at the receiver.

In general, since communication channels are not fixed but time-varying, the equalizer needs to be able to track the time-varying characteristics of these channels. Such equalizers are called adaptive equalizers.

In general, the most popular adaptive channel equalizer is a decision feedback equalizer (DFE) using a Least Mean Square (LMS) algorithm. The DFE regards the most energy-intensive path as the main path when the received signal enters through a multipath channel, and considers all remaining paths as neighboring signal interference (ISI) or ghost signals coming through the reflection path. After the phase and magnitude correction are extracted only for the signal coming through the main path, the signals coming through the remaining path are removed.

The adaptive channel equalizer has the advantage that the structure is simple in that it does not require the autocorrelation value of the correct input signal or the cross correlation value between the input signal and the desired output. It also does not require complex operations such as matrix inverse operations.

However, when the channel distortion is severe, a judgment error frequently occurs in the judgment value entering the feedback filter. In addition, an error propagation may occur in which the equalizer performance deteriorates as the infinite loop rotates in the feedback filter due to an incorrect determination value. In this case, the decision feedback equalizer does not equalize the received signal properly.

Therefore, an equalizer capable of preventing the occurrence of error progression, which is a problem that the decision feedback equalizer has, and producing a consistent equalization performance, is shown in FIG. 1 and has been filed by the present applicant (Application No. Filing date 2003.10.20).

Referring to FIG. 1, a front end filter 110, a noise removing unit 120, an error signal generating unit 130, and a weight control unit 150 may be included.

The front end filter 110 is a time-domain linear adaptive filter composed of L fixed taps, and compensates for the distortion of the received signal transitioned to the baseband through the weight controller 150.

However, the above-described patent has a fixed number of taps, so that the larger the number of taps, the more the calculated value must be calculated each time the equalizer coefficient is updated.

The present invention is a complement to the above-described patent, and an object of the present invention is to provide a channel equalization apparatus of a digital broadcast receiver to improve the equalization performance by varying the number of time-domain equalizer taps according to the estimated channel condition through a channel estimator. To provide a way.

In order to achieve the above object, the channel equalizer of the digital broadcast receiver according to the present invention estimates a channel from a received signal and changes the number of taps of the time domain equalizer according to the estimated channel value, It is characterized in that the distortion of the received signal is compensated by performing tap coefficient update by applying the LMS algorithm.

The channel equalizer of the digital broadcast receiver, which has a hardware structure, includes: a channel estimator configured to estimate and output an impulse response of a channel from a received signal passing through a transmission channel; After setting the effective channel tap area according to the estimated channel characteristics by the channel estimator, the set effective channel tap area is determined, the number of equalizer taps is determined, and tap coefficient update is performed using the received signal and the equalized signal. A weight control unit to perform; And a time domain equalizer for varying the number of taps under the control of the weight controller, and compensating for the distortion of the received signal in the time domain according to the changed tap coefficient.

The apparatus may further include a noise canceller configured to predict a colored noise amplified at the time of equalization from the output of the time-domain equalizer, and then subtract from the equalized signal to whiten the noise.

A determination unit which is connected to an output terminal of the noise removing unit and outputs a determination value closest to a signal in which the noise is whitened among a plurality of preset determination values, a preset training sequence in a training interval, and A selector which selects a determination value output from the determination unit and outputs a reference signal, and a subtractor outputting the difference between the reference signal output through the selector and the equalized signal as an error signal to the weight controller; It is characterized in that the configuration.

The weight controller increases the number of taps of the time domain equalizer if it is determined that the length of the channel estimated by the channel estimator is long, and reduces the number of taps of the time domain equalizer if it is determined that the channel length is short.

The channel equalization method of the digital broadcast receiver according to the present invention,

(a) estimating the impulse response of the channel from the received signal passing through the transmission channel;

(b) setting a valid channel tap area according to the channel characteristics estimated in step (a) and performing tap coefficient update using a signal equalized with a received signal in the set valid equalizer tap area; And

(c) compensating for the distortion of the received signal in the time domain according to the tap coefficient of the effective channel tap region set in step (b).

Other objects, features and advantages of the present invention will become apparent from the following detailed description of embodiments taken in conjunction with the accompanying drawings.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention that can specifically realize the above object will be described. At this time, the configuration and operation of the present invention shown in the drawings and described by it will be described as at least one embodiment, by which the technical spirit of the present invention and its core configuration and operation is not limited.

The same components as in the related art are denoted by the same names and the same reference numerals for convenience of description, and detailed description thereof will be omitted.

FIG. 2 is a block diagram showing an embodiment of an adaptive channel equalizer according to the present invention. The channel estimator 210 estimates a path (impulse response) of a transmission channel from a received signal and outputs it to the weight controller 220. ) Is further provided.

2 includes a front end filter 110, a noise removing unit 120, a reference value generating unit 130, a subtractor 140, a channel estimating unit 210, and a weight control unit 220.

The noise remover 120 extracts only colored noise from the output of the front end filter 110 to predict the amplified noise at the time of equalization, and the noise predictor 121 from the output of the front end filter 110. The subtracter 122 subtracts the predicted noise and whitens the noise.

The reference value generator 130 is connected to the output terminal of the noise removing unit 120, the determination unit 131 for outputting a determination value closest to the signal whitening the noise of a plurality of preset determination values, In the training section, a preset training string is selected, and in the data section, the selector 132 selects and outputs a determination value output from the determination unit 131 as a reference signal.

In the present invention configured as described above, the received signals y (n), ..., y (n-L + 1), which are transmitted to the baseband through the demodulation process, are transmitted to the feedforward filter 110 and the channel estimator 210. Is entered.

In this case, when the transmission signal is called x (n), the impulse response of the discrete equivalent channel is called h (n), and the white noise is called w (n), the received signal y (n) input to the receiver is represented by the following equation. It can be expressed as 1.

를 가중치 제어부(220)로 출력한다. Therefore, the channel estimator 210 receives the received signal y (n) as shown in Equation 1 and estimates the impulse response h (n) of the discrete equivalent channel that the original signal x (n) has passed. Finite impulse response estimate

Is output to the weight controller 220.

In this case, the channel estimator 210 may use a simple correlation method (SCM) or a least square method (LSM) estimation method to estimate an impulse response of a transmission channel. It is the choice of the system designer to decide whether to use SCM or LSM in the present invention.

의 행렬 연산을 하여 전송 채널의 임펄스 응답을 추정하는 방법이다.That is, the SCM method assumes that the training signal periodically added to the transmission signal is a white signal, and obtains a cross correlation value between the training signal passed through the channel and the training signal already known at the receiver. It is a method of estimating an impulse response. In contrast, the Least Square Method (LSM) estimation method obtains a cross correlation value p between a training signal passed through a channel during a training time and a training signal already known at a receiver, and calculates an autocorrelation matrix of the training signal. Find R Then remove the autocorrelation part in p, which is the cross-correlation value between the received signal and the original training signal.

It is a method of estimating the impulse response of a transmission channel by performing a matrix operation of.

If the channel estimator 210 estimates N as the number of channel estimating taps that are passed through the channel estimating process, is it an environment in which there are many near ghosts or long ghosts through channel estimation? Can figure out.

That is, the weight controller 220 determines, for example, whether the environment has many ghosts or the environment of distant ghosts according to the characteristics of the channel estimated by the channel estimator 210 to be used for the front end filter 110. The number of taps to be used is varied.

For example, if the length of the channel is long, the number of taps to update the filter coefficient is increased accordingly. If the length of the channel is short, the number of taps is reduced while the channel is updated.

, n=0,1,...,N-1라고 할 때, 채널의 센터값을 max{

}, n=0,1,...,N-1이라고 하자. 여기서 유효한 채널 탭 영역을 다음의 수학식 2와 같이 설정한다.The impulse response value of the channel estimated by the channel estimator 210 is calculated.

, n = 0,1, ..., N-1, the center value of the channel is max {

}, let n = 0,1, ..., N-1. Herein, a valid channel tap area is set as in Equation 2 below.

In Equation 2, α is a threshold of 1 or less used to identify the channel region.

When the tap area of the valid channel is determined through n obtained in Equation 2, the weight controller 220 calculates the tap coefficient of the front end filter 110 by the LMS algorithm through Equation 3 below in the effective channel tap area. Update

In other words, the product of the input y (k) and the error e (k) of the filter tap is multiplied by the step-size u again and is updated to the next filter coefficient w (k + 1) by adding the current filter coefficient w (k).

In this case, the weight controller 220 increases the number of taps of the front end filter 110 to be updated accordingly when the length of the channel estimated by the channel estimator 210 is long, and when the length of the channel is short, the front end filter The coefficient update is performed while reducing the number of taps at 110.

Therefore, when the number of taps of the front end filter 110 determined by the weight control unit 210 is L, the total sum of the values multiplied by the L arbitrary filter tap values appears as the output of the front end filter 110.

The noise predictor 121 of the noise remover 120 predicts the colored noise amplified at the time of equalization from the output signal of the front end filter 110 and outputs it to the subtractor 122. The subtractor 122 removes the predicted noise of the noise predictor 121 from the output signal of the front end filter 110 to remove colored noise that has been amplified during equalization.

The final output from which the amplification noise has been removed by the noise removing unit 120 is input to the determination unit 131 of the host value generator 130 to generate an error signal for updating a new equalization coefficient.

The determiner 131 outputs, to the selector 132, a decision value closest to the signal whitened by the noise removing unit 120 among a plurality of preset decision values.

The selector 132 is a kind of mux, and selects a training string which is already known in the training signal section, and selects a determination value of the determination unit 132 in the data section and outputs the reference value to the subtractor 140.

The subtractor 140 obtains a difference between the output signal z (n) of the front end filter 110 and the output signal d (n) of the selector 132 and outputs the difference to the weight controller 220 as an error signal e (n).

The weight controller 220 multiplies the error signal by the step size for updating the tap coefficient, and adds it to the existing tap coefficient value to regenerate the tap coefficient of the front end filter 110 for the next time domain equalization.

This time domain equalization produces the output of the equalizer for every input signal (or symbol) and calculates the coefficients to be used for equalization of the next symbol.

Therefore, the error progression occurs in the feedback portion due to the severely distorted channel in the conventional DFE structure, but the present invention does not occur such an error progression by taking the FIR structure.

The time domain LMS channel equalizer having a variable equalizer tap number through the channel estimator proposed in the present invention can be applied to an ATSC digital TV transmission / reception system, a digital data broadcasting transmission / reception system, and the like.

On the other hand, the terms used in the present invention (terminology) are terms defined in consideration of the functions in the present invention may vary according to the intention or practice of those skilled in the art, the definitions are the overall contents of the present invention It should be based on.

The present invention is not limited to the above-described embodiments, and as can be seen in the appended claims, modifications can be made by those skilled in the art to which the invention pertains, and such modifications are within the scope of the present invention.

According to the channel equalizer and method of the digital broadcast receiver according to the present invention described above, the effective length of the channel is determined by estimating the impulse response of the channel from the received signal, and the number of taps of the time domain equalizer required for equalization is varied. Let's do it. Therefore, the amount of calculation for coefficient update is reduced compared to a time domain equalizer with a fixed number of taps regardless of channel conditions.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the present invention.

Therefore, the technical scope of the present invention should not be limited to the contents described in the embodiments, but should be defined by the claims.

Claims (9)

A channel estimator for estimating and outputting an impulse response of the channel from the received signal passing through the transmission channel; A weight controller configured to set an effective channel tap area according to the estimated channel characteristics by the channel estimator, determine the number of equalizer taps in the set effective channel tap area, and perform tap coefficient update using the received signal and the equalized signal ; And The number of taps is controlled by the control of the weight controller, and the channel equalization unit of the digital broadcast receiver comprises a time domain equalizer for compensating for distortion of the received signal in the time domain according to the variable tap coefficient. Device. The method of claim 1, And a noise canceling unit for predicting the amplified colored noise during equalization from the output of the time domain equalizer and whitening the noise by subtracting from the equalized signal. The method of claim 2, A determination unit connected to an output terminal of the noise removing unit and outputting a determination value closest to a signal in which the noise is whitened among a plurality of preset determination values; A selector configured to select a predetermined training string in a training section and to select a decision value output from the determining section as a reference signal; And a subtractor outputting the difference between the reference signal output through the selector and the equalized signal as an error signal to the weight controller. The method of claim 1, wherein the weight control unit And an effective channel tap area is set by applying the following equation.

Where

Is the impulse response of the estimated channel, max {

} Is the center value of the channel, n = 0,1, ..., N-1, and α are the thresholds used to identify the channel region. The method of claim 1, wherein the weight control unit And if the length of the channel estimated by the channel estimator is long, increases the number of taps of the time domain equalizer, and if it is short, reduces the number of taps of the time domain equalizer. (a) estimating the impulse response of the channel from the received signal passing through the transmission channel; (b) setting a valid channel tap area according to the channel characteristics estimated in step (a) and performing tap coefficient update using a signal equalized with a received signal in the set valid equalizer tap area; And and (c) compensating for the distortion of the received signal in the time domain according to the tap coefficient of the effective channel tap region set in step (b). The method of claim 6, And predicting the colored noise amplified in the equalization process of step (c) and subtracting the equalized signal to whiten the noise. The method of claim 6, wherein step (b) A channel equalization method of a digital broadcast receiver, characterized by setting an effective channel tap area by applying the following equation.

Where

Is the impulse response of the estimated channel, max {

} Is the center value of the channel, n = 0,1, ..., N-1, and α are the thresholds used to identify the channel region. The method of claim 6, wherein step (b) And if it is determined that the estimated channel length is long, increase the number of valid equalizer taps, and if it is short, reduce the number of valid equalizer taps.

Images